June 1999 - Zope - Zope lists

RE: [Zope] Split the list again?
by Ken Manheimer 16 Jun '99

16 Jun '99

Perhaps it's time to reraise the possibility (spectre?) of having a newsgroup - with sub groups etc. Or at least splitting the newslist. Once again, i won't have time to mess with this for another few weeks - but i'll put it on the chart when i resurface. Ken Manheimer klm(a)digicool.com > -----Original Message----- > From: Michael Bernstein [mailto:mbernstein@profitscape.net] > Sent: Tuesday, June 15, 1999 4:54 PM > To: zope(a)zope.org > Subject: [Zope] Split the list again? > > > I think it may be time to revisit the issue of splitting the main > zope(a)zope.org list. > > Proposed sub-lists: > > zope-install (compiling, installing zope, apache issues, etc.) > > zope-announce (zope and product announcements, including commercial > services, moderated) > > zope-bugs (zope bugs, what else did you expect?) > > What do you all think? > > Michael Bernstein. > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://www.zope.org/mailman/listinfo/zope > > (For developer-specific issues, use the companion list, > zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

1 0

RE: [Zope] Split the list again?
by Ken Manheimer 16 Jun '99

16 Jun '99

> -----Original Message----- > From: Alexander Staubo [mailto:alex@mop.no] > Sent: Wednesday, June 16, 1999 11:15 AM > To: Zope Mailing List (E-mail) > Subject: RE: [Zope] Split the list again? > > > I never figured this on out. There's nothing harmful about Reply-To. Because you don't see it doesn't mean it's wrong!-) In brief, reply-to is not intended for the mailing list to grab replies. Munging of it breaks that intended purpose. The actual purpose is to enable senders to specify an alternative address, other than the one from which they're sending the message, as the destination for replies. This is very simple, and very useful for people who send mail from multiple places and want to be assured that they'l get the replies in one place. It's also used to specify different response addresses depending on what's being sent. In any case, it's a tool for the original sender, not the intermediate mailing list. Having the mailing list system stomp the reply-to for its own purpose loses that essential information. This is "evil", in programming parlance. Having the mailing list system try to compensate by, eg, altering the 'from' address to preserve the reply-to info loses the from info, and compounds the sin. This is the path to irredeemable corruption - known in some circles as excessive bureacracy:-) There are other reasons for not munging the reply-to - i think the cited article goes into them, i don't recall exactly which. It's sufficient to say that (1) the fact that you don't understand the "reply-to munging considered harmful" rationale is not adequate reason to dismiss the rationale, and (2) the controversy alone, plus the above specific reason - plus (3, 4, 5...) reply-to's actual intended purpose, plus the principle that minimal intervention is desirable, (particularly when it comes to email!) is enough to convince us to think twice before changing it back. Wouldn't that be enough for you? In another message on the subject, Phillip Beazley wrote: > >reply-to list also leads to a higher incidence of mail loops, stupid > >vacation messages, and other rubbish, being sent to the list. > If that's the case, the list administrator needs to get a better list server. I won't take offense, though i'm one of the developers of the list server in question (mailman, which has its pluses and its minuses). I do think it's presumptuous - particularly in light of the litany of reasons above - to dismiss the cited article, and opinions of many savvy email folk. There are good reasons for the decision - and we're not planning to change it. Ken Manheimer klm(a)digicool.com

1 0

RE: [Zope] external search engines (altavista, etc.)
by Brian Lloyd 16 Jun '99

16 Jun '99

> > Can search engines such as AltaVista mine the information > out of a Zope > > website. I have seen responses to other questions that > there are ways > > to use a search engine to search a Zope website, but can > the commercial > > laces search the website. It would be nice for other places to know > > about things in your website, unless you want it to be private. > > Just for the record, I've seen several robots crawl > through my Zope sites > and later found them indexed on the comercial engines. So, > yeah, it works! Just for the record, the main thing you need to watch out for as far as search engines indexing your site is funny links that make use of acquisition. A side effect of the way Acquisition works is that an object can "acquire itself from itself" - in other words a url like: http://myserver.com/foo/foo/foo/foo/foo/foo/foo/foo/ ...is perfectly valid (and will be found by Zope through Acquisition) even though there is only really one top level foo object :) If foo in this case happens to be a document, having a link like: <a href="foo">...</a> in it will probably cause some search robots to lose their minds :) Brian Lloyd brian(a)digicool.com Software Engineer 540.371.6909 Digital Creations http://www.digicool.com

