Andreas:

| there are several open issues in the Zope collector with | Zope running on Windows. I can't see anyone worrying | about the Windows platform (I also don't care about | Windows as UN*X guy). So if you are interested | in running Zope on Windows in a reasonable way than *you* | should help to fix the outstanding bugs.

A quick search for "windows" doesn't show too many. As far as I can tell, all but 2 (1689 and 1728) are fixed with my patches. I also note that my patches have been sitting there for a number of months now - so I humbly suggest the problem is not simply a lack of appropriate fixes :)

| I heard some remarks (behind my back) | that Zope is not a stable solution on Windows...possibly.....

In my experience, that is not true. Zope on Windows is generally as stable as Zope on any platform. Zope on Windows does suffer a form of neglect, but that doesn't make it "unstable" so much as "unsuitable" - things like service error handling, lack of log rotation, clean shutdown etc all mean Zope lacks features necessary for serious consideration on Windows, but is not "unstable". So next time you hear those second-hand comments, you can confidently dismiss them as FUD :) [Tres]

I told Alan Runyan at the sprint in Paris that we would be glad to have Enfold's "Windows stability" patches folded back into upstream. He was glad to hear it; I hope to see something from Andy / Alan / Sidnei soon.

I'd really like to see our "win32all" story cleaned up, per Mark Hammond's issues, too.

In case it isn't well known, I am working for Enfold. Thus, patches from Andy, Alan or Sid are all going to be mine anyway :) All except logfile rotation have been submitted as patches to the collector, and the motivation to submit that is low while the other patches remain outstanding. By way of background, the story behind these patches languishing is: * I enter the scene with no clue about Zope or the Zope culture. I slowly discover Enfold are mainly interested in Plone which is currently stuck at Zope 2.7.x. I've still so much to understand :) * I realize Zope 2.7.x is clearly a "maintenance" branch (from Zope's POV), but also a "trunk" branch from the POV of Plone (and therefore from mine <wink>) * All Zope versions are clearly "broken" (as above) in fundamental (but generally the same) ways on Windows, but the status of my patches as "features" vs "bugfixes" is unclear - especially given my lack of understanding of the culture or the Zope world beyond Plone. * Being part of Enfold, I try and punt all these issues off to poor Sid, who also doesn't care for Windows, so doesn't really know how to handle the issue either. * Enfold continue to release a "frankenstein" version of Zope 2.7.x on Windows :) Occasionally zope-dev sees a bit of noise on Windows issues before all again goes quiet... I understand I could probably gain committer privileges but as mentioned, I wouldn't quite know what to do with them if I did <wink> Mark.

[Mark Hammond] ...

* All Zope versions are clearly "broken" (as above) in fundamental (but generally the same) ways on Windows, but the status of my patches as "features" vs "bugfixes" is unclear - especially given my lack of understanding of the culture or the Zope world beyond Plone.

They're clearly bugfixes to me; that shouldn't be a hangup here. What's missing is a Zope committer who runs on Windows volunteering time to commit the patches in 2.7-land, and to port them to 2.8- and 3.0-land (those require more than just mechanical merge, because things like moving to Python's standard logging module require rewriting code). Alas, Zope in Windowsland attracts about as much volunteer work apart from Mark Hammond as, say, Python 1.5.2 in Windowsland did. Mark understands what that means. Everyone else can read it as "except for building the installer, hard to distinguish from none" <0.7 wink>.

Mark Hammond wrote:

A quick search for "windows" doesn't show too many. As far as I can tell, all but 2 (1689 and 1728) are fixed with my patches. I also note that my patches have been sitting there for a number of months now - so I humbly suggest the problem is not simply a lack of appropriate fixes :)

Yes, but it is a shortage of people knowledgeable enough to merge them ;-)

In my experience, that is not true. Zope on Windows is generally as stable as Zope on any platform.

I'll second that.

* All Zope versions are clearly "broken" (as above) in fundamental (but generally the same) ways on Windows, but the status of my patches as "features" vs "bugfixes" is unclear - especially given my lack of understanding of the culture or the Zope world beyond Plone.

They're bugfixes, I don't think anyone would complain if you fixed 2.7 AND 2.8 ;-)

I understand I could probably gain committer privileges but as mentioned, I wouldn't quite know what to do with them if I did <wink>

Well, you could merge your patches. I suspect you're the only person knowledgeable and confident enough to do so. Many people would thank you for it though :-) cheers, Chris -- Simplistix - Content Management, Zope & Python Consulting - http://www.simplistix.co.uk