Comments on: Time to Talk about REBOL's Future
REBOL Technologies

Comments on: Time to Talk about REBOL's Future

Carl Sassenrath, CTO
REBOL Technologies
26-Aug-2012 3:33 GMT

Article #0510
Main page || Index || Prior Article [0509] || Next Article [0511] || 261 Comments || Send feedback

I've been quiet for a long while, and this blog is not easy for me to write.

I'm sitting here with a glass of 2010 Merlot from right out of the barrel... hoping to be inspired on how to write this. But it's difficult, and the 2010 and 2011 vintages aren't very inspiring either. They were troubled by cooler than usual summer temperatures. They lack body and character.

I find myself thinking back to when I began growing grapes and making wine, which seems like a long time ago, and it strikes me that I've been working on REBOL even longer. For more than a decade REBOL has been a labor of love. As a blend of theory, experimentation, and invention, the language embodies elegant and wonderful concepts and properties. It was and is the most productive language I've ever used. I hope your experience has been similar.

I still care a lot about REBOL and its future.

As you know, REBOL began as a commercial venture. But, times have changed, technology has changed, and economies have changed. The old model of how proprietary software products make money has been turned inside-out. And, it's even more difficult to make money in the computing language market. There are so many competitors, most of which are free, most of which are open, and many of which are quite well supported.

So, it's come time to consider the next stage in the future of the REBOL language. I'll tell you that I've not reached any final conclusions. I want to solicit your suggestions and ideas first.

I also need to tell you that I don't have much time to help out with whatever the REBOL future may be. I can help a little from time to time, but generally I'm very busy these days. Since the end of 2010 I've been working with an amazing company and a world class team building some really great products. I must admit that I'm quite addicted to it. I hope to say more in a future blog.

Anyway, I invite you to post your comments below, but here are some pointers to keep in mind:

  1. My time is very limited, so please make your postings relevant and on point. If your post gets to be long and drawn out, I'm probably not going to read beyond the first few lines.
  2. Keep in mind that REBOL is more than just a technology, it's also a community, and all of us have different interests or requirements. If you're a doctor using REBOL in the medical profession, your needs will be quite different from a university research scientist. That's fine. This is a brainstorming session.
  3. Let's keep this discussion within the REBOL community for now. I don't care much about what people who don't know REBOL say or want. They don't understand REBOL, and many never will. This is our thing, we should guide it where we want it to go.
  4. I cannot afford the time, energy, or heartbreak of looking backwards. Yes, REBOL has invented some very cool software technology. Much of it was years ahead of alternatives. But, what's done is done. That's the past; we cannot change it. I want to focus on the future.
  5. Please don't post negative or degrading comments. They will be removed.

If you know REBOLers who might be interested in this discussion, please let them know about this blog posting.

I look forward to hearing from you,

-Carl

261 Comments

Comments:

Graham
26-Aug-2012 22:48:57
Carl you must have me in mind. I'm a doctor and use Rebol for medical work. I need Rebol improved, and blocking bugs fixed. Operating systems and interfaces are changing but we aren't. If you can not work on these issues, please let someone else or the community who can.
-pekr-
26-Aug-2012 22:54:49
I don't know what kind of propositions you are expecting, Carl? You are clever guy on yourself, aren't you? :-) If you have not had enough of time to work on REBOL, and you know it will not change anytime soon, I would like to know - what is actually your proposition, to get REBOL into the future?

The situation with REBOL (in the competing area) is so bad and degraded, that I am not even sure, if open-sourcing everything would even help. But it would definitely help to those still caring, the community, and the users who's solutions depend upon the language.

We have heard some rumours about legal problems with eventually open-sourcing the REBOL. Whatever the truth is, sooner or later, Red and Topaz are going to replace R3. Although different products, those are at least alive.

So - why to waste so many years of so hard and precious work? The only solution I can see is fully open-sourcing R3. I can see no point in keeping it behind the doors, if there is noone who can work on it. If you really care about REBOL future, just answer to yourself, what's the point in keeping it from further progress?

tomc
26-Aug-2012 23:12:55
Hi Carl Glad to hear you are staying amused in a good way. I really don't know what you should do with your baby. I'm in the midst of changing what I do (chances are Solaris won't matter a bit in whatever my next job is.) But I do know people in academia have problems with languages they see as not open. (not that they could do something about it if it were open) On the other hand these open languages designed by committee or with a multitude of forks are horrendous.

Maybe the best thing if you can't put the necessary time into the care and feeding of Rebol is to turn it over to a group before the community fades away too. I know I feel pretty wispy.

Sabu Francis
26-Aug-2012 23:34:28
Hi Carl. I am from India. I've been using Rebol since a very long time. Though I am not a traditional programmer (I am an architect), I think it is people friendly and extremely efficient; even though people may say it is slow from the point of view of measurements at nano-seconds level but I don't care about that. My first love was Prolog. Then Rebol came along and I've stuck to both despite a lot of comments on both. I think there is a big niche that Rebol is fulfilling and there are a huge amount of us there you possibly don't know about: People from non-computer domains who want an intelligent programming language which is quick to code. I do hope you continue with Rebol into Rebol 3.0, remove the glitches here and there and release it quickly. I am not qualified to give my opinion whether it should be open-source or not, but my personal opinion is that making it open source could remove bugs fast. But since you have so obviously proved your talent as a language developer, I think the strategy of Rebol 2.0 of open sourcing the mezzanine functions (while leaving the core kernel proprietary and free) is also adequate for me ... just keep the kernel very short and sweet. BTW, I am possibly one of the oldest users of your SDK and I have developed a lot of programs using it for private use; and am seriously considering going commercial with some of those. If you want support from India, just let me know ... we are famous for our huge population :-) and I am very confident that India itself will be a huge market and I could possibly contribute to help you there. All the best
Rondon
26-Aug-2012 23:35:16
Hi Carl. Glad to hear you again. I'd like 3 things in order or priority: 1) Rebol for Mobile (Android at least and IOS); 2) VID, Graphical Interface such as html5/css, that is, don't worry, just render with WebKit/QT (join this library to Rebol Code), because we can use a lot of libraries (jquery,dojo,ExtJS) attached. 3) I would pay for Rebol for R3, but if you want the community to improve it, release it as open source, but with a license clause, if we make money, we donate to RT a percentage.

Glad you're back!!

Best Regards!

Rebolek
27-Aug-2012 0:10:30
Hi Carl,

I think you know what you have to do. Open source R3. Use GitHub or another system of your choice and let people improve the language. If you want to have some "official" version, pick someone as maintainer if you haven't time to do it yourself.

I really don't see any other way to keep REBOL alive.

Best regards,

Rebolek.

Henrik
27-Aug-2012 0:29:07
I entirely agree with Rebolek. There are at least 5 hardcore developers ready to jump on the source immediately to fix remaining bugs in R3. From our point of view, open sourcing it is a no-brainer. We've been waiting 18 months for any movement.

There are several alternatives: Topaz, Red and World Language that are directly a response to R2's and R3's current state and Red is starting to move quickly now, but is far from completed.

While their goals are slightly different than R3's the basic idea of the REBOL language remains the same.

We need a functioning REBOL or REBOL derivative in the future.

Brian Tiffin
27-Aug-2012 0:50:59
Carl, thanks for asking.

Find a lieutenant. Give our good BrianH or similarly invested individual with the demonstrated skills access to the build machines, if not all the sources. REBOL 2.7 is a sweet system as is, but it needs to keep up with builds.

Have good Carl, nothing but good.

Cyphre
27-Aug-2012 0:55:19
Hi Carl, Glad to hear you are doing well. I understand your hard times regarding R3 decission so here are my suggestions:

-Make R3 core lib fully open sourced so remaining bugs can be fixed, interfaces with Host-Kit part extended and R3 ported to other platforms.

-Don't worry about forks. Just always offer "official" version of REBOL sources on rebol.com site. Noone will complain the official version doesn't have all the "current" changes but people can rely on it as the trusted code base.

-alternatively, if you want to still make some money from the R3 code and don't want to open it fully, please offer at least some source code escrow licensing for companies/organizations that want to use R3 for their products.

-don't forget R3 is still the most advanced and complete "next-gen REBOL" product when comparing to all the REBOL alternative projects people are talking about so letting R3 die would be really waste of all your time and energy you put into it.

MaxV
27-Aug-2012 1:00:48
Hi Carl, I wrote rebol2.blogspot.com, people are excited of Rebol, we get over 11,000 hits in few months. Facebook group is about 250 people and growing... but people feel that Rebol has never be full supported:
  • lack of good documentation
  • lack of a public forum (AltMe is not the best place for newcomers)
  • lack of a guide, you disappear for months
So we lost a huge number of people (Facebook group had at least 1,000 members inscribed and unsubscribed). Everybody knows that Rebol is better than Java and you are a great genius, but it's not enough. Like a child, a software must be followed, corrected, improved.
I suggest you to maintain the proprietary of the official build, but leave the source to the community. You could act just like a chief, deciding what to is in and what is out in the canonical version. Community could start an official forum with threads and you could link to rebol.com. Rebolek is right, put Rebol source on a platform like github, sourceforge, etc; since you are busy, follow the development from far, just put a post once a year saying, for example, "Keep it simple!". Community doesn't expect from you anything, doesn't expect to solve bugs or add features, just to leave us to improve and continue to develop one of your great invention: Rebol.
Endo
27-Aug-2012 1:10:02
If you think that any of the gurus can lead the project and continue development of R3 (and/or bug fixing R2) then you should give sources and rights to them, if there is no license issues etc., or open source it all.

Otherwise, do you see any future for REBOL? If you do, how? Any plan?

We wait so long for any activity, but you don't (and won't) have time to work on it anymore.

"I still care a lot about REBOL and its future." - so you should have something more in your mind, please tell us. Best regards.

Jerry Tsai
27-Aug-2012 1:29:49
+1 to Cyphre's comment.
Maarten
27-Aug-2012 2:43:11
Carl, I've stood in awe of all of your products, so I am *very* curious to your next(current) product.

Having said that: with REBOL, go the McCarthy way. That is, write a paper about the denotational semantics etc. and open up a reference implementation. Then let history unravel itself.

Be a Michelangelo in stead of a Bill Gates.

GrassRoot
27-Aug-2012 3:05:03
I think you'll have trouble getting users to ever again trust Rebol as proprietary software. You killed that confidence with no information, development or support for many months.

If possible - open source Rebol - or the language will die!

Hail Red!

Robert
27-Aug-2012 3:11:40
Hi Carl, any idea was already discussed. So no need to go through it again. It only takes a decision and an execution and R3 will gain speed. And that is:

Open the source to everyone or at least a group without any complicated constraints in the next 4 weeks. If this is not possible tell us now.

For me this is the last chance for Rebol to revive. It's dead as it is.

This will be the last chance I give Rebol. I will wait about 4 - 8 weeks and than move on.

Ryan Cole
27-Aug-2012 3:12:58
  1. Congrats on your new job.
  2. Fantastic work on your last job.
  3. Open source it.
  4. Put it somewhere to catch exposure (sourceforge?).
  5. Tell us REBOL's birthday, so we can all pop open a bottle once a year and properly reflect on its greatness.
  6. Append onward: [] onward
Jerry Tsai
27-Aug-2012 4:38:01
Carl, we've been using REBOL for a long time. We all love REBOL, deeply. Please let us help to keep REBOL alive. You are holding the key.
WuJian
27-Aug-2012 5:11:21
Carl,you brought Rebol to the world. Now,Rebol is dying and will disappear in the history.

No, just Open Source it,give it a new life. Rebol can have a great future.Let's see it.

Everyone can benefit from the success of Rebol,not only the community, and , you will benefit the most. Let's see it.

Peter Wood
27-Aug-2012 6:24:08
Your work on REBOL is to valuable to be lost. If you still don't feel comfortable or aren't able to release it under an open source license, please consider a commercial arrangement to license the software to someone (or a group) you fully trust - Gabriele, Nenad, Robert, Ladislav, Cyphre, Gregg.
Bo
27-Aug-2012 7:23:04
You already know my opinion. Open source with a committee oversight.

The others are right. Let it live!

Brian Hawley
27-Aug-2012 7:51:06
+1 to Bo, Cyphre and the rest who think that it should be open sourced with a committee looking over it. That worked for the open source parts of R3, and it seems to be working for Red.

Forks are good, as Ruby and Python have demonstrated. The computing world is becoming increasingly diverse, with new application models in use. The only way to keep up with these changes is to let people experiment with the new platforms, the new models. We will converge on the best ideas, and the rest will need to be different. Write once run anywhere doesn't work anymore, or is too limiting to be useful on most platforms.

I will be glad to help as much as I can, though that won't be as much as before. At the moment none of the REBOL-alikes support the platform I work on professionally, so using any of these at work would require major changes. I would love to see a REBOL ecosystem that could be flexible and diverse enough for all of us to be a part of it.

James Nak
27-Aug-2012 7:59:06
Carl,

First, I imagine this was one of the hardest letters you've ever written so I am happy for you now that you've finally done it. Secondly, I feel as if I've seen this movie before with C=. All of us can have a different ending this time, right? Please Open source it and keep in touch once in a while.

Steven White
27-Aug-2012 8:04:25
First, do no harm. (Isn't that what the doctors say?)

If R3 is not coming, or is going to come very slowly, make sure R2 does not go away. I suggest this selfishly, because I am finally starting to use R2 enough that my understanding of it is expanding enough that I can start using it more. I am amazed to see how I can beat my head against some problem for hours, read the documentation and try to understand it, and then finally find that the solution to my problem was just a handful of lines of code. If version 2.7.8 is complete, please make sure it remains in existence. If there are known bugs or some critical areas of incompleteness, it would be nice if you could at least fix up those, or make available to some suitable volunteers the resources to keep R2 in existence.

I get the feeling that the documentation is not as complete as might be considered ideal. It would be nice if you could find someone to go over that, and make available the resources to do it. I myself actually like to write documentation, but I don't know how to help with REBOL because I would need documentation to understand it in the first place. VID seems to be particularly lacking. You need a personal clone. (Can you get your kid to work on it? Mine mows the lawn, yours could write documentation. That's why we have them.)

As for R3, I am not qualified at this time to have an opinion.

Steven White
27-Aug-2012 8:17:03
Regarding the open source concept, I don't know much about it, but it is my understanding that you don't just put the source code on a web site and let everyone have at it. These products have an official maintainer who knows the product and can coordinate everyone's efforts and submissions to keep the product intact. That means that you will have to perhaps remove yourself from the coding chores, but remain in the maintainer role, or pass along the maintainer role to a new and trusted maintainer, AND, spend a certain amount of time with the new maintainer to make sure he understands your vision and how you have executed that vision up to the current time.

So it looks like you are going to have to spend at least some time, no matter what method you choose for removing yourself from the REBOL life.

Nick
27-Aug-2012 8:40:35
Carl,

It's clear that there are people who not only care about REBOL, but rely on it. You initiated this reliance by creating REBOL, and if you can't support it, I hope that you will find a way to release the code with an open source license. That provides the possibility of continued use in the future, for existing users, and provides a hope for new life among a modern crowd. My gratitude for your creation is deep, so I have mixed thoughts about asking you greedily to give up control of your baby, but you will always be understood and appreciated as REBOL's creator. I think you can still maintain control of which releases are considered "official", and you will attract serious developers who will the continue your work. Your vision of what REBOL can do for people has been fully realized in my life and work, and I hope others will find the same use for it. For me, it's been a profoundly useful tool. There are other people who also believe in it and need it. If you'd like to see REBOL fulfill it's potential, and if you cannot support it, I hope you will find a way to open the source. All the work and brilliance that you put into it deserves a chance to be continued.

Michael Church
27-Aug-2012 8:40:38
Carl, you've done great work and I would love to see it continue. If your time is greatly limited at a minimum I would set up a group of people that can move it forward that would have access to the source. My preferred would be that the entire source would be opened so that people could take it and bring about the true potential, but have an official release (very similar to the Linux kernel) that goes through trusted hands before being put out there.

Once again thank you for all the great work!

Nick
27-Aug-2012 8:52:44
PS - I'm still using REBOL2, as it's the only complete product for my needs. I understand that most thoughts about pushing REBOL ahead are with R3, but I think that a lot of potential REBOL users are attracted to the completeness of REBOL2, so releasing that source is just as, if not more, important, in my opinion. Is that possible?
gg_tk
27-Aug-2012 9:13:19
Hi Carl !

What about the idea of dual-licensing ?

* A fully open-source, "community" version using a copyleft licence, like GPL 2/3. That is to say if someone forks REBOL, or integrates it into an other project, he has to distribute the software and its source code under the same licence / conditions.

* A "commercial" version, bought by companies. It's exactly the same software, but the company isn't required to give back its modifications : they can develop proprietary products.

It's what Trolltech/Nokia does with QT, for example (http://qt.nokia.com/products/licensing/).