1 0

RE: [Zope] Split the list again?
by Alexander Staubo 16 Jun '99

16 Jun '99

I posted mine at 17:16 (local time), received your reply at 17:31, fifteen minutes. That's not bad if the response to a query is good. :-) If speed is essential, nothing is stopping you from cc:ing to the person you're conversing with. -- Alexander Staubo http://www.mop.no/~alex/ "`Ford, you're turning into a penguin. Stop it.'" --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ >-----Original Message----- >From: zope-admin(a)zope.org [mailto:zope-admin@zope.org]On Behalf Of Mike >Pelletier >Sent: 16. juni 1999 17:31 >To: Zope Mailing List (E-mail) >Subject: RE: [Zope] Split the list again? > > >On Wed, 16 Jun 1999, Alexander Staubo wrote: > >> Non-fiddling with Reply-To is, imho, more harmful than not. >How? Well, >> for every follow-up I get, I also get one explicitly directed to my >> email address, because people hit "Reply to all" and instead >of having >> their follow-up sent to just "zope(a)zope.org", they send it to >> "alex(a)mop.no; zope(a)zope.org". Needless to say, for every >follow-up I get >> _two_ emails. > > I have to say I find that very convenient. Often I'm on >lists that take >ten to thirty MINUTES (an ETERNITY!) to deliver a message to >me via the list. >The courtesy copy gets to me in a second or two, which is >really handy when I >have a problem stopping further progress. As a test, I >haven't CC'd you, and >I replied immediately upon receiving you email. Now you can >let me know how >slow the Zope lists are. > >-- > --- | Mike Pelletier | Paid to be a geek! > /opeware! | mike(a)zopeware.com | You'd have to >hire KryoTech > --- | ICQ:7127228 | To get much cooler. >http://www.zopeware.com | (519)725-7710 | -Haiku >by Retu Pmoc > > >_______________________________________________ >Zope maillist - Zope(a)zope.org >http://www.zope.org/mailman/listinfo/zope > >(For developer-specific issues, use the companion list, >zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

2 1

RE: [Zope] CVS - broken
by Ken Manheimer 16 Jun '99

16 Jun '99

> -----Original Message----- > From: Phil Harris [mailto:phil@philh.org] > Sent: Wednesday, June 16, 1999 11:26 AM > To: zope(a)zope.org > Subject: [Zope] CVS - broken > > > Hi all, > > I just tried to do a CVS chekout and had an error message: > > cvs [checkout aborted]: recv() from server zope.org: Broken pipe Thanks for the heads-up - inetd on the machine has been restarted, and it's working now. (This is the second time in the last few weeks that this happened - we're going to have to look into a long term fix.) > Just thought I'd let you know, ;) We appreciate it! For future reference, this is the sort of thing that's better directed to root or webmaster @zope.org, so it'll get more immediately to our attention - stuff can get lost in the volume of the newsgroup. Ken Manheimer klm(a)digicool.com

