Hi Christian: > >Hi Eric, *, > >On Mon, Jun 30, 2008 at 2:01 PM, eric b <[EMAIL PROTECTED]> wrote: >> Le 30 juin 08 à 13:08, Christian Lohmaier a écrit : >> [...] >> No, that's not my idea, but just a fact: people (not me) believe Maho's >> builds are "releases". Here is the problem. > >That might indeed the problem. But your solution is wrong. Not >providing milestone builds surely is not a solution to a communication >problem. > I've been watching this brew for a while so here goes the 'mash':
1. Eric has a legitimate complaint that Maho was building when it was not worthwhile. However, without these builds we would have not known the fixes did not work. 2. Eric provides 'private' builds with the fix. Where is the CWS for the fix? (See comments later in this message.) 3. Milestone builds should have a readme.txt that states they are for testing only. Not for use in a production environment. Data loss may and does happen (take a look at the calc problem.) 4. We need not 'bash' Maho. He is doing what he said he would do and continues to provide builds for the Mac PowerPC and Intel systems based upon Tiger. See my comments below. >> I think Jason mails is a good summary of what I think. > >But he doesn't mention the root cause either: > >Why do people think that the builds are releases. >Obviously linux and windows users have no problem in understanding >that Pavel builds are snapshots/testing builds. Are Mac users more >stupid that linux/windows users? I don't think so. > Mac users pride themselves on being smarter than Windows and Linux users. I doubt that the problem is the majority of Mac users, but only one or two of them. >And now comes a "weired" theory: >Maybe people download the snapshots, because there just are too few >"official" testing-releases? The Mac Porting group needs to advertise the fact that OpenOffice.org 3.0 is in beta but the builds are not stable and many issues exist that need to be fixed before an actual release to the public with a listing of known problems and workarounds, if they exist. > >The port makes so much progress, that a snapshot that is a couple of >months old just doesn't reflect the current status. >I go further: Such an old status can even be worse than a crash in a >newer version, when the rest of the system feels snappier, feels more >integrated/more impressive. > Older builds with problems should be removed from all sites, not just Maho's. >> [...] >>> It is true that some builds have problems. These faults are a problem with >>> QA. (of the cws that are about to be integrated, or just because the >>> milestones/the combination of the cws >>> is not tested as well as it should). >>> This is not Maho's fault as QA is not his job. >> >> Just about the job, isn't it Maho who does manage QA ? > >of course everywhere I wrote "Maho doesn't do QA" or "Maho's job is >not QA" I was refering to his status against the build he provides. He >is not responsible for doing Mac-Teams's QA work, he is not >responsible for testing every build he provides. > Agreed. There are not enough hours in the day for Maho to test every build on the good-day.net site. >> [...] >>> He spends his time and ressources on doing builds, not development, not >>> QA. >> >> As QA lead, this is not the best example :-) > >being QA-lead doesn't mean one has to do QA on everything himself >(more the contrary, do the administrative stuff, give guidance, but >not do the QA itself) > That is very true. I would love to lead a team with many members where I could assign various tasks to. Unfortunately, this does not exist for every team. >>> That's why it is even more important that people do find those bugs. >>> But he also wrote before: He doesn't follow development and doesn't want >>> to go through the mail-archives himself. >> >> This is not normal, and I contest here : when one has the responsability of >> doing public builds, the minimal requirement is to keep an eye upstream. > >Again: Milestones are meant to not introduce any breakers. If we >cannot rely on that rule, then the whole cws process, the whole >organization of development can be thrown away. > The milestone builds combine one, several or many CWS fixes that may introduce other problems. We need them to keep moving forward. >The way in OOo is: File an issue, put him on cc. Or write a mail and >cc him in that mail. >He has stressed multiple times that he is not active in porting >efforts on the Mac. He is just a provider of builds. He is active on the FreeBSD porting effort. He is not on the Mac effort and more needs to be done to provide feedback to him to insure that builds on his system are not distructive and that builds are marked as development/testing. Complaining about the build process does not fix it. The project needs Maho to provide builds as a public service. This is what he is doing and is doing a very fine job of it. Building 'private' builds and then not completing the process is not. Fixes need to be moved forward in the process and not held, unless the fix introduces additional problems. I don't think the crash fix does this based upon what I'v read here. James McKenzie --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