More informations : http://en.wikipedia.org/wiki/Multi-licensing

Bye, and thanks for all the fish ! :)

sqlab
27-Aug-2012 9:36:41
I would like R2 as open source, as it has served me many times despite some minor flaws. But I imagine, that R3 can soon reach a mature state, if it goes that road, as there seem still enough hands available to push it further. Anyway thanks for R2.
Sergey Shishkin
27-Aug-2012 9:42:51
You must write a text (book) preceded by open source.
Brian Hawley
27-Aug-2012 9:46:44
gg_tk: Trolltech did that for Qt, and then Trolltech had to get bought out by Nokia. Nokia went LGPL, but still had to sell off Qt to another company. There's no indication that this model will work for a programming language without a huge installed base, and even then it can fail if the language has competing implementations with more permissive licenses, as REBOL does.

It's not just the REBOL-alikes that REBOL competes with, it's also other languages. (L)GPL languages haven't been doing as well as their non-copyleft counterparts, particularly languages that generate programs that might be constrained by the license of the language runtime. The only languages that can survive as copyleft are the ones that accumulated a huge installed base long before comparable permissive license competitors existed (pretty much just Java and GCC). Even GCC is losing ground to LLVM now.

REBOL doesn't just have to compete against Red and Topaz, it has to go against Ruby, Python, Javascript, Go, and dozens of new languages that run on JS, JVM, Dalvik, LLVM and CLR. Most of these new languages are free (as in money) and have permissive licenses.

Steven White
27-Aug-2012 10:14:34
It might be obvious, but keep in mind what people really are asking for in the clamor for open source. We want REBOL to live on and evolve. Currently, open source is the best (or best-known) way to make that happen. Another way would be for you to work on REBOL full-time forever and never retire, die, or just get tired of it. If you can come up with a third alternative, that might be good too.
Gregg Irwin
27-Aug-2012 10:47:46
Whew! You're alive. :) We've missed you terribly, but I'm thrilled to hear you're so excited about your current work. I think everyone in the community wants you to be happy, and to see REBOL succeed.

Many of us love REBOL deeply, and rely on it. We need it, for both reasons. We want to contribute, to see it live on, to see it honored for its innovations, and hold it up as an example for other languages to follow (as Red, Topaz, and World have).

If you trust the community to shepherd REBOL with care, I believe we will rise to the challenge. There is enough talent, and people who will provide resources, to make it happen technically. Organizationally, there are no guarantees, but if you hand-pick the bootstrap team and, critically, if you can occasionally make an appearance and provide guidance, there is hope.

None of us can know what REBOL means to you--how much of *you* is in it--but we do care about both. I have more than 10 years of my life in REBOL and, despite its flaws, there is still nothing that comes close that I would rather use. Thanks for that, whatever happens.

If you believe in the power of REBOL, unleash it.

Arnold
27-Aug-2012 11:11:05
Hello Carl, Please open source R2/R3 and 2012 will definitely be the best wine-year ever, promised! ;-)
YekSoon
27-Aug-2012 11:20:06
If the objective is to grow the community and its adoption, I think going the open source route will not hurt much. The community is not that big to begin with. So, we don't lose much here :(

If the objective is more commercial oriented, there will be other considerations.

What will be the key focus or objective here?

I don't think we can have it both ways.

Laurent
27-Aug-2012 11:36:51
Yes open sourcing Rebol will boost its adoption. Then you can make big money by being *the* Rebol expert.
Karol
27-Aug-2012 12:04:26
I suggested You to read all the comments and then disappear for two years to think about what to do next ... just kidding. I still use Rebol and I am worry about its future. My tips:

1. Choose maintainer

2. Open source it on github (don't care about forks, evolution will choose the best)

3. Create official groups, communication channel, wikis etc. using standard, "google scannable" ways like twitter, google groups ...post something from time to time to make it official

4. Put orginal/official version of rebol.exe at rebol.com with commits to main source tree accepted by you or your maitainer

5. Keep reboling

Maxim Olivier-Adlhoch
27-Aug-2012 12:08:04
Funny, I thought it was already all sorted out, behind the scenes.

If there are no legal reasons not to MIT license and fully disclose the source code, please do so. IMHO its the only way your decades of language insight will have lasting effects and be all worth it in the end.

There still ARE people whose livelihoods depend on REBOL to continue on.

Maxim Olivier-Adlhoch
27-Aug-2012 12:10:52
Allowing several others the chance to carry on with the torch, is just as noble as carrying it.

Don't be afraid of Forks, they are the natural selection process of organically evolving languages, just like human languages.

Graham
27-Aug-2012 12:47:54
Carl, there was always going to be a time when you needed to pass the baton on. Better to do it now when it still has some relevance, and can enrich and sustain the community rather than in some distant time in the future when no one will notice.
Gerard
27-Aug-2012 13:48:08
Hello Carl, (J'espčre que tout va pour le mieux pour toi)

I'm glad you found a new treasure to love while you work. My best wishes to you in the future and a lot of thanks for having recreated the SMALL IS BEAUTIFUL reality in the way you did with REBOL 2 and 3.

Hope both of your versions of REBOL will deserve their fullest potential in a near future, like yourself when you'll have organized their and your succession - that is

1) write your will . what you'd like REBOL to go from now on, . a clear way as to how it had been done up to now, . how the road could have been done if you had had the time to do it yourself

. Kind of book as suggested by Marteen may be welcome - ask for writers to help you.

2) Choose the person which is the mostly apt to supervise the execution of your written will

3) Give this person and the others who will help him to render concrete this will, the keys to permit this dream to become true (open source may be the solution you hoped for and is a necessity too if we want to be a bit serious)

4) If you can afford it - keep yourself as an angel to look over the shoulders for the members of this community - just to see how it is going well and growing from now on...

God bless you - your initiative never got equaled (in terms of space and simplicity) - but inspired many newcomers that will be ready soon. Help us to keep up the good work you already did.

Thanks and I'm now ready to get more news from your new work - as I'm now retired and have a lot of spare time !!!

Gregg Irwin
27-Aug-2012 14:25:51
Gerard makes an excellent point.

Carl, what is REBOL's perfect future in your mind?

Andreas
27-Aug-2012 14:26:57
Open source R3. And do it now.
Maarten
27-Aug-2012 15:01:53
Carl, I clearly remember the analogy of REBOL and the red or blue pill in the movie The Matrix.

In the end, the ideas behind REBOL are timeless. I wouldn't call it analogue to Maxwell's equations, but more like that than say, Python.

Either way, thanks. I still use it...

Jerry Tsai
27-Aug-2012 18:30:48
+1 to Laurent. Being "the expert" can actually make money. Take myself as an example. Big companies (MS, Sun, IBM ... ) invited me to their conferences, programmers bought my books, companies hired me to guide their software development, because they thought I was "the" Java Expert in Taiwan and China. Indeed, I was good at Java, but I don't think I was "the" expert then. The truth is, I was good at showing myself, and ,you know, every business is show business. ... So please Open-source REBOL, make it popular, and become *the* REBOL expert and start the show business (which is not so bad as it sounds). The difference between you and I as experts is ... you are the real expert, a guru, and I am just a fake one. :-) Best wishes to REBOL.
Erin Thomas
27-Aug-2012 18:52:22
Bo just pointed me to this blog. Yes, open source is the way to go. This allows both you and REBOL to move forward in the best possible ways. You've seen your technological child through conception, birth, childhood, and adolescence. Now it's all grown up and ready to go out into the world, and it will make you a proud, proud parent.
Tsun Kuo
27-Aug-2012 20:21:12
Though I am not a really hard-core expert on it, I really fascinated by the beauty, concise and easy spirit that it endowed which make programming of tedious works becomes a joy. Hope it will be keep alive forever by those enthusiasts and hobbyists.
Refaktor
28-Aug-2012 0:49:31
In short. You created a language that some of us think is (still) exceptional and would like to continue to use.

You don't "owe it to us" to support and develop it for your whole life.

The landscape is constantly changing and if you, as you said it yourself "don't have much time to help out with whatever the REBOL future may be". Please kill it or put it on GitHub. Let the bird fly. We promise to send you postcards. Maybe you'll like the places it goes...

IMHO the biggest problem we (rebol community) have right now is indefinite waiting. So whatever you do, do it very soon please!

Robert Paluch
28-Aug-2012 0:59:13
I totally agree with Cyphre :-) and I think it is your crossway in life: a)Open yourself,to believe to people and to release R3 e.g. like Open Source or b) Close yourself, kill R3, keeping philosphy "either R3 with me or nothing".

R2 helped me and my customers many times. It is the best language for small and powerfull form apps. Thank you Carl! R3 is for me hope that i can continue to develop small apps with love and not like "code-machine" :-)

Paullys
28-Aug-2012 2:04:19
No existing licence can achieve your goals, Carl. So and because you've got no much time, you have to invent a new licence: -to show the source code -but keep control of creation (comited process or what you want). - ...what you want To resume Think of that new licence and go to the best lawyer you know.
Paullys
28-Aug-2012 2:07:59
Whatever you decide, keep in mind that your decision is important TODAY. TOMOROW time would make the choice for you!
Luis.
28-Aug-2012 3:31:59
As others have mentioned, overseeing the new rise of Rebol would be the way to go. Look at Python and see if you can picture yourself as the Rebol BDFL.

The current derivatives show not only that Rebol merits continuation, it also shouts out the lengths some are willing to go to maintain that vision. Maybe handing over the Source to these developers as the initial maintainers, if they are willing, would be a good first step.

As for financial incentives for the developers: Maintaining a Rebol editor, a subscription service for VID updates (especially considering not only the desktop space but also the mobile platforms), etc.

Cheers,

Luis.

Luke
28-Aug-2012 5:11:48
Open source it, preferably BSD. The kudos for creating the world's most usable deep programming language will always be yours. Good luck with your new ventures!
Giuseppe Chillemi
28-Aug-2012 6:04:15
Hi Carl, you have developed REBOL with the idea of create a revolution in the programming world. Your ideas and concepts are really haead of time. However revolutions needs reasources, mostly money, and some luck to reach the mainstream. I suppose in the last two years you have thought a lot about the past and the future: how much money and time you spent on the project and your dreams about it. The silence you kept for long time meant for me "Carl needs to reconsider everything. He needs this time". Now you are asking for opinions. Here are mine:

1) You should believe in us ! We are able to keep the "make the revolution" part of your dreams. The community is mature enough to procede developing R3 under an open source licence.

2) You are our oracle. We need a committee to supervise the project. You should really be part of it. We don't need a single line of code anymore from you. We will code in place of you. We need only a guide and ideas.

3) Once open sourced we will try to get some funding to help developers work on the project.

4) Rebol Tech. could return someday to provide solutions based on the Rebol Language than producing the Rebol Language itself.

I don't know anything about you except you like wine and you are a great inspirator. I know about many people moving from "all white" to "all black" thoughts about something. It seems you loved REBOL and now you reject it. Please consider an option in the middle about your involvement. We need only 2 days of your time each month.

You are precious !

Regards, Giuseppe Chillemi

Giuseppe Chillemi
28-Aug-2012 8:00:05
Here is a simple ToDo List:

Goal:

1) Open Source REBOL3 2) Complete REBOL3 3) Make it multi Platform

Todo:

- Choose which public licence best suites REBOL3

- Create a Committee of Gurus involving Carl

- Invite Developers to give their opinions about the future

- Define the Developers involved into improvement and maintance.

- Choose a public repository for code

- Release all the code for REBOL2 and REBOL3

- Release all the documentation in editable format

- Choose a public Wiki for Documentation

- Resurrect the DIAGRAM Carl Created to show which will be the structure of REBOL3

- Make public on the the Media that now REBOL is open sourced

- Start Bug Fixing REBOL2 and REBOL3

- Create a development path

- Improve VID for R3

- Port all the available documentation to Wiki and improve it

- Add Features Planned

- Find Funds for the development of REBOL

- Move Rebol.Org Server

- Start Porting to other platforms

....

Jerry Tsai
28-Aug-2012 8:35:39
As you might know, I wrote a R3 book in Chinese, trying to attract more people to the REBOL camp. It sells very well so far. But I would like to attract more than Just Chinese. So I decide to translate it into English and make the English Edition free. It's a slide with 320 pages. This slide is colorful, beautiful, and easy to understand. I would like the community to help me correct my English before I put it to the public. I would also like this gesture to make Carl feel the support and love from the community.
Nick
28-Aug-2012 8:55:57
Jerry, I'll help with that as much as needed. You can reach me at reverse "gro t.o.d zratiug t.a kcin"
PatrickP61
28-Aug-2012 9:37:59
Carl, good to hear from you -- we have all sincerely missed you and your insightful contributions.

I also created a product long ago of my own, nowhere near as creative as REBOL, yet I used it for my own purposes. While it did what I wanted, and I did show it to others, it just never reached any level of critical mass. So it is a relic left on my hard drive. My contribution to this discussion is simply "Where would you like to see REBOL progress to, in the future regardless of the level of your involvement?"

I suppose no language is ever really finished as there are constant bug fixes that seem to drag on. I seem to recall the original idea of re-writing REBOL to R3 to incorporate uni-code as well as many other improvements. It does take an awful lot of commitment, dedication, and desire to bring a new language into the world -- even though it started as a simple idea for your own private use.

So maybe the question to answer is how and what kind of structure would you like to see REBOL supported by? Then you can decide how to transition to it.

Regardless of what you decide, Thank you for creating and sharing REBOL. It is a valuable product.

srinivas
28-Aug-2012 10:03:14
I have been visiting this blog as well as other Rebol related blogs because I was drawn by the simplicity and elegance. Initially my need was to generate documents in a portable way and makedoc had led me to visit Rebol related sites. I haven't used Rebol to do any development however I am interested in seeing this vision move forward.

As a bystander my comments are:

- define the vision clearly - PITS, PITL (REBOL3 and beyond)

- collaboratively develop a roadmap

- define governance structure - define how we things get decided. Carl's active voice is required in design decisions, at least initially.

- let others do the work to ensure the vision is unfolding

- collaboratively define a communication plan - regular communication builds confidence, trust and puts some pressure to demonstrate progress

- have/appoint a drum beater who will move things along. This will ensure that deliverables get out the door at regular intervals

- just open sourcing may not produce the desired results without rallying the community, rebuilding trust is an important facet

- organize support structure to ensure other aspects besides development get attention e.g. documentation, tutorials, showcase of real world usage. Right now it is very fragmented and since people don't have authorship rights on rebol.com, everyone is rolling out in a small way. The community needs a focal point.

- above all - communicate how you (Carl) will communicate and frequency of communication

Brian Hawley
28-Aug-2012 10:04
Carl, I know we can't follow the same path we did before if you can't be more actively involved. That central control that you have over REBOL now requires a lot of work to do, and when you can't the project suffers.

One of REBOL's main advantages is that we've had a chief designer, someone who can keep the language on course, someone who can say no to bloat and cruft and excessive complexity. This is why it has been able to stay so clean, small and efficient. We need to keep that advantage going forward.

I think that we should switch from a control model to an inspirational model. You have enough talented people who know the REBOL way, now these people can inspire others. The official repository can be the spiritual center of the community, and the branches and forks can be the spreading of the word, so to speak. This model has worked well for Ruby and Python, and they didn't have the advantage of being REBOL. Matz and Guido don't have to implement everything, others follow their lead.

Pick a license that we can build a community around and we'll do it. You don't have to do license tricks to maintain central control. If you pick a license that will allow code to be shared with Red, Topaz and such, practical and aesthetic considerations will be enough to keep the community together, and keep our expertise and vision in play. If the same people can work on all of these platforms, they will keep the comparable stuff similar just because they can.

Change the product to an ecosystem, or a movement.

Jordyn
28-Aug-2012 10:10:25
Just got a Raspberry Pi and I would love nothing more than to see that support REBOL. It just might get a lot more interest since the Raspberry Pi is just getting started and selling very well. I think a lot of developers would use REBOL if we could get an ARMv6 port available for the Raspberry Pi.
Nick
28-Aug-2012 11:08:37
Yes, mobile is very important for the future (Android)

... as is the *immediate* time frame. You're about to lose the involvement of Robert and his group. As far as I can tell, that would be a critical loss to the community. If you want the help of a talented group of serious developers, who are genuinely capable of adding to the code base, motivated to actually use REBOL for commercial projects, and willing/able to complete important parts of REBOL3, etc., that appears to be be a pressing matter. People like that will give the project renewed traction.

Luca Truffarelli
28-Aug-2012 11:35:24
Hi Carl,
I was very happy to read your post, it made me think that Rebol would have lived again, but then I realized that's not so simple.

I'll keep it short, people that met Rebol want it to be alive, and the only practical way seems to open source it.
But I think that Rebol is not just a programming language, it is an idea that was born to the real life.
So, to keep Rebol as it has been during this years we need its creator (who owns the idea) to evolve it, the community alone can't do it.

