Comments on: R3.0 Plan for July 2009
REBOL Technologies

Comments on: R3.0 Plan for July 2009

Carl Sassenrath, CTO
REBOL Technologies
10-Jul-2009 17:29 GMT

Article #0414
Main page || Index || Prior Article [0413] || Next Article [0415] || 1 Comments || Send feedback

During June:

Progress on R3 continued to accelerate during June, thanks in large part to an even greater level of contribution from the REBOL community.

Many significant issues were resolved... "nagging" issues like the relationship between binary and string and how do you compare at different levels between datatypes. The result is a more robust and powerful REBOL. And, if you're standing on the sidelines wondering what's going on, all you need to do is glance at CureCode or the R3 change log to observe the advancements that are being made. Good stuff. My hat is off to everyone who helped make this progress possible.

Last month, for a primary goal, I wrote: "It is essential to get the plugin system running this month." Well, it didn't happen. At the heart of the problem are concerns about decoupling -- how to isolate the REBOL kernel from the plugins. This isn't easy. (In fact, I was reading about Google Chrome recently, and they mention similar issues in their system.)

For example, if next year we decide to add a new internal datatype, we don't want to break all the plugin modules. Such breakage would occur if the plugins use a direct method of recognizing datatypes. The more I think about this, the more I think it is a critical issue.

The way to make progress on this issue is to break it down into smaller chunks, get it out to users, then carefully build back up what we need (and desire). A lot of the code is already in existing releases, so what remains to be done is less of a coding issue, and more of a definition and documentation issue.

Goals for July

Here's the master list:

  • Release just one simple working plugin. Easy enough, right? Nothing fancy. A little bit can go a long way. We will take it from there.
  • Finish up security. Put sandboxes online.
  • Add text encoding and decoding refinements for READ and WRITE functions.
  • Add basic file modes control dialect. (More to be added later.)
  • Split-off system and user-script contexts.
  • Start ramping up again on GUI, because it's one of the next big subsystems to finalize. (I spent a week in March making important revisions that need to be incorporated at some point... so we can get more Rebolers involved again on it.)
  • Continue with bug fixing, as necessary.

[Learn more about REBOL3]

1 Comments

Comments:

-pekr-
10-Jul-2009 13:54:50
having beer and fun with friends, I can't miss on posting, right? :-) As for "decoupling", I posted interesting link to article about MS Gazelle browser project, you might find it interedting, even if from MS :-)

Maybe we need to put more thoughts to mentioned decoder system. I will not comment about read/text, you know my opinion, but - is codec system defined so that it can be refined in compatible manner? I miss functionality for streamed codecs. Well, but then the question is, what the design should be, eg if it is not task for threaded app servers (BeOS)? But - I don't want to have my Blue Ray movie whole in memory, right? Even parse should work upon streamed ports, this is my 10 year old wish ...

Ah, next beer :-)

Post a Comment:

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

Name:

Blog id:

CS-0414


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 19-Nov-2017   -   Copyright Carl Sassenrath   -   WWW.REBOL.COM   -   Edit   -   Blogger Source Code