Andrea Aime wrote:
> Am I seeing dark clouds at the horizon that are not really there? 
> Jody, what is your point of view given the planned future of uDig?
I tried to be clear with my link to the Joel article on the peril of 
putting a live codebase on hold while you do something pretty. Mozilla 
did this; and while I like Firefox they really sacrificed the browser 
market to IE in the process.

If we are to have a new GeoTools 3 we first need some ideas to generate 
excitement; and then we need to round up developers who can be excited; 
and make sure we have a nice sensible plan.

Now the plan can be aggressive; but we will need more bodies to make an 
aggressive plan work. Personally I am going to have a lot of energy 
coming out of the OSGeo process; and we will be in a good position to 
recruit.

So if you want an rough outline:
- July - graduate
- August-September recruit and train up new blood
- Tag team module maintainer as mentor / experience with new recruit
- step through the modules one at a time issuing a 3.0.0-Alpha1, 
3.0.0-Alpha2 etc.. for as each module is locked down and the deprecated 
code removed. These alpha releases can be used by projects making the 
migration.
- I am tempted to formally isolate the code into interfaces and 
implementations; ie an interface is either in geoapi or gt-api or it is 
not worth the effort and should be simplified into an abstract class.

My worries on this one are frankly that Geomatys will be move on to 
their own work once Martin's modules are updated (I cannot think of a 
solid business case - other than the fact we need the help). As you 
point out Refractions and OPP are all booked up with their own products 
timelines.

So I agree Andrea; we need a solid plan and resources to do this one. I 
would rather do it with recruiting - getting read of old code dead code 
and half started ideas is something a new recruit will a) enjoy and b) 
have an easier time accomplishing than some of us that were around to 
build it :-) The idea would be to have two module maintainers by the end 
of it.

Now this idea is not perfect - Andrea for some of the orphaned modules 
it is likely that you and I have sufficient experience to offer 
guidance; but other than that we may be forced to let some code die 
(scale back the library to current staffing levels).

I would also like to know Justin's thoughts; he has been able to do some 
amazing things to the code base as a whole; I feel we let him down on 
the testing side a few times but there you go.

The alternative (ie BIG BANG simplfication approach) is also a 
possibility and may be accomplished in less time; but I don't think the 
risk and lack of migration plan would be worth it. Sure is tempting.


Jody

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to