So, moves as you feel better to realize your idea, the community is there to help, and you, somehow, have to be there too.
Bo
28-Aug-2012 13:22:47
Regarding Luis's comment, Carl is the programming language Messiah. Even if he is never involved again (I find that hard to believe) it doesn't mean the faithful fall away from what he has taught us about the true programming language path.

On this note, if Carl wants to consider some new names for REBOL, perhaps the following may be considered:

  • Messiah
  • Divine (D-vine)
  • Enlighten (Enli10)
  • Sassenrath
  • Elevate (Elev8)

But I digress. Remember, in the programming language sphere:

Carl = Messiah REBOL Community = Faithful

If REBOL is released as open source, before making changes to the source, the community should ask themselves "WWCD?" (What Would Carl Do?) If you stray from the true path, Carl may throw a lightning bolt your direction. ;-)

Bo
28-Aug-2012 13:23:59
Oops. I mean "Regarding Luca's comment..."
luminarious
28-Aug-2012 14:12:28
If wishes were bushes, I'd be a bushman. I'm never going to be experienced enough to contribute to REBOL's development as a programmer. I don't have the know-how to decide wether R3 is the final step in the evolution of a language.

But I do have a huge interest in as a quick prototyping tool, both for games and general utilities, so I'm among those that would like the implementation of said language to continue to be functional on the common platforms.

Oh, and maybe a multimedia library that would enable playback of common sound, video and image formats. That would be great.

Gregg Irwin
28-Aug-2012 15:37:46
Brian Hawley +1
Nick
28-Aug-2012 18:54:13
Carl, I'm not sure if you're quite aware of how much REBOL has made a difference in many real lives. Merchants' Village, Pittston, PA (http://merchantsvillage.com) is running entirely on software written in REBOL. *Thousands* of people shop there every week for food and essentials. Hundreds of people make all or part of their living. All point of sale, inventory management, reporting, scheduling, web site management - everything - happens in REBOL. Millions of bar codes have been rendered, and hundreds of thousands of sales processed, entirely by REBOL code. It's a real place, full of life - a huge destination that involves and improves the lives of so many people. It would NOT have been possible without REBOL. Writing the code that runs that business, entire in REBOL, was that place's reason for being. Also, the Rockfactory in Newtown, PA. We were on the front page of the New York Times earlier this year, and we affect the lives of hundreds of students, and provide incomes for many instructors, all organized and made possible because of software written entirely in REBOL. Also P & R Discount Foods in Eynon, PA. They provide low price food for *thousands of people every week. Every single piece of inventory in that place is priced and all inventory is managed with a custom application that was written from the ground up in REBOL. Again, dozens of livelihoods are enabled there, and the lives of real people improved because of that system. There are dozens of other projects that I've been involved in that would not have seen the light of day, if not for REBOL. Did you know that a system that schedules commercial programming for FOX has been written entirely in REBOL (and is used in daily production)? That software was conceived and largely written by a person who could not have otherwise, without the learning curve made possible by REBOL. Please consider the value of what you've created, what you've made possible, and what you may potentially make possible in the future, in the lives of many real people, if you allow people to continue to use REBOL, particularly by opening the source. It really will make a difference. It's not just a computing experiment, or a piece of amazing creative technical engineering. It has a valuable functional purpose in real life. That is what you hoped to create in the end, isn't it? REBOL has a potential place in the lives of so many people, who otherwise could NOT become programmers and who would NOT be able to use the power of computers to build software for use in their very real lives, to accomplish useful/functional/productive, life-improving goals. I've always thought that was the most obvious value of REBOL. It's accessible by normal people, like no other tool. You said in your post that the economics of technology have changed, but the fundamental problems able to be solved with REBOL have not. You can help people in a way that so few others ever get a chance to. You can enable people in a way that is truly profound. You already have - and I bet you've done it for a lot more people, and in bigger ways than you probably know. If you allow REBOL to be continued by others, you are giving a gift that no other person alive can. I'm truly grateful for what you've enabled in my life, and the lives of people around me, and I hope that you will find a way to keep the project going. I'm willing to help in any way I can.
Scot Sutherland
28-Aug-2012 19:41:54
I have an honest question for you all. If REBOL were to be open sourced(OC). How do we propose to protect REBOL from Complexity Pollution (CP)? So far OC = CP.

If that question can't be answered acceptably well. I would rather work toward the purchase a license for the closed source. REBOL's greatest attribute is its immunity to complexity.

Kaj
28-Aug-2012 19:59:02
If that's true, than that trend has stopped. Red is fully open source under a permissive licence and has much lower complexity than other products with similar functionality.
Henrik
28-Aug-2012 20:26:38
Scot,

This has in fact already been attempted for the past 18 months to no avail.

Open source complexity:

There has to be a small body of overseers that are qualified to make decisions on what goes into the canonical repository. You can't avoid a big, dirty cloud of forks and variations, so there has to be one canonical source, to which only the best people have access. Fortunately, open source tools like github allow us to choose exactly who gets to do what in such a place.

The body of overseers is probably not going to be a single person, although I'd prefer this, and I would of course prefer it to be Carl, who would outline the design in broad strokes and let the best skilled community members implement it.

But since Carl is excited about his new job, the design work should probably continue in a company, perhaps with legal bindings, so that only this entity can claim to be the "new official designer of REBOL". I'm perfectly content, if Carl continues to legally own the source and continues to run rebol.com, as long as competent people have access to fixing bugs in R3 and have access to changing the documentation, post release news, etc., if Carl wants the outward appearance of R3 as an RT product to remain the same.

The design of R3 is not done, for examples, media types are not yet implemented and Carl has so far not revealed how he would do that.

However there are a big number of bugs that could readily be fixed by the community, for which the intended design is already set and would likely not make R3 any more complex.

Had Carl remained and worked at his highest activity level back in 2007-2010, I would prefer to keep going with a closed product, but this is no longer an option, especially as some people have already started to build commercial products with R3 and fixing these bugs is critical.

Gregg Irwin
28-Aug-2012 20:41:13
Scot, there are no guarantees. But consider the community, and that the vast, or at least vocal, majority lament complexity in other languages and laud it in REBOL. We have a great framework to go from, including examples of how the base kernel and other elements can be broken out (as in the SDK) for encap purposes. I don't think your average REBOLer wants to make REBOL more like C in order to increase adoption, or add dependencies that would make it grow to the likes of Python or Ruby.

Yes, some of us want a full module system for PITL, and certainly other additions will be asked for, but we may also see versions where the core shrinks and more of REBOL is written in REBOL.

Personally, I love being able to encap an app and have no external dependencies. Love, love, love it. There are also things I want to see added, as will a lot of other people. How to decide what gets added is hard, and won't always please everyone, to be sure.

I want to use REBOL, for a long time to come, but time marches on. OSs change, and if REBOL can't keep up to run on them, I can't justify using it for commercial work. A closed license model will, I fear, alienate most of the people who would contribute to REBOL, who already love it, and will keep it alive. In order for a closed license to work, someone will have to pay others to keep it going. And that will still self-exclude many open source advocates.

Gregg Irwin
28-Aug-2012 20:43:38
Sorry, should be clear that what we laud is REBOL's elegance and *lack* of complexity.
Scot Sutherland
28-Aug-2012 21:13:55
Kaj,

I have watched the development of Red. It is young. Just the few divergent paths mentioned by Greg lead me to believe that complexity will grow, perhaps more slowly but inevitably, precisely because there is not a single decision maker to settle divergent aims.

If there were any mature counter-examples to the OS = CP equation, I would be happy to consider it, but there are none, I've looked for a very long time. So I have to conclude that just as gravity is a law OS = CP is also a law.

Greg,

I have no arguments about the commercial license model for a language. If people can do what they want for free, there is no value in the market, even if the product is superior. Money is made by holding the consumer and the product manufacturer hostage. Complexity pollution solidifies the position of the power brokers.

We need a new and different model and I am interested in exploring ideas. One path would be to center REBOL development in an academic setting the way Net Logo did (now OC and already fragmenting). Net Logo modeled complex emergent systems and became important the fields of complexity science and ecosystems.

REBOL has a much greater potential to impact a larger range of research and development, including commercial enterprises as long as it remains immune to CP.

Scot Sutherland
28-Aug-2012 21:28:55
Henrik,

Your response resembles closely the path taken with Net Logo (Northwestern University). Uri Wilensky developed Net Logo for modeling biological systems about 15 years ago. He oversaw development with a group of handpicked developers. Net Logo benefitted from several rounds of research grant funding.

Now Uri has moved on to other projects and Net Logo as been open sourced. Already it has become part of the markup mess with a push to port the entire thing to javascript.

What's required is something like you describe that has a commercial outlet. R3 comes very close to this model technologically but there needs to be an accompanying funding/commercial piece.

The lesson learned from Uri's departure for me is the need to keep Carl in the game even if it is only as a high level decision maker. Finally, I think we need to learn as much from Carl about the way he writes code and designs systems as we can. I find myself redesigning a lot, from scratch, something that is really only possible with REBOL these days. The code stays small and elegant.

Luis.
29-Aug-2012 0:57:50
Nick +1

This reminds me, in a sense, of the Speccy and the C64: Little boxes that anyone could do anything with at the outset as the language was built in. It's this ease of 'getting it done' in Rebol that harks back to those days when technology was an enabler.

Cheers,

Luis.

Philippe
29-Aug-2012 3:07:46
Hi Carl, glad to read a new post after so many time. What is better than a Master ? Two Masters...or more. - Open-source Rebol with a license or some management rules to control the growth of complexity and keep the code elegant, powerful & simple. - With our Dockimbel & others masters, join the qualities/features of R3 and Red (or Rails) to produce the next generation of langage for all platforms (mobile ...). Let us dream again. Thanks to be back. Cheers
Refaktor
29-Aug-2012 3:33:10
Just saw the news that android will power NASA mini satelites (google it). That is the power of Open, to go beyond where initial authors can go. Yes iOS will be more polished, but android will change the world while closed iOS will stay the apple gadgets OS. Android is also already ported to Raspberry Pi, which somebody mentioned.

Scot Sutherland: perhaps moving something like netlogo (where easy access is key) to javascript is what is required to give it distribution it deserves. Reminds me of processing.org (processing.js) which has part of same audience/goal. Whole 3d engines are being compiled to JS (like BananaBread via Emscripten)

Carl Read
29-Aug-2012 4:56:50
Carl, I'm sure you knew open-sourcing REBOL would be what most would want. And I'm sure you know all the reasons for and against it. But for the record, I'm for OSing it too.

However, REBOL's a messaging language, right? And it's important messages mean the same thing to the sender as the receiver. Which, in REBOL's case, means the datatypes have to mean the same to both sender and receiver.

For this reason, REBOL needs a locked-down specification for its core datatypes, as apposed to the language. And that specification should be how you control an open-source REBOL. ie. The datatype specification needs to be part of the license. Then any fork would need to comply with the datatypes specification to claim to be a REBOL-based language.

Obviously the number of core datatypes can and should grow, and some maybe extended. But hopefully, if well thought out, they should never need to be broken.

Luke
29-Aug-2012 5:07:22
+1 for Carl Read's suggestion about specifying core data types and structuring devices (perhaps block, map, context...).
Nick
29-Aug-2012 6:06:38
There may be lots of additional mileage in the idea of putting various specifications and technical requirements directly in the license. I wonder if that's the silver bullet that Carl needs to ensure development stays on track. Does anyone know if that's been done in other license models?
tombon
29-Aug-2012 6:55:55
open source R3 now. if money ia a concern (don't think so but in case) initiate a crowdfunding to see what's possible and if success pay out mr. invisible. be the kernel-maintainer for commercial R3 and let the evolution do the rest before typing curly braces will be as easy as typing curly brackets.
Steven White
29-Aug-2012 7:43:46
I thought about this last night and an idea came to mind. It made a bit more sense last night when there were only 68 comments, but I'll mention it anyway.

One thing needed by those who use REBOL for production is a stable version they can count on. The current situation is not bad in theory. We have a stable version (R2), a new and improved one in the pipeline (R3), and various highly-compatible products that could be used in place of REBOL if the official REBOL were not quite right (RED, etc.). However, the problem is, as everyone knows, that it is not quite clear if R2 is going to keep up with things like OS changes (Windows 7 and 8), R3 is stalled, and Carl can't keep going forever even if he wanted to. So the following came to mind:

Carl puts his limited time into R2. He polishes it to his liking, fixes bugs, keeps up with OS changes, and maybe writes some documentation that only he can write, things about the REBOL design and inner workings. This keeps a stable trusted version in existence.

As for R3, Carl releases it as open source to a trusted group of maintainers. He remains involved as a guide and arbitrator, to make sure R3 doesn't get the complexity pollution and remains aligned with his vision.

As time passes, R3 will be "done" and Carl can let ease out of any work on R2. The now-static R2 will remain as a model of the great idea that it is. (There ought to be a software museum.)

I have no idea if any of this is even possible or reasonable or how much work it would take. The ability to offer ideas at will is an advantage of the uninformed.

Steven White
29-Aug-2012 8:25:25
Regarding Nick's comment above, I too am one of those who depends on REBOL to do things in my daily activities at work. I have been using REBOL only a little, but over a long time, so I looked to see just how much REBOL we have. A quick check showed over a hundred scripts that are/were in regular use.

Now many of these are trivial, but trivial by REBOL standards. If I had to do some of those things in COBOL (email is one concrete example I have found) it would take a page of COBOL to do what one line of REBOL can do. REBOL has saved me from a lot of work

As a related twist, when I retire I will leave behind those REBOL scripts. We don't do a lot of programming here anymore. We all are generic "analysts" who install and configure software we have purchased. A person who replaces me is not going to be a programmer. At some point he might have to work on one of the REBOL scripts. He won't know REBOL, BUT, he might not know any other programming language either. He will be faced with learning REBOL and modifying a short and heavily-commented REBOL script or learning another programming language and rewriting a REBOL script. Based on the little I know of other languages, the choice is clear.

And finally, maybe REBOL is sort of self-perpetuating. If enough REBOL gets out there in production, maybe it will make no sense to replace it because it would more work to do it in an alternate way. Maybe the rebolution will take not years, but generations.

Brian Hawley
29-Aug-2012 9:33:28
Steven White: If you've used R2 and R3, then you know that having Carl focus on R2 is doing him a disservice. R3 was made for many reasons, and having to work on R2 instead would be disheartening. Let him work on R3 and he'll be more likely to work on REBOL at all (no offence intended, Carl).

R2 won't be keeping up with OS changes; in particular, it won't be getting the Unicode strings that are needed to support versions of Windows from 2000 on, since the necessary changes would basically turn it into R3. However, we might be able to make R2 at least not crash on more recent OSes, and that would be nice. There's a lot of little fixes we could do, but we need to consider R2 to be a legacy platform for legacy OSes.

Steven White
29-Aug-2012 9:48:23
Sorry, I see now that he did say he didn't want to look back. I did not realize the full meaning of that. I have not used R3 at all. I guess that's what I meant by having the "advantage of the uninformed."
Gregg Irwin
29-Aug-2012 10:00:46
Carl Read +1

I thought about mentioning the data format idea, but my messages are already too long. If the the REBOL data exchange format is open, and there is a DLL that makes it easy to load and produce REBOL with other languages, there's a lot of leverage there.

Brian Hawley
29-Aug-2012 11:00:56
Gregg +1

The core datatypes don't need to be locked down for interoperability; all you need for interoperability is a common data interchange format, or more than one for different messaging circumstances. That's the advantage of messaging. Actual datatypes are in-memory, and only need to be metaphorically the same between different systems, having the corresponding semantics for the system's process model, rather than the same semantics across different process models.

Remember, load is just a function that takes bytes and returns internal data structures. You can easily have more than one such function for different streams of bytes, as long as they return the same data structures on the same system. Same with save in reverse.

Kaj
29-Aug-2012 13:19:04
Scot,

This is not the place to have an extensive discussion about Red, but you're wrong that it doesn't have a single decision maker. Nenad is its inventor, designer and chief implementer.

Of course its complexity will grow while it develops, but the point is how it compares to similar products. Currently, Red/System is a complete cross-platform compiler/linker toolchain for a low level language, and its complexity compares extremely favourably with similar products. There is no reason to assume this advantage will be lost during the implementation of the high level Red language. To the contrary, the whole point of the low level language is to reduce the complexity of the high level language as much as possible.

Even if they're in the minority, it's easy to give more counter-examples to your "law". Lua and Io are open source languages with favourable low complexity. Certainly Lua is very mature, with a long-standing ecosystem around it. Syllable is an open source, general purpose operating system with much lower complexity than other open source, general purpose operating systems, and certainly much lower complexity than commercial systems. Genode is a newer, open source operating system with a revolutionary design specifically aimed at reducing complexity.

Open source does not equal complexity, software equals complexity. The most complex software systems in existence are closed source. But also, there's a natural spread filling all niches from low complexity to high complexity. Where you are depends highly on one's judgement. REBOL has a great capacity to teach people what's complex and what is less complex. But they will never know if REBOL stays in its cave to die.