1 0

RE: [Zope] Split the list again?
by Alexander Staubo 16 Jun '99

16 Jun '99

I never figured this on out. There's nothing harmful about Reply-To. The document you refer to tries to build an argument out of inserted Reply-Tos modifying the standard behaviour of the "Reply" operation, which is complete bs. Non-fiddling with Reply-To is, imho, more harmful than not. How? Well, for every follow-up I get, I also get one explicitly directed to my email address, because people hit "Reply to all" and instead of having their follow-up sent to just "zope(a)zope.org", they send it to "alex(a)mop.no; zope(a)zope.org". Needless to say, for every follow-up I get _two_ emails. To avoid having my inbox cluttered with redundant mail, I filter everything into my Zope-list folders, but I still get two emails for every follow-up. Madness! Common sense would tell you that 99% of all replies go to the list -- that's the whole point. Why cater to the 1% where the followup is sent privately? I'm on a lot of mailing lists, and Zope's are the only ones I've come across that has not implemented the courtesy Reply-To fix. It's incredibly, mindbendingly annoying. -- Alexander Staubo http://www.mop.no/~alex/ "`Ford, you're turning into a penguin. Stop it.'" --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ >-----Original Message----- >From: anthony(a)nextTelecom.com [mailto:anthony@nextTelecom.com]On Behalf >Of Anthony Baxter >Sent: 16. juni 1999 16:12 >To: Phillip Beazley >Cc: zope(a)zope.org >Subject: Re: [Zope] Split the list again? > > >>>> Phillip Beazley wrote >> Oh, and how about having the reply-to address of posts that of the >> mailing list unless specified by the user? > > >noooooooo. Bad, bad, bad, naughty. Don't fiddle the reply-to >on mailing >lists. I'm on some lists that do, some that don't - and the ones that >do are _far_ more irritating to work with. > >see also: http://www.unicom.com/pw/reply-to-harmful.html > >Anthony > >_______________________________________________ >Zope maillist - Zope(a)zope.org >http://www.zope.org/mailman/listinfo/zope > >(For developer-specific issues, use the companion list, >zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

5 4

RE: [Zope] How stable is the 2.0 alpha release?
by Michel Pelletier 16 Jun '99

16 Jun '99

> -----Original Message----- > From: Alexander Staubo [mailto:alex@mop.no] > Sent: Wednesday, June 16, 1999 11:00 AM > To: Zope Mailing List (E-mail) > Subject: RE: [Zope] How stable is the 2.0 alpha release? > > > >-----Original Message----- > >From: Michel Pelletier [mailto:michel@digicool.com] > >Sent: 16. juni 1999 16:12 > >To: 'Alexander Staubo'; Zope Mailing List (E-mail) > >Subject: RE: [Zope] How stable is the 2.0 alpha release? > > > > > > > > > >> -----Original Message----- > >> From: Alexander Staubo [mailto:alex@mop.no] > >> Sent: Tuesday, June 15, 1999 9:13 PM > >> To: Zope Mailing List (E-mail) > >> Subject: [Zope] How stable is the 2.0 alpha release? > >> > >> > > > >*Seems* stable to me. > > > >It may be very stable. You may setup a vast production site with it. > >You may find a very excellent bug that may trash all your work. > > > >If it makes you feel any better, we'll fix the bug if it's there. ;) > > Great, what about all the things mentioned in the 2.0.0a1 > release notes? > Release notes mention that alpha 2 "will be released in the > week of May > 26," which obviously did not happen. Are you behind schedule? > This is my personal opinion, as I don't do any 'scheduling', but I think that scheduling software releases in general is a Microsoftian mistake; it's done and ready when it's done and ready. I may be getting myself in trouble here, as the official company policy may differ. I do know that we have been fulfilling several customer requirements lately that were somewhat unexpected, and, of course, we can't turn down a good money making job on the excuse that it will take time away from our free product. We spend most of our time working on Zope, but we have to spread the time around for the people who want to give us money too. We are also working on alot of non Zope core products, like Catalog, Portal Tool Kit, and Tracker. All in good time. Zope is cool, and it will get much, *much* cooler. > The alpha 1 todo list is relatively serious, eg., "Most database > adapters are likely to be problematic [...] when running Zope 2 with > multiple threads" -- eek! Which ones are subject this to this problem? > Probably all of them except for the ones we write (which will be made thread safe). The problem is that most of the python libaries (which we didn't write) that the DAs (most of which we didn't write either) are written against (and in some situations the DAs themselves) are not threadsafe. This is a problem that is *way* beyond our control, but we'll probably try and fix it anyways cuz we're nice guys. > I could not find documentation on compatibility issues with ZODB2, > either (probably covered on this list at some point; would you care to > repeat?), ie. can I just copy by data.bbb into the alpha version? Yes, 2.0 alpha will work fine with an existing Database *if* you run it in ZODB2 mode (no concurrency). If you run it in ZODB3 mode (the default) your Data.bbb file will not even be used, rather, a different file called Data.fs (fs = File Storage, remember, ZODB3 can use multiple backend stores) will be used. Data.fs and Data.bbb are two different formats, and are incompatable. Real Soon Now we will have a conversion script to upgrade from Data.bbb format to Data.fs format, but I for one, will promise no date. > > Do you recommend updating the sources from the public CVS after > installing alpha 1? I recommend for the non developer interested in kicking the alpha tires to follow the release schedule, ie alpha 1, alpha 2, alpha n, beta 1, beta, n...final. If you are a developer type, and you like to get bleeding edge bug fixes, then I recommend you update against the CVS daily, and subscribe to the zope-checkins mailing list, which gives you a blow by blow description of CVS checkins done. http://www.zope.org/mailman/listinfo/zope-checkins -Michel -Michel > > >-Michel > > > >> -- > >> Alexander Staubo http://www.mop.no/~alex/ > >> "`This must be Thursday,' said Arthur to himself, sinking low over > >> his beer, `I never could get the hang of Thursdays.'" > >> --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ > >> > >> > >> _______________________________________________ > >> Zope maillist - Zope(a)zope.org > >> http://www.zope.org/mailman/listinfo/zope > >> > >> (For developer-specific issues, use the companion list, > >> zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) > >> > > > > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://www.zope.org/mailman/listinfo/zope > > (For developer-specific issues, use the companion list, > zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

