If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.
You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!

MeetingNotes004

Page history last edited by PBworks 16 years, 1 month ago

Squeak Release Meeting 4

Monday, March 24, 2008 at 18:00 UTC

On #squeak IRC channel

Open meeting for the unofficial SqueakReleaseTeam, regarding the planned Squeak 4.0 release

Agenda items

Answers are in italic

Questions for Edgar:

Is FunSqueak meant to be FullSqueak?

No. it is meant to be a distribution, like squeak-dev

Should MinimalMorphic/Ladrillos be part of 4.0, or a separate release effort, possibly 3.11 or 4.1?

Probably. We didn't discuss this much

Should we all work on Relicinsing for 4.0, then all work on bug fixes, or both in parallel?

was not discussed

License questions:

Are all contributions for 3.9 and 3.10 still under Squeak-L?

It is safest to assume that all contributions made after November 2006 (just prior to 3.9 final) are still under Squeak-L. The detailed answer is, the effective date of the license agreement differs per contributor

Do we therefore need to redo KernelImage from 3.8?

This would be a good idea for four reasons:

All contributions are definitely dated after 3.8.1
3.8 is the base of nearly all the other forks
3.8 is a bit smaller and simpler than 3.9 and following, as it does not have traits and Monticello
3.8 would be simpler to build a history database from, as it uses the update stream simply, not in a curious roundabout fashion like 3.9 and 3.10 do

It would be a very bad idea for one reason:

Losing 3 years worth of bug fixes

It is hard to say whether or not we will lose these bug fixes anyway by redoing the core of squeak. This is definitely a big issue to discuss. See Squeak40Base for details

For very small, obvious, Squeak-L licensed methods, is it better to leave them alone, or rewrite them with likely the same content?

Size doesn't matter. Copying does. If they have the same source code as Squeak 1.1, they can be freely replaced with that

How do we license-audit KernelImage?

How did Pavel audit KernelImage the first time?

Out of about 5900 methods:

350 methods are definitely dirty, as their latest contributor (as of 3.10) has not signed the agreement
1000 methods are definitely clean, or might as well be, as the source is the same as it was in Squeak 1.1
The remaining 4500 methods are not determined, and would require archeology work

This audit ignored the dates of the contributor signatures
For details, see the LicenseAuditing page

Andrew is doing something about a complete method history. What is the status?
Is there any new information in Andreas's Squeak History, or is it just a better packaging of what is already on ftp.squeak.org?
Also, Maurice Rabb has been working on a squeak history. He says it will be about done by late April
For more details, see the HistoryDatabase page

Exactly what process will we use to clean the image?

Does this document explain it correctly?
Will we write tests for every unclean method?
Will we write some kind of formal spec, a la STEPs?
Do we need to have a separate person write the test/spec than writes the actual new method?

None of this was discussed at all

Scope of Squeak 4.0

The image should be usable. Should we officially support Installer?
Should we make a build server?
Should we incorperate any bug fixes?

None of this was discussed at all

Misc

What is the mailing list for the release team?

It will be release@lists.squeakfoundation.org as soon as Craig sets it up

Transcript

This meeting was much longer and less organized than the previous two. Six hours from start to finish, with 3.5 hours of actual discussion. It is recorded at ircbrowse.com