Scot Sutherland
29-Aug-2012 14:16:37
Kaj,

Thank you for your response. I have looked at all the systems you mention. None have a front to back complete solution. If an OS were made in REBOL nothing else would be required. Glad to hear Red has a single decision maker. Never allow that to change and Red has a chance.

Honest question. If there is Red, why do we even need REBOL?

Brian Hawley
29-Aug-2012 14:31:19
Scott, compiled and non-compiled languages have different semantics, different advantages. Red, once it exists, would be best for certain kinds of tasks, and REBOL would be better for other kinds of tasks. There is a lot of overlap of course, tasks for which either would be suitable. It's better to have both.
Kaj
29-Aug-2012 14:39:16
Lua, Io and Syllable are complete solutions. Academic differences can always be pointed out; they're not REBOL. But even REBOL couldn't give you a complete operating system: it would be part of the C code in which REBOL is implemented. Consequently, you would also still need a C compiler toolchain, and a traditional operating system to run it on. In Red, however, a complete operating system could be implemented, where the kernel would be written in Red/System, and the Red compiler would run on the system itself. (In operating systems, this is the important milestone of being self=hosted. Syllable has also achieved this, but with the traditional open source toolchain.)

About why we need REBOL: I wonder myself. In its current state, I'm sad to say it has confined me more than it has empowered me. This could have been very different if it had been more accessible, and this problem can still be solved.

Compared to competitors, REBOL is still more mature and more integrated. The high level Red language is not here yet, and it is not planned to get its own cross-platform GUI. This will make for better integration with host platforms, but it will be less portable. The one thing where REBOL is still unmatched is its integrated GUI. However, the R3 version is stalled, the R2 version is more and more in shambles, and if this doesn't change soon, this last attraction will also disappear.

-pekr-
29-Aug-2012 21:11:18
Kaj, please don't spread your own vision about whether Red will or will not get its own cross platform GUI, as it can cause a false alarm. I know you are a proponent of native or GTK or Enlightenment or other over-bloated solutions, but there is still enough ppl, who will use Red only if it has GUI awailable, and only if it is simple View like solution, and that includes me, so it will come, whatever way it appears ...
GreG
30-Aug-2012 3:46:03
First of all: Welcome back! A 4 days old post and already about 100 comments, this only shows that REBOL is still alive. To me, first thing to do would be a poll: - who wants to use REBOL - who wants R2 or R3 evolutions in which domains - who could support REBOL (documentation, coding, testing, sponsor, ...) I really think we need to assess how the REBOL community looks like and what it is ready to do. So we need a little REBOL script to run this poll :)
Kaj
30-Aug-2012 8:08:53
Petr, I'm not spreading my own vision, just the official Red roadmap. You know full well that I have already provided a cross-platform GUI in GTK+ and am planning another in Enlightenment. They are both cross-platform and native to the platforms that have chosen them as such. It's just that they aren't written in Red.

A GUI written in Red, like REBOL VID written in REBOL, is a huge work that is not realistic at this early time. Even the REBOL GUI is not based on REBOL, it's based on the cross-platform AGG library written in C++. If there is to be a native Red GUI in the future, it will most likely also be based on a cross-platform rendering engine, instead of being written purely in Red.

Scot Sutherland
30-Aug-2012 9:04:59
Brian H,

I appreciate your position on the semantics of compiled vs interpreted languages, but I am inclined to disagree. If it is better to have many languages, why is it that the world increasingly communicating in a single language?

Also REBOL can be implemented as an interpreted or compiled solution, and front to back, top to nearly bottom complete. C became the standard way to abstract and optimize assembly code which is why it was used to create. The only layer left to conquer is the operating system itself and extension of REBOL could be made to do the job extremely well. Plugging all kinds of wonky hardware into the Amiga was an absolute dream. Carl did it once and there is no reason it can't be done again.

The difficulty isn't technological, it is a strong funding/commercial model that will attract hardware developers on the bottom and solutions providers at the top.

Scot Sutherland
30-Aug-2012 9:07:31
Sorry... Second paragraph sentence 2...

C became the standard way to abstract and optimize assembly code which is why it was used to create REBOL.

Scot Sutherland
30-Aug-2012 9:18:13
Seems to me this discussion boils down to two pills as mentioned before:

Keep REBOL as a complete top to bottom alternative that can talk to and influence everything else.

Question: How can REBOL become influential?

----- OR -----

Allow it to become another piece of everything else.

Question: Why do we need REBOL?

tomc
30-Aug-2012 11:17:07
(at)Scot S

Question: How can REBOL become influential?

By being used.

As to used within everything or as an end-all solution the answer is YES. Let it be used and the future will figure itself out.

Question: Why do we need REBOL?

To help enjoy solving more problems than we create

Oldes
30-Aug-2012 13:07:11
Even rusty hammer can beat the nail! I'm still using REBOL and will be in the future, but anyway, please make your decision, whatever it will be, just make it now, because the waiting is terrible. Thank you.
Observer
30-Aug-2012 13:12:11
Folks, REBOL offered some innovative ideas but JavaScript/HTML5/CSS3 is the future for now. Time to move on.
Kaj
30-Aug-2012 13:13:45
Scot, you're right about the breakdown in these two options. Fortunately, we now have more than a decade of data about the first option, to keep REBOL pristine according to the Amiga model, as an isolated island of pureness unpolluted by complexity. Carl's blog finally admits that it has failed.

The second option is what everyone else has been doing in the meantime, and why REBOL is now obscured by hundreds of other languages and operating platforms. They have filled every imaginable niche by being agile, by being open source. But as I said before, without REBOL it turns out that people can't imagine many niches. These are still unfilled, and Red allows us to fill them. Notwithstanding that, one of those niches is a pure unpolluted Red-only system. The existence of niches where Red adds value to other systems does not prevent a pure Red-only system to also exist.

REBOL still has the option to decide to also add value to other systems that are already relevant, and thereby making itself relevant again. I have been gnashing my teeth for a decade seeing all those opportunities and not being able to grab them with REBOL.

Bdug
30-Aug-2012 18:08:40
Rebol is the best language I have ever used but I have always regretted its "unfinished" aspect (bugs) and sometimes hard to predict behavior. Lack of documentation has also been a problem.

Open source Rebol is the only viable option. This seems to me so obvious.

Also, I think rebol has a strong potential to interest academicians (researchers). I dont know if there exist research papers about the formal foundations of rebol but such papers would help to increase interest for rebol.

Bo
30-Aug-2012 21:50:33
To "Observer": I can also create furniture using the back side of a pipe wrench instead of a finish nail gun. It would not be fun, it would not be efficient, and it would not create a quality end product. Have fun with your pipe wrenches.
Observer
30-Aug-2012 22:03:30
Nice try Bo but you have it backwards. As you'll see from many of the above comments, it is REBOL that has fallen behind and is unfortunately slowly becoming antiquated.
Bo
30-Aug-2012 22:27:05
REBOL 2 can do amazing things, and REBOL 3 will be able to do even more amazing things. For example: I'm working on a REBOL project right now for video surveillance. It captures video from IP cameras, has advanced motion detection with variable sensitivity, and records motion including the four seconds before and after the movement period into MPEG videos.

My current proof of concept is recording from two cameras simultaneously at 15fps 640x480 resolution, performing the advanced motion detection entirely in REBOL (30fps total), doing the pre- and post-buffering of images for the motion detection and converting to MPEG on the fly and displaying the moving video in 9 different faces and only using about 30% of the CPU on my first-gen quad-core Phenom-equipped PC (3 years old). And I still have optimization to do and a bunch of debug code running.

Not to mention that my entire business runs on REBOL. Did I mention that I love REBOL?

Thanks, Carl!

Bo
30-Aug-2012 22:35:59
Oh yeah, 13KB total code to do the above including script headers and comments. Let's see Javascript/HTML5/CSS do that. REBOL's VID 'style word makes CSS look like a tangled ball of yarn. Some "solutions" are like using a 53' tractor-trailer to deliver a pound of bananas.

As Carl doesn't have the luxury of time to invest in REBOL, open source the code. I have faith even more amazing things will be possible.

Jaime Vargas
31-Aug-2012 8:02:38
Carl congratulations on your new venture, it is a great to see you engaged and enthusiastic. As for REBOL it is time to let the child face life on its own, form new relationships and find a community to settle.
James_Nak
31-Aug-2012 8:20:04
Carl,

Observer is right: "JavaScript/HTML5/CSS3 is the future for now."

Now, please help SAVE US from this present future!

That "technological trio" is the exact and continued embodiment of what you created Rebol to replace. Its users might not suspect there is a better way but we Rebolers sure do.

Yeah... Pretty please with sugar on top.
Observer
31-Aug-2012 8:45:38
"Now, please help SAVE US from this present future!"

I happen to agree with this! JavaScript/HTML5/CSS3 are amateurish ... but that wasn't the point. Unfortunately that is the direction things have gone and are going, the platform to target for cross-platform apps is the web browser. This is largely driven by the standards process.

Rob
31-Aug-2012 10:06:07
I've used REBOL for years, all the way back to 1999 on the Amiga. I've purchased commercial licenses as recently as last year.

I would like to see REBOL completely open sourced. The reason is simple: I use it in some commercial applications and, if a bug rears its ugly head, I need to be able to fix it quickly.

Also, let me say thank you, Carl, for all you've done, both with REBOL and - more dearly to my heart - the Amiga. I have an original Amiga (later called an Amiga 1000) with all of the team's autographs (including Mitchy's paw print), a boing-ball decal signed by R. J. Mical, and lots of other old Amiga stuff. Still the greatest computing platform ever.

Edoc
31-Aug-2012 12:52:25
I also agree with Observer (although I wouldn't discount REBOL's innovation). A productive conversation would focus on getting REBOL to integrate with platforms which leverage JavaScript/HTML5/CSS3-- as Rondon alludes to above.
Scot Sutherland
31-Aug-2012 15:03:35
Another observation:

OS people and hobbyists want to provide services and solutions...sell their own expertise.

CS people want to provide products.

We decidedly DO NOT have 10 years of data on from closed source approach. Most of the data for CS is not and will never be available because it involves proprietary information. I for one have a lot of IP nobody knows about and can't know about. There are others, including RT itself.

The embedded systems and product folks are just not as vocal on these discussions as the hobbyist/service/solution providers. Carl and Cindy probably have the best idea the balance sheet looks like.

Even then there are some products out there that have REBOL as in important part of the embedded solution that do not have a royalty payment attached to the software portion.

I don't think I have much more to offer in this conversation. We'll have to see what Carl decides.

Kaj
31-Aug-2012 17:00:14
Scot, that's irrelevant. We've heard from the horse's mouth that no money is being made nor invested. That's the situation that needs to be dealt with.
wotWok
1-Sep-2012 7:23:47
"Nature finds its way" .....
Bo
1-Sep-2012 8:16:05
(at)wotWok, it's true, "nature finds its way..." into chaos and disorder. Fortunately, we're talking about intelligent humans designing and dealing with software created by intelligent humans.

I feel a more accurate statement will be that "the community will find the way."

Giuseppe Chillemi
1-Sep-2012 9:37:52
I think the discussion is now a litte off-topic.

Thr summary of the blog is:

We are a cummunity ! We are ready do take the development of REBOL We wish that Carl remains with us to be our oracle.

That's all

Giuseppe

Scot Sutherland
2-Sep-2012 15:17:47
Sorry everyone. Thought I was done.

Kaj,

Whether RT is making money is a very different question than whether AMD or ING Bering are making money in part because of REBOL.

As for investment, isn't that partly what we are discussing here? Just because RT has been unable to make money, doesn't mean money can't be made, and it doesn't mean that REBOL might not become an important part of a funding/business model that does make sense in the market.

Just because those voices are not the loudest or even visible ones, doesn't mean they are not out there.

Respectfully, --Scot

Kaj
2-Sep-2012 16:09:36
I always agreed with that, but RT has been unable to capitalise on all those opportunities. How do you think that happened and how do you think it would change?
Graham
2-Sep-2012 19:24:52
Carl can't spend time on REBOL yet retains an interest in its future. It won't have a future the way things are going. Doesn't matter who is making money or where with Rebol, RT isn't. And thus nor have the investors. Release R3/R2 into the open source world. Rebol lives on. Nothing to stop Carl coming back and creating a closed source commercial product that then builds on the work of others if the license is right. And I have no doubt that he could. That is, releasing the sources is not an end to RT being able to monetize the work in the future.
Maxim Olivier-Adlhoch
2-Sep-2012 20:24:06
Scot, even if RT makes money, even if Carl where to return to R3 full-time...

R3 should be open sourced.

Open sourcing has nothing to do with making money or not. Its about freedom. Freedom from the decisions you are proposing.

A string of decisions, which have relegated REBOL to the fringes.

If R3 had been open sourced 2-3 years ago, the current scene would be very different:

  • It would be past beta stage, probably finished.
  • It would have a version running on all modern platforms including iOS and Android.
  • It would have several GUI engines, networking, multi-threading, devices,etc.
  • known bugs fixed.
  • a growing community
  • a slew of modules for most of the frameworks and engines out there.
  • World & Red might even have been a fork of the main project, instead of a completely new one.
  • There might also have been another face to face devcon.
All this would have cost RT nothing. Open sourcing it wouldn't prevent RT from promoting it just like it was if it wasn't open sourced.

Instead, now, we have 3 different clone projects, each with its own and different semantics and an abandoned, partly implemented R3.

If you want to see just how much complexity can be controlled, look at the red project which retains a lot of REBOL's key ideology. its 100% open source, and its improving quickly, with the help of several contributors.

Carl Read
3-Sep-2012 1:55:14
As a business model, RT could probably do worse than follow some of methods of Jason Fried and 37Signals, those behind Ruby on Rails. Just search for some Jason Fried interviews. There's plenty to choose from.

And curiously, they've been working in a similar space and over a similar timespan to RT. Rails, being just a framework for a language, isn't comparable to REBOL of course. But their core products, aimed at business collaboration and communication, are similar to the products built with REBOL by RT. The big difference? Theirs run in the web browser.

Pierre
3-Sep-2012 14:46:26
Hello, bonsoir,

First, I'm very glad to hear again some news from Carl. Rebol is still alive! I must admit that I had became a bit resigned about Rebol future, these last months/years. Thanks, Carl, for the news, and a lot of courage!

Being an open-source fanatic, I think, like many people here, that open-sourcing Rebol may be an excellent idea.

But dreaming that open-sourcing would almost instantly solve all Rebol's limitations, bugs, etc. may be a little bit optimistic. There is also some danger that the code may become ... less simple? There needs to be a Leader, this point was clearly made already.

A little sad story. I am not a core programmer, but rather a scientist, hacking programs from time to time. And I did feel the "WOW!" effect, years ago (I love this expression! And, by the way, this famous effect HAS to be kept, for advertising purposes, to convince people, etc.). I coded many things in Rebol, mostly little scripts, utilities, daily jobs, etc. At some point, my Rebol console had become my primary shell on my machine at work.

A few years ago, I wanted to port a quite large application that I wrote a long time ago to Rebol. I also wanted to develop several software units dedicated to earth sciences, in Rebol, of course. Since I don't have much time that I can dedicate to programming, I subcontracted this job. The subcontractor said something like "Is Rebol so great? fantastic! Is it free/open source? No? Oh, I won't even try it, then." Too bad. Python and Qt was the choice.

Years passed, I used Rebol less and less; I dreamed of a Rebol on my palm; I waited for R3... And now I write my scripts in Python, I don't think the Rebol way any more. Sorry...

NOW could be another chance for Rebol. I thought about an alternative business model:

Forever:

  • Release an "old" (couple months, couple weeks? depends on the development pace) version of Rebol as open-source;
  • RT keeps working on the next version, close source, saying publically "we are working on this and this part of Rebol. It is going to be amazing!" During this time, the rest of the developers worldwide know that they should better work on other parts of the code, to avoid difficult merging afterwards.
  • RT completes the work, and sells the brand new version of Rebol, still closed-source, using the "old" business model. Only the licence is modified, to explicitely mention that this software will be open-sourced in X months/years. This is to reassure users that this fantastic technology will not disappear and become abandonware.
  • RT ALSO sells expertise, on-demand developments, support, formation, etc. to client companies.
  • The previous closed-source version of Rebol is then merged into the main open-source trunk (this may be the tricky part, especially if the pace is too slow).

By doing so, one could mix the two approaches, open and closed source.

It may also be a stupid idea, though. But I though I'd just mention it; just in case.

Sorry, my post is too long...

Giuseppe Chillemi
4-Sep-2012 9:25:05
As far I have understood Carl has no resuource and willingness to procede to the development of REBOL.

Having it closed source or partially open will stop the whole project.

I prefer a full open source licence and the community will take care of everything during the years.

Giuseppe Chillemi

Maxim Olivier-Adlhoch
4-Sep-2012 9:55:47
Although a noble idea, the "old", open source version would easily surpass the hidden one in a short time-frame.

