Hi,
  Just got here from...  
http://www.reddit.com/r/Python/comments/yp6ib/has_anyone_here_used_python_pipelines/

I wonder if it would be worth moving the development to github?


I think this would give - better visibility, easier participation (people 
can easily open tickets/pull requests and fork if they need to).

I remember checking out Kamaelia a while ago and it certainly is promising, 
it just doesn't seem to have the visibility it deserves.

If it's going into maintenance mode, it might be better somewhere like 
github where I reckon theres a better chance that someone else might pick 
it up.


(Although the newer/more exciting; certainly is exciting ...  the same 
things apply here).

Cheers,
Stuart


On Tuesday, July 10, 2012 1:00:20 PM UTC+1, Michael wrote:
>
> Hi, 
>
>
> Some thoughts, feedback welcome. 
>
> As people may've noticed, Kamaelia itself has had no major features or 
> extensions in some time.  There have been extensions to applications 
> using Kamaelia, and this largely reflects the project's status - stable, 
> and in use in a small number of projects.  (Indeed, the SVN history 
> reflects this pretty well IMO :-) 
>
> When it was started, there wasn't really much like it - in terms of 
> trying to make concurrency easier to use/applicable. It's been nice / 
> interesting to see a number of other projects take the ideas forward. 
> (Which was in part the point really) 
>
> As a result, given recent questions, my suggestion is as follows: 
>
>  - Kamaelia goes into a long term maintenance mode (not dead, just 
> resting) 
>
>  - This being realsitic needs to be in versions that are 2.7 and 3.3 
>    compatible. 
>
>  - The website is changed to reflect to represent this change in state. 
>    Hopefully simplified as well. 
>
>  - When (or if) a follow on project starts up, the website will change 
>    to reflect that, and will be able to use the 2.7/3.3 codebase in a 
>    stable known fashion.  (Probably via a code bridge) 
>
>  - Also, the copyright statement will change slightly to reflect this. 
>    I'll also ditch the need for new contributors to sign a contributor 
>    agreement, because if the project is not in ongoing development, it 
>    needs to default to use clause 5 of the Apache License instead.  (Also 
>    clause 5 of the Apache License is perfect for this) They'll still be 
>    welcome though, and effectively still act to protect contributors. 
>
>  - Open question here is this - switch to github? I've been using git at 
>    work a lot recently and I can see how lots of things are easier using 
>    it. 
>
> I'm not disappearing, but I think for things to move on, the project 
> needs to explicitly say "this code is now end of line and something new 
> will replace it, but don't worry, it *is* maintained, looked after and 
> used". 
>
> After all, for reference, in addition to Jim's projects, I still use 
> Kamaelia for a number of different things, so I still need it to work 
> and stay relevant.  As a result, bug fixes for different versions of 
> Ubuntu/etc make sense. 
>
> However, I think stamping this version of Kamaelia as "the" version for 
> Ubuntu 12.04 LTS is probably the way forward.  When 12.04 is end of 
> life, I think the same will hold for Kamaelia. 
>
> For those that worries, Ubuntu 12.04's EOL is 2 years after it ceases 
> being "the" LTS release, which puts that at 7 years from now, with bug 
> fixes relative to 12.04 for the next 5. 
>
> If in 7 years time I'm still using it, I'll be surprised. The first 
> lines in the codebase thought are now 10 years old, which is older than 
> I expected and older than it has any right to be, and it's still useful, 
> so that's pretty neat. This code is probably the oldest for reference: 
>   - 
> https://kamaelia.googlecode.com/svn/trunk/Sketches/MPS/Old/microProcess.py 
>
> Hence the suggestion though of slipping to a dependable maintenance mode, 
> to make room for something newer and more exciting :-) 
>
> I think though the path to getting that to that stage is probably to 
> make the codebase python 3.X compatible first. 
>
> I think in that process debianising the packaging as well makes sense. 
> A bit of effort happened there in the past, but I've recently done this 
> for some other things, and doing the same to Kamaelia would make sense 
> if there's interest. 
>
> It's also probably worth mentioning that all work on kamaelia itself is 
> something I do on my own time these days and have for a few years now, 
> so any assistance is welcome. 
>
> Thoughts? 
>
>
> Michael. 
>

-- 
You received this message because you are subscribed to the Google Groups 
"kamaelia" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/kamaelia/-/AtY6jqED10sJ.
To post to this group, send email to kamaelia@googlegroups.com.
To unsubscribe from this group, send email to 
kamaelia+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/kamaelia?hl=en.

Reply via email to