Team proposal: Morphic Splitters
Dan at SqueakLand.org
Wed Feb 23 16:57:17 UTC 2005
>Anyway, here's a team proposal. I'm Team Leader Ad Interim - if someone else wants the job, there's a good chance I'll step down because I'll be busy enough as it is.
I support you, and I think you deserve at least one fun project.
>Goal: split up Morphic into 'Mini Morphic' and 'Rest Of Morphic'
This seems do-able.
I have a number of thoughts...
1. One simple first step would be to strip an image down to just the basic tools, run the iterative part of major shrink, and then see what is left of Morphic.
2. I would like to do the same (perhaps they have already done it) for Pavel and Steven's "Small-3.7" and QuiteSmall images. This is excellent work, very much in the spirit of this project. I hope they will sign on, at least in an advisory capacity.
3. Similarly I'd like to see what Juan has already done. It may turn out that there is very little to do.
4. I would also like to get from John Maloney one of his early Morphic tutorial images, plus hear any suggestions from him about this project (note the CC).
5. If we augment the discard process in major shrink (or whatever script Pavel, Steven and Juan may have) so that it does a fileOut of everything it discards, then building the FatMorphic package should not take more than a day or two.
6. The only serious remaining part of this project (wait a minute, *is* this part of the project? ;-) is to make EToys into a package. Here I think we need some discussion about where we draw the line as to what remains in Basic, and what goes into the EToy package. There are a number of novice-oriented facilities that are not really EToy facilities, that may or may not be deemed to be an appropriate inclusion in the basic system. Here again, I think the QuiteSmall work may save us some time.
7. We will need some tests to tell us when the EToy package is working fully. Important here, I think, is to have a *person* or several who is capable of doing this testing, and able to fix simple problems until it works.
8. Lastly, it would be great to get someone on board who is interested in the Abstract UI project. This minimal set of tools and minimal morphic should be a nice simple context in which to convert things over to an abstract interface. All of the ugly Morphic/MVC compatibility methods are there just waiting to be cleaned up, and I think it would be a lot of fun to do. With one or two motivated people on it, i think it would be quite reasonable to see an AUI layer and MVC as a separate package in the same June/July time frame. And wouldn't that be cool!
More information about the Squeak-dev