With many people working concurrently on it and solving real-world problems under delivery pressures they face, I can't see how Carl/RT could keep up.

Jos'h Fuller
4-Sep-2012 11:26:45
Carl,

I love REBOL. I especially love VID*. Thank you so much!

REBOL may be in an analogous situation to Forth (another language I love...) at the NRAO. They liked it, but didn't want to support it and couldn't figure out how to commercialize it so they handed it back to Chuck Moore, who promptly placed it into the public domain. To quote him: "Hindsight agrees that Forth's only chance lay in the public domain. Where it has flourished."

I want REBOL to flourish and be recognized. Even if not immediately, it should be waiting, like Anakin's lightsaber in Obi-Wan's attic, to be raised again anew in the fight for truth, justice and simplicity.

Thank you!

* I could write a book on how awesome VID is.

Jos'h Fuller
4-Sep-2012 11:59:58
Anybody remember Chumby? It was actually fairly successful as an Internet-connected appliance, but (if I recall correctly) you had to program it in Flash, of all things. I remember thinking how nice it would be to have REBOL on that instead.

I wonder, with the recent success of the Raspberry Pi, if a device dedicated to REBOL ("Like BASIC, but better(tm)!") might get some sustainable traction. Set up a very small Linux distribution (enough to launch programs, etc) so that no operating system design is required, but with REBOL as the sole programming language and interface (no Perl, no X-windows). Sell it as an all-in-one beginning computer like the old Sinclair and Commodore machines, but hint at its hidden power -- it would be useful for industrial control/monitoring (where simple GUI construction would be highly prized, "internet of things" outposts, etc., etc.)

This allows Rebol Technologies to avoid open-sourcing (which may be problematic) because it would be tied to a desirable chunk of hardware and people wouldn't notice it so much. I would very much like to have REBOL on something like my Ben NanoNote or the Raspberry Pi. This (might) be a viable way to do something like that.

RT may not be in a position to immediately fund such a hardware venture, but a KickStarter with the "designer of the Amiga" involved ought to pull in quite a lot of the nostalgia vote that powered the recent game console fundraiser?

Tim Guimond
4-Sep-2012 12:06:45
As much as I like Rebol and have used it successfully in a number of projects that I have worked on, I probably wouldn't use if it was open sourced.

Why? Two examples, Linux, and Nagios. We use both at work. Linux is fragmented, putting fixes on it always breaks something else, our servers haven't been updated in three years since our Linux Administrator left. Nagios, beautiful monitor but no one has time to update/support it so we have purchased a replacement.

Commercially successful products need more than a few visionaries, they have solid sales teams, professional support teams, lots and lots of exposure and regular updates.

Carl, find a buyer with venture capital to move Rebol more into the main stream and the 21st century and work with them as a technical advisor.

Steven White
4-Sep-2012 12:43:39
Regarding the concepts of REBOL on a dedicated machine, and a buyer for REBOL, how about QNX (www.qnx.com, a very small real-time operating system that used to have a demo of the OS and a web browser on a single 3.5-inch floppy disk). I know basically nothing about it personally, but I have used the demo disk and it was quite amazing what could be done in such little space. It was like the OS soul mate of REBOL.
Brian Hawley
4-Sep-2012 13:31:34
Steven, QNX was bought by RIM and is now the core of Blackberry OS 10. QNX is not longer likely to be usable for that kind of thing, but there are some others.
Greg
4-Sep-2012 13:56:55
Why not making REBOL open source and creating a REBOL foundation? Anyway, at this stage, I feel that either REBOL won't evolve anymore and its popularity won't expand, OR it is open sourced and get a second chance. I remember the first REBOL day in France, many people discovered REBOL thanks to Dream magazine and publications from Olivier Overlot. I'm also from that Linux generation. Many people there won't even consider something if it's not open source. Just sharing some thoughts to add to this brainstorming.
Chris
4-Sep-2012 20:55:29
Glad you are enjoying your new venture! Will echo the sentiments of fellow REBOLers - open source it, maintain as much an advisory role as feels right. REBOL is the future and must continue to evolve.
Luis.
5-Sep-2012 2:23:42
+1 (at)Jos'h Fuller on the Chumby style idea.

-1 (at)Jos'h Fuller on Closed Source :P

Now, bets on when Mr Sassenrath will reply?...

Cheers,

Luis.

Massimiliano Vessi
5-Sep-2012 5:06:06
Some open source project very remunerative:
  • Red Hat
  • Twitter
  • Facebook
  • Linkedin
  • vmware
Dear Mr. Sassenrath, what are you waiting for?

Jos'h Fuller
5-Sep-2012 19:57:22
Luis, there seems to be some impediment to releasing REBOL as open source -- patents, licenses, whatever -- only Carl knows. I was simply trying to think of a solution that would allow it to stay current and funded without changing the terms of any of the agreements that RT is already entered into (open sourcing would change those in a BIG way, I bet).

If there are proprietary/encumbered libraries used by REBOL that could be replaced by open development (image libraries, etc.), I would be happy to help! But those pieces would have to be identified and the replacements integrated, which is time and money for someone even if the new library is "free". So the question is not easy.

Certainly opening the source could open some doors -- that was one reason I couldn't use it at my day job. But deriving an income from it is no cakewalk -- Guido van Rossum is employed by Google, not Python.

Luis.
6-Sep-2012 6:33:55
:) And he's still the BDFL for Python...

Cheers,

Luis.

Jacob Goodson
6-Sep-2012 13:40
Although this would increase the size of the executable... I think you should build a rebol that runs on the web, whether that be js or dart(hehe) targeted, I do not care. I think rebol should come with an ide, it should be open source, and it should be on kickstarter so that I can give you money.
Edoc
6-Sep-2012 14:12:14
11 days have passed since Carl re-emerged to talk about REBOL's future...

Carl: quick! Please open-source REBOL and be done with it before life gets in the way.

-pekr-
6-Sep-2012 23:17:29
Carl, I am 40 today! So, any b-day presents? :-)
Brett Handley
7-Sep-2012 1:33:59
Carl, first thank you for REBOL.

I'm glad that you still see a future for REBOL.

My context for using REBOL is now simply personal use. Even so, I want to be assured that it will be maintained to support changes in operating systems. While I can't bring myself to learn another lanugage which is not REBOL, we each only have so much time to spend, so I'm not interested in building my libraries any further if they will have only a short lifespan.

I want it to be robust and resolved - I use R2 because R3 is in development.

I want it to be more relevant as technologies, platforms and my needs change - ie. I want it understood in more places like tablets, phones and clouds. I love the small footprint, but maybe it should run on already widely installed platforms like .net or on a jvm so it can be instantly more useful and drive adoption in a larger scope.

I think you'd need an army of coders to achieve these things so open source with a strong REBOL central makes sense to me and/or perhaps a "White Knight" company to support it.

I don't think there's much point protecting against it risks which may not happen - better to let it out of the cage to see if it will thrive. I'd already thought it had died, but since it hasn't why not see how far it can spread?

Brock
7-Sep-2012 11:19:14
I can only thank Carl for his efforts over the years. Rebol is a fantastic tool. PERIOD.

I'm another non-programmer who has made his life easier and produced some fantastic time savers for both my personal life and my professional life.

I'd love to be able to use Rebol into the future. I dream of writing code in Rebol to run on my many Android devices, present and future.

Open sourcing seems to be the right way to go.

Thanks again Carl.

Leon
8-Sep-2012 8:39:28
I've always thought rebol should have been free and open with a licence fee required when commercial software was developed with it. It's never too late I guess so I hope you leave us with at least this kind of business model.
Lennart Fridén
8-Sep-2012 9:54:43
Once upon a time I was greatly interested in and inspired by REBOL. Like Carl wrote, time outside the world of REBOL didn't stand still so I moved on when REBOL didn't evolve in the way I had hoped. Nevertheless, there are certainly still aspects of REBOL that hold merit and in order to spread these, in order to give REBOL the best shot at not completely wither away, I strongly encourage you to open source REBOL.

Put it on Github, give commit access to trusted members of the community and let it take things from there on.

Giuseppe Chillemi
8-Sep-2012 11:00:38
There is a lot of work to complete REBOL3 and make it up-to-date to the latest platforms.

It must be completed and bugfixed

It must be extended and to the latest communication technologies.

Documentation must be modified

All previous technologies updated

Porting to moder platforms

We shouldn't also forget about moving all the documentation and technologies from closed to open source.

It is an huge work that only an organized community can complete.

Fully open sourcing REBOL is the only way. A company needs at least 300000 USD to accomplish all the tasks.

Giuseppe Chillemi

luce80
8-Sep-2012 11:37:03
If you do not want to open source it, make at least a shared library with the necessary functions to "do" some code so we can build all the rest (gui etc.) around it.
HRO
8-Sep-2012 11:41:20
Computer Programmer here. First time I comment in this community, but I have always kept an eye on REBOL, since the days of C= Amiga. For scripting purposes, it is hard to forget how REBOL feels: practical, easy, clean and terse. Cannot say the same about the untidy knot that is JS/HTML/CSS. Whatever the decision, I would like REBOL to survive and be available.
Kaj
8-Sep-2012 13:16:20
Luce80, that's how REBOL 3 already is. Boron, too, by the way.
Pierre
8-Sep-2012 14:55:39
Interesting points of view have been exposed here. The number of comments posted here gives an idea of the level of interest and expectations about Rebol: high.

Now, I feel that we all are quite anxious about your next post, Carl.

To Maxim Olivier-Adlhoch: in fact, if things are well organised (yes, open source CAN be organised... at least, I hope), the main branch can automatically incorporate (after a few checks, of course) all open-source changes. The "hidden" part, where closed-source development takes place, has to be publicly mentioned, so that no open-source developer should interfere with the closed-source dev.

To Tim Guimond: there are also examples of successful open-source developments: qGis, for instance, has evolved at a supersonic pace over the last years.

To Greg: it looks like we have read the same Dream magazine, and the same book from Olivier ;-)

To Jos'h Fuller: I heard a few times about the impediments you mention. What would be the problem, exactly? Wouldn'it be possible to isolate the "litigious" parts of the code, while releasing the rest as open-source? Then, obviously, the missing parts should be rewritten from scratch, or taken from various existing libraries. To Kaj: and Red, also?

Now, Carl, whatever your decisions are, please show some signs of life. And, just in case, please note that my birthday will occur in less than a year... ;o)

Alex A. Naanou
8-Sep-2012 14:58:35
Open sourcing is the way to go, IMHO.

I've been following REBOL for years and the only reason I'm not using it in either my teaching, research or production work, is the licensing/openness requirement for most of what I do.

If you do not want to lose control I'd suggest GPL/LGPL and if you really want to make the project open to the community use BSD/MIT (my choice for my own work).

Thanks!

KSH
8-Sep-2012 15:19:37
Carl,

Would it be feasible to open source Rebol 2 immediately, and delay a decision on R3 for six months or a year? That would ensure a growth path for people who have invested time & energy into R2 solutions, and should also generate interest among non-Rebol open source advocates in the programming community.

It would also likely lead to divergent code bases, but I don't see that as a huge disadvantage. After all, Indo-European split into Latin, Slavic & Germanic branches, but English mixes aspects of Latin/French and German/Anglo-Saxon into a language that's reasonably expressive while still maintaining its ability to be erudite & earthy.

Once an open source ecosystem has grown around R2, RT could make a decision whether to complete R3 either as a proprietary solution or as an open source project under a foundation similar to Mozilla. I'm thinking that, in the same way SQL serves as a database language in both proprietary & open source forms (Oracle vs. MySQL), REBOL could serve as a general programming language in both open source and proprietary forms.

If, by the time a decision is to be made on R3, the R2 code base has diverged into Lisp-y, Haskell-esque, Java-like and Smalltalk-ish variants, it would be up to RT or the foundation to decide on what would constitute a suitable, flexible, general purpose and (hopefully still small) core for the release of R3.

One concern would be whether issues like a lack of unicode support in R2 would be serious enough to prevent open source advocates from getting involved in the project. Unfortunately, I don't know enough about what's under the hood in R2/R3 to know how difficult it would be for an open source community to port some of the more modern features of R3 back into R2.

I agree with all the comments above about REBOL's unique merits, and would hate to see future programmers miss the chance to experience its elegance, simplicity & productivity. So if the above option isn't feasible and it comes down to a choice between open sourcing everything or letting it die, I'd strongly support open sourcing it all.

Whatever happens, thanks for all the time and energy you've put into the project so far. Maybe the 2020 or 2030 vintage will be above-average, and you'll be enjoying it in semi-retirement with some sort of unofficial title as either the Godfather, King, Czar or Guru of REBOL.

shadwolf
8-Sep-2012 16:40:27
Thank you carl for the invitation to comment your 2 years after the war mea culpa. I invite you to read the comments people left on your blog during that time you took out of the town. and read your rebol world.

One think that makes laugh so mutch is you put so much effort into rebol to end throwing the baby with the bath water...

Glad to hear from you even if it's a pile of B-S and whiny comments, about your own lacks and own ways to disrespect your followers.

shadwolf
8-Sep-2012 16:51:22
your time is limited so here is the truth. Rebol community dead. You will never ground a financial issue on them after those 2 years they are disgusted of rebol and you

Rebol is stamped since 10 years a failure. If you want to reboot the project you will have to release to public domain the source code act as a inspiration lead the folks but let rebol live on it's own and accept what the people want to see in it is not what you want.

Rebol is dead actually after the two years of silence you put us through there is no easy comming back dude.

Actually rebol is making no monney put rebol in LGPL put a link in donation and I will donate each time something that please me is released by the freedom devteam-

Carl you speak about revolution ? your revolution is a communist one leaded by the worst ever dictator. You forget their ain't no revolution without freedom.

aka the only banned guy from your rebolworld-

shadwolf
8-Sep-2012 17:00:02
CArl all we ever asked to you as community was to give a ending to things in rebol.

For example VID we never said VID was the worst thing ever we just wanted a more meaningfull set of widget that is why rebgui was made en proposed.

Reb webbrowser was awsome dude ... think of not flash java or silverlight are on par with it but YOU choose against all advise to focus on activeX and IE this is YOUR decision and YOUR messed up. Since then you totally forget the idea.

Etc. etc. etc. rebcode?

Sure r3 introduce better things but it is stamped as alpha unstable and actually give you no money this is dumb. see kickstarter projects see blender seee libreoffice. you need to associate with a big shot and make rebol explode this is your work this is your lack. Or fully delegate. The actual situation is the worst

trans
8-Sep-2012 18:50:13
The suspect the best advice I can offer you is simply this: Learn to let go.
Proteusguy
8-Sep-2012 19:16:51
Every successful language/platform ultimately wins because it is open. The more the better. I refuse to use proprietary systems at our company unless we don't mind rebuilding it from scratch in 9 months (which is never thus far). I agree with the comment above that REBOLS's situation reminds me very much of FORTH's.

I would follow Python's example. If REBOL is ever to truly thrive then a truly open license (i.e. NOT GPL) owned managed by a foundation of volunteers with commit privileges will be the best chance to get it there.

I'm particularly interested in seeing REBOL get into restricted resource environments like mobile. Being able to work cross platform in mobile/desktop but still keep strict control of my resource usage is something I can only do in C++ right now and most of those apps don't warrant the complexity of C++.

I love cool programming languages. I'd love an excuse to bring REBOL to bear on some of our projects. Can never happen if it's proprietary stuff or GPL.

Nikita Dudnik
8-Sep-2012 21:55:06
Just open source it. I looked into Rebol before, liked it, but didn't use it because it was proprietary.
John S
9-Sep-2012 3:15:17
Carl, I have two suggestions for you: 1) Adopt a dual open source/commercial license for Rebol like BerkeleyDB used at the beginning with its Sleepycat License. This will allow you to open it up, but control commercial use of the system and perhaps even pay for additional development and/or maintenance.

2) Establish a language foundation for Rebol and a community proposal process. Python may be looked at as a model for what this sort of process could be. This will open up governance of Rebol to more people, while preventing the sort of free for all that would be bad for the language. It would also open up development to an even wider community.

Doing these two things would give Rebol the best shot at a lasting future. Progress in the system would no longer solely rely on your efforts, but you would not be giving up total commercial control of your project.

http://en.wikipedia.org/wiki/Sleepycat_License http://www.python.org/psf/ http://en.wikipedia.org/wiki/Python_Enhancement_Proposal

Kind regards, John Siegrist

Robert M. Münch
9-Sep-2012 5:27:40
So, when will something happen now?
just saying...
9-Sep-2012 6:46:36
why don't you and Doc Kimbel unite your forces together? he seems to be the only one serious enough to start a new language to address the weaknesses of rebol and its dev release model ?
Nick
9-Sep-2012 8:04:36
Carl, I have years of gratitude saved up for you and for what REBOL has provided me, so I will be endlessly respectful and patient. But please be aware - we're all holding our breath here, now, since you posted. I hope you are encouraged by the continued interest in REBOL. Everyone here clearly has respect for you and what you've created, and would love to see REBOL continue. I have to wonder how many more people will be excited to see this post, and this conversation (not too many people had reason to check this blog regularly during the past year). You asked for suggestions, and initiated a conversation. The response so far is resoundingly hopeful about open sourcing REBOL and allowing it to flourish.

