Hmm...OBEX
Mag748
Posts: 266
I'm here to "talk about what might be happening" to the OBEX. Anyone know?
Thanks,
Marcus
Thanks,
Marcus
Comments
There's still the old obex... http://obexclassic.parallax.com/
Ken mentioned it a few days ago:
http://forums.parallax.com/showthread.php/147468-About-to-put-first-object-into-OBEX-Can-someone-review-it-first?highlight=OBEX
A lot of work would be lost if they didn't.
I hope this transition can be implemented with smooth forwarding of links from old to new. It makes more sense to me to leave the old one at obex and call the new one newbex or something like that.
Yes, there's so much going on and I've been quiet about dropping hints. When the gag order is lifted I've got more to talk about that you could imagine. OBEX has been redesigned from the ground up in a Drupal framework that integrates with our new web site. This is a long-term effort to make a functional, reliable tool with single-sign in, fast response, and to provide support for more C. In this process we have done many things:
- Moved all objects to the new system, eliminating many along the way.
- Rewriting descriptions, looking closely at the whole object.
- Implemented a new rating system with three levels.
- Developed a new signup process (everybody will get an e-mail from us shortly, written by Jeff Martin).
- Redefined the prior-known Gold Standard, to be required for the top level "core" objects.
- Removed all references to C from Imagecraft or Catalina, and it will now only support PropGCC code libraries.
I'm in Taipei right now with Jessica to train their universities on Prop GCC for education (about 30+ Learn pages you haven't seen with new libraries, some under development) so I'm a tad out of sync and will remain that way until OPC when I get home. But last time I heard, the new OBEX isn't live because of the need to divorce the former system's e-mail from the licensing claws of Microsoft's SQL server and Exchange (we're moving to Google). That's the only hangup since the web address names seem to be properly resolved. I'm sure Jim E is working feverishly to solve this problem.We've had amazing effort into this project over the last 90 days from Bump Jacobs, Jeff Ledger (@Propeller Powered), Jeff Martin, Daniel Harris, Jen Jacobs and Jim Ewald. This group pulled together a great effort and a result far beyond what I expected.
@Tracy A: that's something I can't answer, unfortunately. I think I know the answer (we're probably using new links) because the backwards-compatibility would have thrown us into a long-term coding cycle that we couldn't exit [leaving us with NO new OBEX in the end].
We hope you will enjoy the new OBEX as soon as it becomes visible!
Sincerely,
Ken Gracey
You say "eliminating many along the way". Are you removing objects that don't serve a certain purpose, are just not well written, or aren't actually useful? Does that have to do with the "looking closely at the whole object" thing? Seems like a lot of work to look at all the objects, I'm guessing you are only looking at the Parallax-created ones?
Sorry about the confusion. Actually, we've removed very few, < 1% and we applied specific rules to decide whether or not to eliminate them, such as "Do they appear to be infected with a virus," "Do they conflict with our goals for the Propeller," or "Are they majorly broken and appear abandoned." Nearly all existing objects made it over.
If any are found to be missing, please contact us at obex.support@parallax.com so we can resolve it.
Wow, that isn't good for Catalina folks..
Check the updates: they're working on adding it back.
I think t's reasonable for Parallax to eliminate programs dependent upon dev tools that they do not control or are not open-sourced.
-Phil
Catalina is open source and they are allowing PropBasic which I believe is closed source.
I think RossH and ImageCraft might be a bit miffed at having their compilers referred to as "attempts". They are both complete, quality compilers with a lot of effort behind them.
@Phil
I can understand wanting to remove things that are dependent on "foreign" tools. However this is C source code and C is C regardless of the compiler you might use. I know these things won't compile out of the box under GCC due to library differences but there may be some useful nuggets in the source anyway.
Is Catalina open source? I can't seem to find any source code links on the Sourceforge page. Do you have a link?
Catalina is fully open source apart for the Optimizer - I used to charge for that, but I don't any more, so I will post source for that as well at some point..
The source code for Catalina is (and always has been) available on Sourceforge. If you have downloaded Catalina you have also downloaded the entire source code tree, although it is optional whether you install it or not.
As far as I know, there has never been any Catalina code in the OBEX, so removing C programs from the OBEX may affect ImageCraft and GCC, but does not affect Catalina at all. One of the main reasons for this is that Catalina code tends to be straight ANSI C source code, and this is most likely to be GPL licensed or public domain, not MIT licensed. All OBEX code is required to be MIT licensed, so in general Catalina code can't be put in the OBEX anyway.
Ross.
I think you might want to rephrase that. I casual reader might get the wrong impression and conclude that they cannot use any license they like for their code built with Catalina. Where as the opposite is true (I hope).
Quite right - I meant code downloaded from the "big"OBEX - i.e. the internet - and then compiled with Catalina, not C programs written by users (which can of course be fully MIT licensed if desired).
Ross.