2 1

RE: [Zope] CVS - broken
by Michel Pelletier 16 Jun '99

16 Jun '99

> -----Original Message----- > From: Phil Harris [mailto:phil@philh.org] > Sent: Wednesday, June 16, 1999 11:26 AM > To: zope(a)zope.org > Subject: [Zope] CVS - broken > > > Hi all, > > I just tried to do a CVS chekout and had an error message: > > cvs [checkout aborted]: recv() from server zope.org: Broken pipe > > Just thought I'd let you know, ;) > Thanks, we'll look into it. -Michel > Phil > > phil(a)philh.org > > > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://www.zope.org/mailman/listinfo/zope > > (For developer-specific issues, use the companion list, > zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

1 0

RE: [Zope] ZODB3 question
by Michel Pelletier 16 Jun '99

16 Jun '99

> -----Original Message----- > From: Jimmie Houchin [mailto:jhouchin@texoma.net] > Sent: Wednesday, June 16, 1999 11:35 AM > To: zope(a)zope.org > Subject: [Zope] ZODB3 question > > > From thread: RE: [Zope] Large Portal requirements (was: > Other web servers?) > At 03:15 PM 6/14/99 -0400, Michel wrote: > >In the future, we will be have the client server model of > ZODB3 (which > >will be a commercial product) that lets you have multiple > Zope frontend > >processes (possibly on different machines) accessing > multiple database > >backend processes (possibly on different machines). Also, ZServer is > >currently multi-threaded, but there is no theoretical limitation to > >making it multi-process, in which each Zope instance will > have it's own > >Python and thus it's own global Python lock, thus giving the OS the > >ability to schedule multiple processes among multiple processors, a > >little heavier than threads on multi-processor, but *vastly* > easier to > >impliment and far more stable. > > And then... > > From thread: RE: [Zope] How stable is the 2.0 alpha release? > At 11:12 AM 6/16/99 -0400, Michel wrote: > >Yes, 2.0 alpha will work fine with an existing Database *if* > you run it > >in ZODB2 mode (no concurrency). If you run it in ZODB3 mode (the > >default) your Data.bbb file will not even be used, rather, a > different > >file called Data.fs (fs = File Storage, remember, ZODB3 can > use multiple > >backend stores) will be used. Data.fs and Data.bbb are two different > >formats, and are incompatable. Real Soon Now we will have a > conversion > >script to upgrade from Data.bbb format to Data.fs format, > but I for one, > >will promise no date. > > > To clarify my understanding. > > Will their be a non client-server version of ZODB3 which > offers concurrency > that will be a part of the non-commercial, open source part of Zope? > Yep. Just the client server part will be commercial. -Michel > Jimmie Houchin > > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://www.zope.org/mailman/listinfo/zope > > (For developer-specific issues, use the companion list, > zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

1 0

CVS - broken
by Phil Harris 16 Jun '99

16 Jun '99

Hi all, I just tried to do a CVS chekout and had an error message: cvs [checkout aborted]: recv() from server zope.org: Broken pipe Just thought I'd let you know, ;) Phil phil(a)philh.org

1 0