Can you give us any feedback?

florin
9-Sep-2012 11:26:39
Watching rebol now for a couple of years.

Cautiously and respectfully I say, the manner in which Carl treats the rebol fans is profoundly insulting. Not only he is no leader, but he lacks in common human decency. Whoever continues to follow Carl, displays a sorry sense of self respect.

Kaj
9-Sep-2012 12:31:14
I can't disagree.
Gregg Irwin
9-Sep-2012 15:53:45
Wahoo! Thanks Carl! Wait, that won't tell you anything. I imagine the partying will begin as soon as the license is finalized.

THANK YOU! THANK YOU! THANK YOU!

P.S. I'm glad no woozles were there to make things worse.

Kaj
9-Sep-2012 16:05:03
Say what you will about REBOL, but it still has the best trolls.
James_Nak
9-Sep-2012 16:05:05
I'm glad I read more than the first two lines of your response Carl. Doomo Arigato!
Gregg Irwin
9-Sep-2012 16:47:12
Grrrrrr. Thanks for catching it Kaj. I'll get that troll for impersonating me in his reply to Carl as well. ;-)

Desperation breeds belief.

Brian Hawley
9-Sep-2012 16:49:34
Good news everyone!

Be careful about creating your own license. "Open source" is trademarked, and they use that trademark to enforce certain limitations on what you put in the license. We don't want to be tripped up like last time.

Still, it really is good news and we hope to see it soon!

:-)
9-Sep-2012 17:12:31
Come on Brian, read whole thing.
Jerry Tsai
9-Sep-2012 18:10:57
Real Carl should be marked as yellow. I don't think this is him.
John
9-Sep-2012 21:32:20
Open source is the best way to go forward. The sooner the better.
Robert M. Münch
9-Sep-2012 23:07:16
Carl, please just pick one license. There are surely ones that fit. Don't re-invent the wheel. It's a waste of time. Use MIT to make things simple.
Arnold
10-Sep-2012 0:14:25
Jerry is right. Though the style of the post is convincing and partly what 'we' expect to read, it really should be 'Carl Sassenrath' and it should be yellow. (See for example http://www.rebol.com/cgi-bin/blog.r?view=0187#comments ). So nice try, REBOL does have the best Trolls. Now Red can print 1 we soon will have our needed Open Sourced REBOL.
Jerry Tsai
10-Sep-2012 2:20:58
It's not the first time that fake Carl showed up here in Carl's blog. It's not fun at all to do that.
Carl Read
10-Sep-2012 2:30:13
Florin, try and see things from Carl's POV. If he knew what to do with REBOL that satisfied his needs and the users' needs, I'm sure he'd have no problem explaining it.

But my guess is every path he's thought of so far does not tick enough boxes in his eyes. If that's so, anything he could say now would probably be unsatisfactory to all concerned. Best to wait until he can say something definitive.

Carl Read
10-Sep-2012 2:37:51
REBOL's Possible Future:

It's released as Open Source. It grows nicely into an up-to-date, complete and reasonably popular language while retaining it's core (and 'simple') virtues. Then, a few years down the track, Carl writes an OS for (or with) it.

Arnold
10-Sep-2012 2:49:04
That would be CarlOS?

REBOL's best thinkable future at this moment: REBOL becomes more obsolete month by month. Lucky for us one of it's offspring is growing slowly into maturity. Then it is bye bye and thank you REBOL, hello Red.

Time is running out for REBOL.

MaxV
10-Sep-2012 4:40:13
http://en.wikipedia.org/wiki/Open_Software_License
RobertS rshiple
10-Sep-2012 6:54:36
Glad to see discussion of open source options.
Bo
10-Sep-2012 13:35
So, I looked at Git-hub and I couldn't find anything. Am I gullible?
Carl Sassenrath
10-Sep-2012 14:39:12
Just a note from the real real Carl, not the virtual Carl who posted above. (Sorry fake Carl, I've removed your comments, but they're in the right spirit of things.)

Someone is spoofing blog comments under my name. But, not a big problem... as they're fairly accurate... almost.

I'll get it fixed tonight and get back to you.

-Carl

Carl Sassenrath
10-Sep-2012 15:06:22
Removed div and span tags from those allowed in comments.

Nick
10-Sep-2012 16:43:46
Excited! ...
John
10-Sep-2012 21:35:46
Let me guess ... at least half of us here have written a one liner rebol script to alert us to when this page changes :-)
Jerry Tsai
11-Sep-2012 6:03:24
Make a decision now when it still matters to us.
Avasi
11-Sep-2012 6:55:15
...tic...tac...tic...tac...tic...tac... (at libidum)
Pierre
11-Sep-2012 6:57:47
forever [prin "tic... " wait 1 prin "tac... " wait 1]
Fork
11-Sep-2012 17:03:02
I generally stand by the idea of "better late than never". So to my ears, this is good news!

Oddly enough: I was going to let my "freebol dot org" site lapse on the 17th. But if this is for real, and I might actually get to try and build my "Firefox" to Rebol's "Mozilla"...then I'm certainly in to give it a shot.

It has not been clarified if Rebol's implementation depends on third-party technology that would limit the ability to open its source (or on investment capital that restricts Carl from making the license whatever he wanted). If those are issues, it changes things a lot. So it would be nice to know if this is in the picture.

I personally believe in the Stallman notions of software freedom. So I'd have suggested the GPL except that this is one of those textbook cases where he argues less restriction is needed to come in from behind in a competitive arena. Seems the LGPL is the fit: let people distribute their Rebol-based programs with an optionally modified Rebol executable, just as long as they give the source for the mods to anyone they transfer the program to.

(If anyone thinks their Rebol modifications are commercial gold and don't want to lose their competitive advantage, they need to write Carl a check. Also, it's always easier to relax a license than to restrict it...so no need to go straight to MIT/BSD right away.)

For branding: do what Firefox and Chrome did. Keep the Rebol trademark, name, and logo in the legal hands of RT. Since Carl doesn't want to be managing these builds (or the website), delegate these chores to people who are trustworthy enough to do it.

Let me know soon so I can tell whether to renew that domain or not! :)

Pierre
12-Sep-2012 0:43:24
(at) Fork: I have just read your site: concise, wise. I fully agree. Please, do NOT let it vanish.

Just in case, I copied it on my machine (what a pleasure to hack some Rebol lines: I didn't do it for several years, I feel like a happy child playing Lego!)

>> write %freebol.org.html read http://freebol.org/

Now, this is a bit more disappointing:

###ROOT### < 15/08/2012 13:03:44 > root(at)autan:/etc# apt-get install freebol Lecture des listes de paquets... Fait Construction de l'arbre des dépendances Lecture des informations d'état... Fait E: Impossible de trouver le paquet freebol

I am dreaming of the day when this command will be successful...

Art K
12-Sep-2012 5:11:40
Open Source does not mean free. Go Open Source.

New business model:
- nominal royalty for View, say $2/yr (less in places where $2 is a lot of money...)
- much, much lower SDK prices

Revenue pool for major developers, royalty for Carl, awards for top library scripts.

If the user does not renew the $2 fee, product stops working.

REBOL is a great intellectual/programming success, but needs a strategy to keep it keeping on.

Luis.
12-Sep-2012 6:10:54
(at)Art K: -1 on the "product stops working."

Anything but crippleware!

Cheers,

Luis.

Brock
12-Sep-2012 8:11:47
Crippleware is sure harsh, but I like the concept. Maybe do as all others have done and create some kind of store with minimal charges for apps or code. People think nothing of buying apps for $1 or a few. Increase view to a minor $10 one time fee, same for the other useful modules for $5 ie; sql, odbc, jpg, encap, etc. Apps and other code gems or $1 or $0.50 and everyone is making money.
Maxim Olivier-Adlhoch
12-Sep-2012 11:00:07
I don't think this is about the money as much as the management of the project (ownership, design, management, coding, releasing, "evangelism", etc.) .

Carl, you should just release it, give one or more trusted peers commit access to the "official" repository, and let it go from there.

REBOL, and its trademarks are still yours and you will retain the right to remove commit access and/or fix any divergence from your POV later.

It worked very well for the Host kit, and many of us did great things with it in a matter of hours. I suspect the same will happen with the rest.

The web site maintenance can also be given to trusted volunteers. There are people who would be glad to give it the work it needs without rebooting it using wordpress.

Fork
12-Sep-2012 11:12:26
I'm fairly certain that for Rebol to gain traction it has to offer at least one non-commercial open-source story. Offering a LGPL license option is fine. The only issue is if RT wants to bring in changes from the ecosystem, then be able to grant specialized licenses on a Rebol.exe that includes those changes. Then the relevant contributors need to grant permission for that.

This would prevent integration of existing LGPL code into RT's binary, but it's doubtful that Carl would ever want any of that code anyway. Probably just BSD/MIT and from-scratch contributions from people in the existing community, none of whom I think would have a problem with that clause.

Only R3 need be opened up. There are going to be enough divergent efforts as it is...and hacking half-baked Unicode solutions into R2 (or other revisits to the past) will just divide community effort and attention.

-pekr-
12-Sep-2012 11:13:46
Carl, if REBOL's going to be open-sourced, please save us from crappy licences like whatever-GPL, this will burry the thing under the minute it is released. That's not how openess should look like ...
Fork
12-Sep-2012 11:55:25
Don't disrespect the GPL. Think about the pervasiveness of computation in our lives...in medical devices implanted into our bodies...running the cars we drive. Even more of this in the future.

We should have a right to inspect the code running on our computers (or ask trusted experts to do so) so we know what it is doing. We should have the freedom to modify and adapt the work if it does not suit our purposes as written. If Rebol people fancy themselves "gurus building the future" then the utter utter obviousness should bonk you on the head.

(It really shouldn't just be sci-fi writers like Cory Doctorow -- or mediocre software designers like Richard Stallman -- who are the only ones who get it.)

While Stallman's software aesthetics may not be to my tastes, his philosophy is prophecy:

Why Software Should Be Free

Rebol is a great tool for makeing software which does "big things" need not itself be big...literate and domain-specific expression. It dovetails perfectly with the open source movement. If software has a revolutionary angle, then a GPL-style license would help prevent a situation like the Apple OS tyranny we're seeing happen right here (in our own time).

mad informatique
12-Sep-2012 17:02:22
hello ! Carl, do you remember Moissac France 'mad informatique' Franck, the circle is closed... (i'm not drunk) after Amiga, Windows, Apple OS X, Linux and their spécialites, an big idea come to me ! and going to 'amiga impact' the french forum dedicated to amiga, for find enthusiast help me in this big adventure, and reading first post ; and i say is not possible : the future of rebol ! ... is link for start another circle... i'm serious Carl, i can't explain an NDA is required, if you contact me, sure you wil not disappoint. please search 'mad informatique' 9 Place des Récollets 82200 Moissac France, Franck
Brian Hawley
12-Sep-2012 17:06:54
Nice to see you back, Fork!

All other things being equal (which they rarely are), (L)GPL software doesn't do very well against permissive licensed competitors, particularly programming languages and libraries. You have to consider in this case that most of REBOL's direct competitors have permissive open source licenses, even some REBOL derivative languages. Past experience has shown that (L)GPL licensing has discouraged many qualified people from contributing to REBOL-like languages, i.e. Boron, R# and Orca.

You also have to consider that the set of people who are qualified to help advance REBOL is not very large right now when compared to the number of people advancing non-REBOL-like languages. Many of these people are also contributing to the REBOL derivative languages, mostly Red and Topaz, or are interested in doing so. We want to use a license that would allow Red and Topaz developers to help REBOL as well.

The more we cooperate, the more we can advance together.

Louis
12-Sep-2012 21:56:55
Hi Carl,

It is good to see your post. You have been a great friend, and I pray for God's blessings to be upon you.

I've been doing some programming in Steel Bank Common Lisp (SBCL) for a while now, and I cannot help but compare it to REBOL.

=== REBOL is much easier to learn. I've written programs in 8 different languages, and REBEL was by far the easiest and most fun to learn.

=== REBOL is much easier to use. No difficult setup. See results instantly.

=== REBOL is much smaller. A SBCL executable is 20 megabyte minimum!

=== REBOL is much faster for developing GUI programs. Developing GUI programs is suprisingly difficult with SBCL.

=== REBOL is much better at messaging.

SBCL has faster execution speed (which I needed for one particular project), a much larger users group (which makes it easier to get help), and is open source (which speeds up development by attacting new talent, assures that Common Lisp will not die, and helps the users group to grow). Several of the developers of SBCL stay quite busy providing paid support to other programmers, and doing custom programming using SBCL.

For most of my programming needs REBOL is by far the best language to use. With the speed of modern CPUs, speed of learning and using a programming language is generally far more important than speed of execution. Every tool has its use, but some tools (such as REBOL) are more generally useful than others.

I suspect that REBOL use will skyrocket once it is made open source. Open sourcing REBOL is the key to its future. Over the years, many programmers turned their back to REBOL the moment they learned it was not open source. Maxim Olivier-Adlhoch, Gregg Irwin, Bo, Brian Hawley, and Henrik have given good advice, I think; there needs to be an official version of REBOL, maintained by the programmers you know to be the best. Note that almost everyone wants REBOL to be open source. The risk of using closed source software is just too great.

But most important of all, you need to make public a clear decision immediately, so that people depending on REBOL for their livelihood will know what they have to do.

Luke 6:38 Give, and it shall be given unto you; good measure, pressed down, and shaken together, and running over, shall men give into your bosom. For with the same measure that ye mete withal it shall be measured to you again.

Many thanks for all you have already given. REBOL has enriched our lives, and enabled us to do things we never even dreamed of before.

Fork
12-Sep-2012 22:10:44
BrianH: Thanks...but I only left AltME. I've been ready to pounce on any Carl blogging when it happens. Especially when it's this significant. :)

The more we cooperate, the more we can advance together.

Well put.

To me, money is scaffolding...the sooner we destroy "trusted authorities" who print it, the better (God bless Bitcoin - we should all be our own banks). But while we wait for that Utopia...if I were Carl I'd do a dual-license and keep that wedge, to try and stop hostile forks of Rebol that don't give the code back. Say what you will about Qt, it's still getting bought and sold...and was one of the few pieces of value Nokia could capitalize on when they went for broke.

I know Carl isn't that concerned about the money (he's the type who need not be in general, and esp. since he's got a new gig). But keeping a tight license that embraces the do-gooders can't hurt for a start. If he wants to loosen it later he can... much easier than going the other way!

Brian Hawley
12-Sep-2012 23:38:43
The problem with a tight license, the problem with control, is that maintaining that control takes a lot of work, more work then he has time for now. If we want people to do that work then we need to get rid of the barriers that would prevent them from doing so. There are looser, more subtle ways to keep things on track.

REBOL is an easy enough language that people can be more valuable than code. When Red adds a module system, very little of the code that I wrote for R3's module system would apply there, but everything that we learned while writing it could inform the design, and our experience would be invaluable. The same goes for porting Topaz's parse extensions to R3, or using the knowledge we gain about JNI integration in both Red and R3. More important than keeping the code portable, we need to allow the people and knowledge to be portable.

As for forks, hostile or not, we need them. There are OSes in common use today with very different operating models. While the R3 model is great, it won't be suitable everywhere. Sometimes we need to be different on different platforms. Only divergent projects can do this. But the fork is not really hostile unless it inhibits sharing with the other forks.

Ryan Cole
13-Sep-2012 9:51:15
After giving it some thought Carl, I now believe REBOL Technologies has much potential. It just got derailed a bit, but the situation is very recoverable.
  1. This may make some of you puke and shiver, but for REBOL needs an HTML/CSS/Javascript client. Don't think anything as nightmarish as REBOL 2 or 3, but mostly just a REBOL VID type client that can run in a browser, and some basic core functionality. This is now quite doable.
  2. RT's product is not the language. RT's product is professionalizing the REBOL developers. Docs are great, but there needs to be a pro REBOL developer program that really means something incredible. Pros don't just pay $225 for a license as they do now, they would be an exclusive corp of expert developers who had to pass RT's version of the 12 labors of Hercules to earn keys to the castle. This way it is something for all REBOL coders to inspire to be and spoken of with a great sense of pride. To this end, the website needs to make REBOL developers seem like titans. The merits of language are a side note, because people first grasp the emotional content anyway--of which the current site is devoid.
  3. A Java port is also important.
  4. Integration into voice API's will soon be critical, as developers will be leaving their comfort zones to explore conversational computing as it takes hold.
Ryan Cole
13-Sep-2012 10:10:08
I should add that it self boots.

