[Ktechlab-devel] Lots and lots of projects to do, can't afford to spend any more time. =(

[P Zoltan]

On Fri, 29 May 2009 08:14:57 +0200, Alan Grimes <agrimes at speakeasy.net>  
wrote:

> That is not to say I'll be able to actually do any of the things I need
> to do while I hack the hell out of ktechlab. =P
>
> There are tons and tons of fairly closed ended projects that ppl could
> help me with.
>
> A couple of thorns in my side:
>
> Now that the math engine is fairly well refactored, write a unit testing
> suite for libktechmath that will test it out on matrix sizes up to
> 10,000 (solve and check).
>
> (I might have some code on hand that does that, that just needs to be
> re-adapted, I have my own linear algebra implementation but I don't
> think it's quite as fast, it is fully tested though with unit tests...
> Problem is that it uses a completely different interface...)
>
> I'm starting a massive overhaul of how ktechlab handles currents. I'm
> currently moving towards a lazy-mode evaluator for element currents.
>
> One significant problem is the hack used in logicOut to display
> voltages! that's a nasty hack that needs to be backed out and replaced
> with something long term so that logicOut and LogicIn can be broken up
> and refactored.
>
> Also, someone should spend some time getting flowparts to work again. =\
>

  The last one is done (hopefully).

  About the subprojects: we should encourage people to contribute to the  
project, so things can get in place faster. Here is an interesting blog  
post on this topic:
http://ondrejcertik.blogspot.com/2009/05/my-experience-with-running-opensource.html