Chris Velevitch wrote:
> On 8/30/06, Aral Balkan <[EMAIL PROTECTED]> wrote:
>> I think we need to identify the main issue here: Is it lack of funds
>> or lack of expertise or a lack of need?
> 
> Wasn't the issue lack time time by the original developer to continue
> developing ASDT? And didn't he ask for volunteers to help out or take
> over the project?

well theres a few of us signed up as ASDT developers but I can only speak for 
myself.

My main constraint was time and also knowledge of the eclipse platform. Now 
i've 
just completed a project built on the Eclipse RCP i have a reasonable amount of 
knowledge, but still little time.

I think Hank really got to the critical point about taking ASDT / HxDT further 
and that is they both require a solid core to build other features on top of.
Which means you need a decent parser, preferably one that can recover from 
errors, but also to generate a good model of the source code i.e. a DOM.

Once you have that in place you can build a lot of features that people expect 
from a modern IDE and that make using something like the JDT so productive. 
Syntax highlighting, auto complete and rare but incredibly useful tools like 
refactoring support. Although that now has language neutral support in Eclipse 
(http://www.eclipse.org/articles/Article-LTK/ltk.html) you still need a DOM to 
manipulate.

So as I see it the initial problem is not a lack of Java / Eclipse Platform 
knowledge but something much more scarce, people who understand and can build 
parsers and compilers.

Unfortunately I was mainly asleep or missing for my university compiler course 
so im not a great deal of use in that area.

Maybe its worth creating a project specifically for dealing with creating a 
parser for as2 / as3 / haxe which can interoperate with which ever editor 
platform be it based on eclipse, scintilla or a custom solution.
Or at least creating the specification for how this piece would have to work to 
allow such an integration. Then when Nicolas has a free weekend he can do some 
of his ocaml magic and abracadabra, its done ;)

seriously, i think once its more clear what is required then there is something 
concrete for people to take on. If it requires money as an incentive / 
compensation then im willing to pay into a fund, but it has to have clear 
goals. 
Perhaps like a bug bounty system such as 
http://www.mozilla.org/security/bug-bounty.html.

thats my 2 MKD

martin.

thats my thoughts on the matter..

_______________________________________________
osflash mailing list
[email protected]
http://osflash.org/mailman/listinfo/osflash_osflash.org

Reply via email to