These "labors of hercules" (more accurately rites of passage or challenges) give you the developer power for a HTML/CSS/Javascript client. Once it is shown that REBOL can simplify the HTML/CSS/Javascript quagmire, rebol WILL get many new developers--and many old ones back.

Warren
13-Sep-2012 13:14:12
Artk: You really thing that developers would leave a "disable on license renewal check" in an open source program? You really are out of touch.

REBOL should be open sourced, and GPL would actually be a smarter license for Carl to choose if he wants to make money. But the money comes from people who pay for your brains and your help, not for silly things like license locks and expiration timers.

W

blah..blah..blah...
13-Sep-2012 13:20:51
blah..blah...blah tic...tac...tic...tac.. blah...blah...blah... tic...tac...tic...tac.. blah..blah...blah tic...tac...tic...tac..

No wonder why Rebol is where it is today...

claude
14-Sep-2012 0:47:27
i will be happy to pay for a rebol3 with a great VID,ODBC,SQLITE,WIN,OSX,LINUX,etc.... like i did for rebol2 sdk ;-) Carl don't stop REBOL futur !!!!
yebird
14-Sep-2012 0:51:02
Giant Carl.Please let us stand on the shoulder of you.
-pekr-
14-Sep-2012 2:06:55
What we need is a final decision. I think that we have expressed our personal wishes quite extensively already ...
Art K
14-Sep-2012 3:48:47
It depends...The "Contributor License Grant" might be used to control developers. But, maybe none of the model Open Source licenses are suitable.

It was my understanding that open source software allowed participation, but not in an open slather way that would hand the farm over to everyone and anyone that some of the open source licenses seem to do.

One might have to write something unique to the ambitions that are held for REBOL, allowing participation, but maintaining some form of control.

And, who would begrudge such a reasonable fee? Its just a way of encouraging participation and future development, as well as acknowledging that writers should be paid for their work - as much as practicable.

http://source.android.com/source/licenses.html
http://www.ilaw.com.au/public/licencearticle.html

Luis.
14-Sep-2012 5:34:56
(at)Art K: Granted, there is the possibility for REBOL, and often is the case, that Open Source licences fork out to disparate versions of the initial system/release.

Regardless, most will tend to go for the origin or the one that trends, in their view, to what is closest to the original. If a fork is followed it may be that it is either an improvement on the original or represents a better fit to the current user.

I suppose a fork, in a sense, can be used as a measure as to 'best fit'.

Cheers,

Luis.

Certificate Industry's e
14-Sep-2012 12:03:32
If it is the certificate industry you want to pull into REBOL mr Ryan Cole, that is what I can do without. I truly disgust this industry where all kinds of people and company's filled with parasites are sucking money out of programmers pockets.
nemy
14-Sep-2012 12:05:27
!
Elwood
14-Sep-2012 12:19:03
> I must admit that I'm quite addicted to it.

It seems this other project is very important, maybe more important for you than Rebol. :-) If some third party developers are willing to work on Rebol core source then release it as open source but only if people are interested.

I don't know the license scheme but you need to do licenses for companies. This is where the money is ;-) Rebol then may be free for personal use.

Fork
15-Sep-2012 1:03:47
(at Pierre) Well, since I had at least one proponent...I did renew freebol.org, for one more year. I hope a decision comes much sooner than that! :-/

(Site narrative note: Clearly I would not second-guess the unicode implementation of the R3 core if it were open sourced. That part of the plan was only for expedience in cloning from a spec without source.)

If legal issues tying Carl's hands are no barrier, it doesn't seem that big a deal to GPL license Rebol's source today...at noon! Might the license be too restrictive in an ultimate sense for the current programming milieu? Okay, sure, maybe it's not the endpoint. But at least in the meantime people like me (who actually prefer the GPL all things being equal) could look and think about what to do. Not all of us have had BrianH & co's level of access, so we don't really know what state the artifact is in! It's hard to advise in that condition.

Yet I do have a little bit of X-ray vision as far as computing.exe goes, and I think I get the gist of what's finished and what's unfinished. Rebol doesn't have a solid multithreaded story yet...and I doubt it will be productive to try and sort that out for R3. I'd save it for "R4" and just hold close the "it's so small that running multiple processes that message each other is okay"...worked for Node.js, didn't it? Rebol could knock them on their ass in less than a year by copying the approach with stronger foundations.

Incidentally: my Freebol fork would beeline for taking out everything I've call a "speed bump". No more func... the existing function would be remapped to what funct is today. func would become something like macro and the existing useless "function" would die. I'd bring in C++ style and Javascript comments instead of the semicolons. I'd make nationalized versions that remapped the words and let people program in their own alphabets. You bet I'm going to rip out that /only semantic debacle too. :)

Point being: I have visions, and so do others. Carl and BrianH may bristle at them...but am I actually worse than letting the rest of the garbage win? Mozilla was saved when Firefox saw the quality of work being done which was disconnected from reality. While they're no longer the best browser, that revolution shifted the discourse.

Rebol may not be the crystalline endpoint of programming (I'm actually a C++11 fanatic and my group is hosting Bjarne Stroustrup here on Wednesday), but it beats the pants off of interpreted languages in use now. A little bit of liberation could be the key to letting it inject some actual good ideas into the noosphere of programming. This opportunity will not assure success (nothing in life is for certain), but I pray Carl gives us a fighting chance to try.

Carl Read
15-Sep-2012 3:01:06
Fork said: "I'd bring in C++ style and Javascript comments instead of the semicolons."

And I say nooo!!! I like REBOL's white space and little clutter. I want to focus on the words and not semicolons and full-stops and commas and empty parens and the rest.

Kaj
15-Sep-2012 5:59:28
Carl made the best argument why not to use closed source in the past two years, and Fork just made the best argument for Carl not to open source...
Carl Read
15-Sep-2012 6:33:35
Indeed. Development grinds to a halt when he who has the closed-source goes AWOL. And open-source forks could produce an incompatible mess.

Which is why I suggested controlling an open-source REBOL with datatype specs. Compare it with web-browsers and HTML/CSS/Javascript. Whether the browsers are open or closed source doesn't really matter. What matters is that it's agreed what the HTML/CSS/Javascript specs are and that the browsers aim to interpret webpages according to those specs.

In REBOL's case, the browser is parse, it expecting values that conform to its datatype specifications. The datatypes are what needs to be known as REBOL.

Giuseppe Chillemi
15-Sep-2012 6:48:48
We need to focus on solutions than the language to make money. REBOL is the tool we need to use to have solutions. We have two choices:

1) Focus REBOL Tech. forces on the language and then solutions.

2) Focus on Solutions only and give to the community REBOL.

The second one is more suitable for the current market and resources of REBOL Tech.

Giuseppe Chillemi

Fork
15-Sep-2012 7:27:58
Several years ago I supported not fighting with Rebol as a standard, rather to have a published spec. But that was years ago. At this point we'd do a lot better to throw some things at the wall and see what sticks and what emerges. Centralized design has failed us, and failed us hard.

Unlike some people, I'm not going to say Rebol needs a string concatenation primitive, or to abandon a raw character type. But I'm not also clueless enough to think that the existing codebases in Rebol have any value relative to what the language could become if it actually looked like something that wasn't downright archaic. Most of my suggested changes are mere mezzanine shifts.

I also know all-caps REBOL vs Rebol makes a difference in perception. The Ikari warriors power-up icon is hideous, and I spent my own time and money developing a far better logo. Please note that Kaj's "try rebol" variant persuaded exactly zero people that Rebol was anything other than an outmoded joke. (All to preserve compatibility with Konqueror, which then switched to Webkit BTW.) I'll stop rehashing that history now.

Also remember that Douglas Crockford tried fairly hard to engage, but left and went on to be highly influential in the web community. I've been warned on my blog that Rebol is a "language that attracts cranks", and that's scared away a lot of people. But check the Tiobe language index and notice Rebol ain't on it...so someone deserves a chance to try their hand at getting a Rebol-based variant on that list.

Would you prefer the other 100 languages to be in the top 100? Is it REBOL in all caps with ugly distinctions between func and funct, and /only ; OR NO REBOL AT ALL?

As BrianH says: "As for forks, hostile or not, we need them." I like the GPL because then everything shares DNA, you reject what you don't like and take what you want.

Red just made its first executable, so it's time to agree...even if we agree to disagree. Waiting another year for a license decision isn't sane.

Brian Hawley
15-Sep-2012 9:34:17
Carl Read, you might want to consider that the internal in-memory datatypes are what makes Rebol. The external syntax, which I normally like, is just surface stuff. The load function is just that, a function. In R3 a good chunk of that function is even mezzanine. You could just as easily write other loading functions that implement different syntax. As long as the datatypes that the loader generates are the same, the other code that uses the data need not care about the difference.

Fork, it's not the design process that's failed us, it's the social process. That's what needs fixing. But we don't need to replace one failed social process with another one that will also fail.

Can you point to a language in wide use, just the ones where the language infrastructure is combined with your code to make a complete application, with a GPL-licensed runtime? Consider that encapping is basically statically linking, from a legal standpoint, and that most code that people write is, for the moment, necessarily proprietary. A GPL programming language with a runtime model like that of Rebol wouldn't promote sharing, it would bring legal problems to most Rebol programmers who saw the GPL'd source at all. Plus, it would kill the chances of Rebol, because all of the other languages in the top 100 can be used to make proprietary apps, and Rebol wouldn't be able to. There are even major platforms with app stores that aren't legally compatible with GPL-licensed code, or any license that requires distribution of the source with the binaries.

The rest, the comment style is just a syntax thing, though you might miss having a modulus operator. The function naming thing is just that, you can rename anything you want in your own stuff, though you might want to do that in a module for compatibility. There isn't much about Rebol that's very fixed - that's the beauty of the model.

KSH
15-Sep-2012 11:00:53
I'm going to agree with Brian that it looks more like a failure in the social process than the design process. I haven't followed Rebol for the last few years, so I'm finding the mixture of intelligence and childishness in these posts kind of strange.

Hopefully there's not a significant perceptual difference between REBOL in all caps and Rebol, so I'll continue my earlier convention of using REBOL to refer to a hypothetical standard (similar to HTML, CSS, & JSON) which specifies a general method of packaging and processing code and data to enable easier computing in the internet age.

Rebol in init-cap, then, will refer to a particular implementation of that standard which has been developed into two versions, R2 and R3, by Rebol Technologies (which I sometimes call RT, although some may quibble and say it should be rT since the "proper spelling" is reBOl TeCHNoloGiES).

In my earlier post, suggesting a "test drive" of the open source development process by releasing R2 immediately and delaying a decision on R3, I had to hedge on issues like unicode processing, since I don't know much about the differences in internal processing between R2 and R3.

If R2 differs from R3 in the same way a V6 engine differs from a V8, I still think a real useful and modern general computing language could be built relatively quickly from the R2 foundation by an open source community. On the other hand, if R2 differs from R3 like a V8 differs from a Wankel, it would be much more problematic.

But I'm troubled by the number of people who seem to feel like they "deserve" immediate and full access to the R3 source. A few of them seem like they might have difficulty programming their way out of a wet paper bag, and I can easily foresee a situation where R3 doesn't just fork into incompatible branches, but rather splinters into a thousand pieces of useless dust.

For that reason, I'm going to suggest that RT look into releasing R3 only to the group of developers who have been working helpfully on R3 so far. If that group of developers feels like they can't handle it alone, there might be a shot at handing off R3 to a university or some other consortium that would be better able to develop it.

And, of course, if R2 is released as open source, the better developers on that project could also be considered as candidates for R3 development further down the road.

But, for everyone's peace of mind, the people who are intent on pushing for immediate solutions to complex problems should really consider sitting at home and typing one-liner 'tick tock' scripts to themselves. Not everything in life is as free and easy as we'd like.

-pekr-
15-Sep-2012 12:30:51
Fork - you are an example of person, who wants to lead REBOL somewhere else than probably many in the community imo. You exagerrate things a bit, no? Who cares about /only? Is that the end of the world? Ditto REBOL vs Rebol. Ditto Tiobe - new languages pop-up here or there, and the last thing authors care when thinking about creating one is to be in top 10 on Tiobe. C/C++ comments - why? And GPL? I know where you lead - but - that's hardly a freedom. I, and I am sure MANY other ppl will not ever touch any clone, with something like GPL licence, easy as that, and I think BrinH provides some very reasonable reasons.

So - in the meantime of our chatter and the chance of Carl reappearing in next 18 months or years :-), Red's going to be in late alpha, supported by many in the community, done by person with good REBOL track-record, and that's all we need ...

And please - a bit more respect to other guy's efforts - re your Kaj's reference ...

Fork
15-Sep-2012 13:03:01
I'm going to hold my tongue on a lot of things I could say that would not be productive.

But yes: REBOL vs. Rebol makes a difference. I will offer that if a change is so "minor" and "negligible", why not just make it?

The lack of the ability of anyone to remix, adapt, and present the concepts of Rebol properly--in a frictionless way--has crippled it into absolute obscurity. Meanwhile, languages with no merit whatsoever have succeeded due to catchy names, social butterfly evangelists, fancy websites, easy deployment.

If you like that situation...because you you feel you're part of a secret club that knows enough to jump the hoops and see past the website that looks like it's from the days of NCSA Mosaic, then you are 100% absolutely correct: I am an example of a person who wants to lead Rebol somewhere else entirely. I don't believe in hazing.

Rebol claims to be easy, but you can't sudo apt-get rebol...you have to wget http://www.rebol.com/r3/downloads/r3-a111-4-4.tar.gz and then you have to gunzip r3-a111-4-4.tar.gz and then you have to tar -xf r3-a111-4-4.tar and then you have to chmod +x r3-a111-4-4 and by this time no one cares to figure out how to put it in the install path or what to call it.

No: I don't represent the community, because most anyone who thinks like I do has left for good to be part of the greater ecology. (Douglas Crockford, for instance...or my friend today whose response to my question of "what will happen to Rebol" was "I wish I cared".) We haven't been given the chance to try our ideas. On that note, (at) KSH, it's not an issue of feeling one has a right to the source, but rather what serves the best interests of a world spiraling into complexity.

This is bigger than us.

MaxV
15-Sep-2012 17:28:55
Linux deb package with all dependencies:

http://www.maxvessi.net/rebsite/Linux/rebol-2.7.8.4.3-4.i386.deb

All other Linux packages on:

http://www.maxvessi.net/rebsite/Linux/

Fork
16-Sep-2012 17:34:11
(at) MaxV...that's cool. I'd actually never noticed that permission was granted in the license to redistribute the executables:

http://www.rebol.com/docs/words/wlicense.html

Nevertheless: that license won't fly with the mainline Debian packagers. But a GPL would open up ground on issues like that. The copyright owner can release software under as many licenses as they want, and I'll stand by saying it's a conservative first step that keeps a lot of the ball in Carl's court. He can ask for payback from secretive commercial applications, while showing goodwill to academia and idealists.

What BrianH says about there "not being time to exercise control" rings a bit true...sort of. Admittedly I can't point to a heck of a lot of successful language cores under the GPL license, outside of GCC. And Clang is aggressively gaining ground on GCC, where the MIT/BSD-style license made it favorable to Apple to join in and use it for XCode. (It's a cleaner codebase, too.)

Guess I just think it's okay if the industrialists have to wait a little bit for the freedom to mix in their private IP without sharing the DNA. It's been a quiet two years, and I think there's a few non-profit-motivated people who are ready to jump in and help make some noise.

Brian Hawley
16-Sep-2012 18:39:18
GCC has a different execution model than Rebol: It's a compiler, not an interpreter, and doesn't include a GPL'd runtime in the applications it generates. It goes out of its way to make proprietary use possible, for that matter.

MIT and BSD licenses meet the Debian guidelines too. You don't have to stick us with a copyleft license to get Rebol into distributions. A permissive license would do.

Steb
16-Sep-2012 18:41:53
So we are at the crossroads. Where can Carl go?

Looking at models for running a language some which come to mind are: 1. BDFL: benign dictator for life (eg Guido) 2. core group (eg the Lua triad). 3. a specification by a core group who vary eg RnRS in scheme (who have kept it recognizably close to the founding ideals over 30 odd years), leaving implementation alone. 4. get the core ideas together into something coherent and then release into the wild and hope it emerges from the primordial slime in a recognizable fashion. Lisp perhaps? 5. get enough people using it in (mostly) one fashion then rely on critical mass keeping it pretty much the same when it is released into the wild eg C

5. is irrelevant to us given that rebol is a minority preference. GPL or MIT is version 4: from comments here not uniformly popular, and certainly risky. Model 2 seems to be the closest next step to 1. if there is no obvious 'next' BDFL.

Whichever is chosen is to my mind not so important as doing it with some swiftness so that the community has something to marshal behind. Ongoing delay is only going to persuade people that there's nothing to see here folks, just move on.

