On Wednesday, March 5, 2003, at 09:06 PM, Daniel Vainsencher wrote:

[about one quick fixes release]

Doug said:

...

That does make some sense. A "very good" release might be a good start before getting down to splitting up the image and other work, as opposed to a more average release. A reference release of sorts. And good PR, I suppose.

It would delay the "getting down to splitting up the image" by at least a month, though.

People are posting "Remove <whatever>" packages, though slowly. I think I can help people to speed up and focus this process independent of the release schedule, as long as the fixes made do not break refactorings by including too many things. So as long as the fixes release is focused on the important stuff, I see no conflict.

When we start 3.6, we'll have a few more removals ready to go.

Yes, that's true... that work could continue during a short 3.5 bugfix release.

...

And I'm not sure it's *really* that important. So I'm still on the fence, awaiting a few more opinions.

Yup, I'd like to hear more too, but keeping in mind that -

...

But we do need to decide on the 3.5 release timing very soon...

Daniel

Well, I haven't seen any more opinions coming in against the idea of a short bugfix release for 3.5 (although not many in favor have come in either, aside from a few on squeak-dev).

After thinking about it for awhile, I'm inclined to go along with the idea of a short 3.5 bugfix release. We'd need to set the release date for about 1 month from now, or perhaps 1 month from when we begin harvesting fixes (which from the looks of it should be very soon). 2 weeks alpha, 1 week beta, 1 week gamma, or something like that.

I'm assuming that future releases (3.6 and beyond) will have a longer, normal release cycle, which would allow enough time for enhancements, refactorings, and package removals, in addition to bugfixes. Somewhere between 3 and 6 months per release is what has been tossed around, but we don't have to decide that just yet.

If we do go ahead with this short plan for 3.5, we will need to be strict about letting in only bugfixes, and even then trying to avoid fixes which are overly invasive or widespread. As an example, the Morphic Cleanup Project stuff seems to be mostly refactorings as opposed to bugfixes. Very useful refactorings, of course. :-) But as such, I would think most of that sort of stuff would wait until 3.6alpha to be included. (Which isn't really *that* far away anyway, and an organized project like MCP would probably warrant consideration for inclusion right at the beginning of 3.6alpha.)

IMHO, possibly the only non-bugfix item worth considering for 3.5 would be removing the Apple fonts (assuming we decided to go with the simple Accufont solution at first since that's pretty well tested), if we were worried about delaying progress with Apple on the licensing issues. But even that might be a stretch.

Anyway, my two cents.

- Doug Way