Fork
16-Sep-2012 19:15:01
(at) BrianH - Carl seems to have decided to embrace a more open process and is only questioning which one. I'm only talking about a first step that opens doors to almost any non-commercial application, while leaving the option for Rebol getting a profit model if commercial interests want to negotiate. A GPL-licensed release seems like a low-risk step that doesn't give away everything all at once.

Plus, it's not irreversible if it doesn't work. It would make at least one person happy (me)--as I have no commercial interests whatsoever. As far as I'm concerned, "money" as we know it is totally utterly fake.

As for the mixture of interpreters and GPL, it's admittedly tricky, but covered in the FAQ:

http://www.gnu.org/licenses/gpl-faq.html#IfInterpreterIsGPL

A bitcoin sceptic
17-Sep-2012 3:30:22
Yes Fork you are right money is fake. Certainly since the golden standard was given up. But it has a community value that the bitcoin lacks. I can use it to pay rent or buy a car or to travel over the community-built road to the supermarket to buy food for the family. Bitcoins have no value, though I know some who would accept them, they were digitally digged up by computing things (what?) giving the issuee at least a head start collecting them. I trust these bitcoin instances less than the government.

Next point, I just clicked the right downloads from a webpage so no sudo apt-get sh*t at all. It was easy and it worked right out of the box on all platforms I tried it on.

Then, please leave the C/C++ style out of REBOL or make your own dialect with as many foos and bars and () and {} and //comments as you like. And don't forget requiring a ; or a dot at each end of a line.

NEXT!

;-)

Maxim Olivier-Adlhoch
17-Sep-2012 5:18:57
"Carl made the best argument why not to use closed source in the past two years, and Fork just made the best argument for Carl not to open source..."

hahahaha

Fork, though I agree with some of the issues you raise, I find an irony in trying to make Rebol more like what it aims to change.

Thing is, trying to change the syntax like you proclaim, will not make Rebol better.

Just take this one: switching ; for // has absolutely NO value. Its a personal preference, one which obviously stems from your previous knowledge of C/C++.

C's syntax is just plain terrible... but we learned it, got used to it and now we (I) even find it "normal" so much so that every other "language" that comes out, just replicates it completely. It used to be that languages actually had different spellings. I never understand people talking about a "new language" when its just another flavor of C.

When I used to be a teacher for film 3D & VFX it was always funny to see people claim that "their" 3D app was the best. it was 90% of the time the one they learned first. All they would claim is better was simply how they where molded first. is a square peg better than a round one? depends which kit of tools you brought with you.

Rebol's syntax is its own design. there are things I'd change, but then doing so would alienate my code from all the rest out there. there's no point. We'll all change it in different ways.

If I want Rebol to evolve, I'd rather work on the core missing parts of the language. My goal, for example, is to bring dataflow-driven lazy computing within the core.

This and other missing things like multi-threading, devices, new platforms and more are the reasons we need the language open sourced... not to add new cosmetics and makeup to it.

Fork
17-Sep-2012 11:24:15
(I'm going to again hold my tongue in saying things that are not productive.)

If comments were the only thing I didn't care for in Rebol's choices, then no. I would not want to touch them for the sake of coming into more alignment with other languages that are popular. I merely give it as an example of a change that could be tried along with other changes to see if it could broaden the appeal. I'll again use the term "speedbump" to describe every step in a new user's learning/perception:

the website - speedbump
the install - speedbump

append/only [a b c] [d e] ; speedbump
func, funct, function ; speedbump, especially since the only one with a decent name is rarely used
append [x y z] q ; a speedbump so bad the interpreter looks like it crashed out

I could make lists all day. Yet I've seen this repeated reactionary behavior to desires to fix them. It's coming from people who'd rather have hundreds of languages which are more popular than Rebol...than imagine giving anyone a chance to try doing the impedance match.

There's no reason why someone couldn't go into various languages and translate the entire Rebol keyword set into the native tongue...even Chinese. Drop it into schools and let kids learn it on their own terms. Rebol is one of few languages that has this utter flexibility to do that; but the idea gets rejected on the basis of your "alienation".

Y'know what's even more incompatible with Rebol than a Rebol with different comments and keywords? Ruby. Python. Any of the 100 languages in the Tiobe index. An official branch of Rebol(tm) to rally around is a great thing, but it shouldn't be the only thing allowed in the ecosystem.

Kaj
17-Sep-2012 14:07:52
Boron sounds perfect for you. Why haven't you forked that?
Carl Sassenrath
17-Sep-2012 20:41:08
It's really good to hear from everyone again. Thanks for taking the time to post your opinions and suggestions.

I've read your comments and made a decision. It's time to move forward. I think most of you will be happy with the decision.

I'll post a new blog to describe the next steps, and I'm going to need your help. We should be able to resume REBOL development again quite soon... but of course, with an entirely new approach to how it's been done with REBOL before.

Thanks again.

Maxim Olivier-Adlhoch
17-Sep-2012 22:18:11
Happy to hear from you Carl and eager to read your next post.
claude
17-Sep-2012 23:44:45
thank you to move forward with REBOL
I'm a nobody
18-Sep-2012 0:06:14
With that in mind please ignore my ignorance. I first learned of REBOL about 6 months ago. I am new to programming and found the current software stack of using PHP, HTML, CSS, JS or Ruby /Ruby on rails with MySQl .... somewhat cumbersome. I was trying to figure the best platform to accomplish a new app would be. In my search I found another app who team had chosen REBOL for their platform. The other similar is as you know the Ruby on Rails came from. My point being is what good is REBOL without a community? What better way to create a community than by showing what can be done and getting people interested. If a REBOL framework/library could be made to replicate what ruby on rails does or replace the wordpress cms and it's plugins and theme makers, you would attract a huge new community of users for REBOL that would then bolster the support of it's more advance usage.

Then again what do I know.

Ken Singleton
18-Sep-2012 1:54:50
Carl, thank you for not letting REBOL die. REBOL is like a beautiful wine nestling among many variations of vinegar.
Arnold
18-Sep-2012 3:20:20
I worked my way through a Ruby on Rails book once, not really long ago before I knew of REBOL so I am a newbie as well, building a "webpage" using the rails stuff. All kinds of impossible steps to follow using some mvc model. I can only say I was utterly disappointed with the end result that was just a table of retrieved data from the mySQL table that I could have produced in a way that is much more logical, understandible and maintainable. I never even looked into Rails and Ruby itself again.

But anyway back to the point that matters. Glad to read here that 'our' real Carl has come to a decision and I am very interested in the next blogpost. Even if it cannot be the best solution to everyone, I think everybody here can agree that a clear new situation is to be preferred above uncertainty. Now just hope for the best for all!

Luke
18-Sep-2012 5:45:27
Thanks Carl for taking a decision to move REBOL forward. I hope it will now be an open community process. There are many smart people who can contribute. I'm trying not to hold my breath, but I'm hopeful nonetheless.......
Arnold
18-Sep-2012 6:30:25
I bet it is going to be a competition: R2 open sourced and R3 the same as today. Carl gives a generous headstart to the community to debug and improve R2 and he will advance R3. (Which he secretely already finished when posting this blogpost.)

Could we speak to Mrs Sassenrath for moment? ;-)

Brian Wisti
18-Sep-2012 13:39:09
Sounds like a decision has already been made and will eventually be described, so I'm just going to ramble a moment.

I'm beside myself with joy that something is happening in the REBOL world. I know Carl is busy with many things, and wouldn't take that away from him. But I can't see his other projects or drink his wine, so REBOL is what I focus on.

I cringe a little every time I hear a Ruby dev use the term "DSL." I sigh whenever I have to download another hundred (or more) megabyte package to play with some new language on another supposedly universal virtual machine. REBOL has always been my favorite universal platform for writing DSLs.

Love-fest aside, please don't go GPL. I appreciate the purity of the license, but I would not be able to use it at work. This would make me sad.

Okay I'll hush up and wait for whatever Carl's got to say. Maybe I'll go look at the Red repo or something. Or work. Yeah, that's a better idea.

KSH
18-Sep-2012 15:19:20
Really glad to hear this! Fork probably felt a bit picked on lately, but there's alot of logic in his last statement: "An official branch of Rebol(tm) to rally around is a great thing, but it shouldn't be the only thing allowed in the ecosystem."

I'm hoping the final decision will allow people to branch off in their own directions if needed, while retaining a strong "trunk" under the guidance of a smart group of developers.

Unfortunately, I'm not one of that smart group--being more of a hobby-programmer at this point--but please let us know how the less technical can contribute (evangelism, documentation, etc.).

I guess the way to measure success will be to compare the quality of what future developers produce compared to what Carl & RT have already done. To my mind, the bar is set extremely high, but I'm looking forward to seeing what the community can do.

Long live the REBOLution!

Steven White
18-Sep-2012 17:43:32
Regarding the GPL issue, this note from the Free Software Foundation seems to mean that one can write a non-free program with a GPL tool, even if the final product is "encapped" the way the REBOL SDK does.

Can I use GPL-covered editors such as GNU Emacs to develop non-free programs? Can I use GPL-covered tools such as GCC to compile them?

Yes, because the copyright on the editors and tools does not cover the code you write. Using them does not place any restrictions, legally, on the license you use for your code.

Some programs copy parts of themselves into the output for technical reasons—for example, Bison copies a standard parser program into its output file. In such cases, the copied text in the output is covered by the same license that covers it in the source code. Meanwhile, the part of the output which is derived from the program's input inherits the copyright status of the input.

As it happens, Bison can also be used to develop non-free programs. This is because we decided to explicitly permit the use of the Bison standard parser program in Bison output files without restriction. We made the decision because there were other tools comparable to Bison which already permitted use for non-free programs.

Brian Wisti
18-Sep-2012 17:52:44
Steven White, my limitation is based more on a twitchy legal department at my employer than the GPL's actual restrictions.
florin
18-Sep-2012 18:09:18
Make it September 26th. It's my birthday, and I'll forgive your winemaking escapades.
Brian Hawley
19-Sep-2012 8:33:27
Steven White, that note from the FSF means that a GPL'd tool that generates other programs that include some stock code or data, can be used to generate non-Free programs if the part for inclusion is licensed in such a way that it can be linked with closed code, basically Classpath or more permissive. In the case of Bison, the part meant for inclusion is a small part of Bison, and that part is licensed with a linking exception similar to Classpath.

In the case of REBOL encapping, the part that is included in the encapped program is basically almost all of REBOL. That means that almost all of REBOL would need at least a Classpath-style linking exception.

It works the other way as well. Running a REBOL script on a closed-source REBOL without encapping is legally similar to running a Java or .NET program on their proprietary runtimes. If you GPL the script, it has to be GPL 3 or greater for this to be legally permitted, since that is when the clause allowing that was added. Not GPL 2 or earlier though - this is why the Classpath license was invented.

Arnold
20-Sep-2012 5:34:36
Still waiting for the new blogpost. Because it will be number 0511, we all may have to wait until 5th november or worse yet the 11th of may. (Note I do not mention a year.) :( Bored by the false positives by my script, another spampost. I thought there was a way to report these to the site-manager, but no.
-pekr-
20-Sep-2012 6:08:17
Arnold - there's a way to review comments - just go back to the main Carl's blog page, and in the header's bottom right corner, you might see links to review last week or last month comments.

But - be warned - I did so cca two months ago, and at that time, there was no reaction and spam was still present, which means - none in RT to handle the request, or something like that ...

PS: I like you dark humour about possible next blog article post date :-)

Arnold
20-Sep-2012 7:20:11
Dark humour brightens up my life ;-)

BTW. I reported all spams of last week, many topics should be closed for comments.

Paul T - The prophet of
20-Sep-2012 21:39:50
Carl, I can't perceive what REBOL can now bring to the table. You fell victim to accomplishment and creative genius. Genius is hindered by its hightened awareness of capability. That awareness leads to a comfort in accomplishment and potential that leads one to rest on the merit of their efforts. The true leader perceives this and overcomes. Overcome Carl...

Fork
21-Sep-2012 5:26:40
(at) KSH thank you for the empathy, I do feel picked on, but I expect it. Feel free to write me anytime.

http://hostilefork.com

(at) Carl, glad you've come to a decision...but I hope it does not involve open-sourcing R2. I know that might seem contradictory given the destructive anarchist label some here want to stick me with...but I want a strong trunk, with Unicode and all the other progress at the foundation that you and BrianH and everyone else spent years-upon-years building.

(at) Kaj, This is why I haven't taken an interest in Boron...the license is noble, but the codebase lacks Unicode, protect, the Mezzanine, and all the awesome twiddly bits that must be under the hood to make our experience of the product have the magic it (sometimes) has.

Kaj
21-Sep-2012 5:54:45
You should have a better look before rejecting the alternative you so crave. Boron has Unicode, and I don't know what you mean by "the Mezzanine", but Boron has mezzanine functions just like REBOL. It even has the Qt GUI that you say you want on your website. Surely if you were serious about starting your own REBOL clone project, you should be able to just add those twiddly bits to Boron. In any case, much the same stuff that's missing from Boron is also missing from R3, if and when you could use that.
MaxV
21-Sep-2012 9:10:10
REBOL/Open License 1.0

This software is Copyright REBOL Technologies. All rights reserved. REBOL is a registered trademark of REBOL Technologies.

Redistribution and use of this software, in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  1. Redistributions in source code must retain the above copyright notice, this list of conditions, and the following disclaimer.
  2. Redistributions in binary form must display the above copyright notice, and the documentation and/or other materials provided with the distribution must also reproduce the above copyright, this list of conditions, and the following disclaimer.
  3. The right to distribute this software or to use it for any purpose does not give you the right to use Trademarks of REBOL Technologies.
  4. The original REBOL Technologies sources and binaries can be download from www.rebol.com

Disclaimer...

Maxim Olivier-Adlhoch
21-Sep-2012 10:11:55
The SDK license could simply allow you to re-distribute without the requirement to reproduce the copyright within your own releases.
Cheap Designer Watches
23-Sep-2012 5:18:36
You assume like a tight-knit group,www.topreplicawatchesus.com
W^L+
23-Sep-2012 20:53:25
Carl, it is good to see that you're still around and still interested in Rebol's future. As far as licenses go, I'd like to suggest the Mozilla Public License, which seems to pass corporate scrutiny where stronger user-empowering licenses like the GPL do not. I also feel that an explicit "your code is your own, license as you will" statement would go a long way to soothe the fears of those who are anti-copyleft.

Like many of the posters above, I believe that a core team should have access to a main (Github/Bitbucket/Launchpad/whatever) repo for the main (canonical) implementation, and that a language reference and conformance standard is needed. I think the Ruby way is a mistake (conformance is "imitates MRI's behavior") in a project that needs multiple forks and independent implementations.

Also, the spam-filter needs tightening and I'd suggest rejecting the HTML "a" tag to reduce spam.

Tony Casanova
30-Nov-2012 20:16:35
Open Source it. Get it running on Raspberry_Pi. http://en.wikipedia.org/wiki/Raspberry_Pi Thank you.
Drakopoulos Anastasios
20-Feb-2013 22:39:52
Rebol is the future today! Please, don't give up!

Best Regards!

Giancarlo
10-Mar-2013 10:19:58
I used to work in the computer industry as a hardware tech until retirment. I did some programming in the times of machine language and assembler but nothing since. I used to love computers and think that they would change the world. They did but at what cost! No privacy left! The program controls the computer and the computer controls you. Whoever controls the program can do what he wants. The end user must write his own program on open source Os. For curiosity I checked some of the current languages perl python ruby tcl/tk to see if it could be possible. But I , a non programmer, could not do it. Then i came across rebol and it is a dream. You can do many things if you learn the functions. You created a magical language completely different from anything else. The language deserves to be widespread to help free the users. Will it be possible? I do not know but if you give up your language will certainly die. It is still too small to take off. It would be a terrible waste. Do not give up, it is too important for the world even if it produces no income. If you manage to find a sponsor of some sort to make it known to the pubblic it may do the trick.
Ade
9-Aug-2013 19:44:58
Put a Bitcoin address on Rebol for donations, may not happen now because not that many have Bitcoins, but in the future...
Thea
19-Sep-2016 1:01:29
Hello, your article is very interesting as I read it. Please do visit my site too if you have time. Thank you and God Bless!.

gofastek.com www.gofastek.com

SP
29-May-2017 16:30:56
I see that a link to the blog no longer appears on the front page.

There is also a mention on StackOverflow that Carl is working on R4 or next gen. Curious to know if indeed something else is coming.

Post a Comment:

You can post a comment here. Keep it on-topic.

Name:

Blog id:

CS-0510


Comment:


 Note: HTML tags allowed for: b i u li ol ul font p br pre tt blockquote
 
 

This is a technical blog related to the above topic. We reserve the right to remove comments that are off-topic, irrelevant links, advertisements, spams, personal attacks, politics, religion, etc.

Updated 16-Oct-2017   -   Copyright Carl Sassenrath   -   WWW.REBOL.COM   -   Edit   -   Blogger Source Code