IRC log for #koha, 2005-11-24

All times shown according to UTC.

Time S Nick Message
12:45 hdl hi osmoze
16:04 chris morning
17:57 kados hey hey
17:57 thd hello kados
17:58 kados drum roll please ... /me has an announcement
17:58 kados
17:59 thd my NetBSD CVS has been compiling for many hours and I have made a mistake oops
18:00 chris congrats joshua
18:00 rach wooo cool joshua
18:01 thd kados: does that mean that you have now left Koha for PINES? :)
18:01 kados thd: nope
18:02 kados thd: it means LibLime will be working on both Koha and PINES
18:02 kados which is a good thing for both projects
18:02 chris yep
18:02 thd kados: I assumed that you always were working with both
18:02 kados thd: well yes ... but now I can get paid to work on both ;-)
18:03 thd kados: I assumed that is why both projects share some common bugs/features :)
18:03 thd kados: getting paid is good
18:05 thd chris: can you reconstruct koha list subscribers since May from the headers in archived messages?
18:05 chris probably
18:06 chris tho im not sure ill try
18:06 chris its the unsubscriptions that i wish i could reconstruct
18:07 thd chris: I expect some people like myself, who subscribed since that time, will assume that your mail system is still down and not bother checking the current announcement at
18:09 thd chris: no one really ever wants to be unsubscribed from the koha list.  All the worlds most important information is there :)
18:09 chris youve obviously never seen my inbox then
18:09 chris all the ppl who cant read the headers, or the footer at the bottom of each message want to unsubscribe :)
18:12 thd chris: I have managed a subscription list before and remember some very unfriendly demands to unsubscribe.  They all secretly wanted to subscribe though.
18:13 chris :)
18:14 thd chris: Do you not have an archive of your inbox with those requests to unsubscribe?
18:14 chris nope
18:15 chris i have about 11000 messages in my inbox, but not the unsubscribe notices
18:15 chris once i actioned them, they were deleted
18:15 thd chris: well, they did not really want to unsubscribe, as I said.
18:16 chris heh yeah, they'll ask again if they really did
18:20 thd chris: assuming that you will not recover the post May subscribers instantly I will subscribe myself again now.  I assume that you will avoid duplicate mailing list subscriptions for those who resubscribed manually.
18:21 chris yeah mailman is good like that
18:22 chris it wont let duplicates subscribe
18:26 thd chris: that is a good reason for the popularity of mailman.  I have never used it myself.
18:30 thd chris: I cannot even subscribe myself twice.  Error, you are already a member.
18:30 chris excellent
18:30 chris thats all good then
18:32 thd kados: Are the new Perl-ZOOM bindings really likely to be finished and fully documented in a week?
18:36 thd chris: what had caused the failure of your mailing list archive backup aside from not testing it?
18:37 kados thd: should be ready this week
18:37 chris it wasnt running
18:37 kados thd: next week by the latest
18:37 thd chris: Do mean the cron job stopped running?
18:39 chris it was a double failure, we are building a new backup server and are having problems with the raid controller, so the offmachine backups werent running, and the mirroring to the spare disk in the machine wasnt happening either, just a big mess, we had offsite backups off everything barring the mailing lists (koha is just one of them unfortunately)
18:40 chris and the disk physically died, ie we sent it off to specialists and they tried everything, new heads, etc and couldnt get a thing off it
18:41 chris wasnt a fun weekend rebuilding
18:41 thd kados: Does that include complete documentation with full explanation of options, syntax, descriptions of what they are used for, and detailed usage examples?
18:44 thd chris: Fortunately I only electrocuted my keyboard controller with static but I do have some problems with the hard disc I am working on now where only home is comprehensively backed up.
18:44 kados thd: you writing a book?
18:44 kados thd: :-)
18:45 kados thd: we'll have to see what they release
18:45 kados thd: then, if it's not adequate, I get to say what we'd like improved, etc.
18:45 thd kados: Yes, several books always.
18:45 kados hehe
18:46 thd kados: There is a standard of documentation that used to be universal.
18:47 thd kados: It is still followed at Microsoft as one example for what they want to publish, although, they may only ship a ten page manual with their products.
18:48 thd kados: The documentation that I describe is still available if you order it for their programming tools.
18:50 thd kados: Which with Microsoft is almost all their applications since they all incorporate some form of Visual BASIC or .NET API.  Not that I use them anymore, nor did I ever use .NET.
18:53 thd kados: my point is merely that what I described above had been the standard for all documentation at one time and the current version of net::Z3950 falls somewhat short and is much more difficult to use as intended on that account.
18:53 kados thd: it was produced by a volunteer ... not a multi-million dollar corporation :-)
18:55 thd kados: I do understand the economics of the situation but guessing at how something is supposed to work is not always easy :)
18:57 thd kados: actually, I would expect that the documentation would point outside itself to other documentation.
19:00 thd kados: Documentation would ultimately point to Z39.50 standards documentation.  However, that official documentation itself is completely inadequate for the database manipulation commands which are significant features for the new Perl bindings.
19:01 thd kados: I was entirely unsuccessful finding any proper documentation of how such Z39.50 extended services were actually supposed to work anywhere ever.
19:05 thd kados: all I ever found was just a general vague idea of what the Z39.50 extended services actually did without anything more than usage syntax that could only be helpful if you really understood how the extended services actually worked.
19:08 thd kados: Do I understand correctly that Z39.50 extended services are used for reading, inserting, and deleting records in the Zebra database?
19:08 kados thd: they can be used to manage all aspects of a zebra database
19:11 thd kados: the Z39.50 standards documentation or anything else that I have seen fails to document there use in any way that would be adequate unless you already knew how to use them.
19:11 thd there/their
19:14 thd kados: I hope the new net::Z3950 documentation remedies that deficiency along with others.  Grasping how something works by intuition is fine except where the necessary intuition is not held in common :)
19:20 thd kados: The deficiency that I am referring to is not exclusive to the net::Z3950 documentation but to the NISO Z39.50 documentation underlying, at least for extended services.
19:23 thd kados: I assume Mike knows how to remedy that on his own for his own documentation.  I understand that you have not hired the Library of Congress.  They have a difficult time addressing a problem in one year never mind a few weeks.
19:28 thd kados: (The Library of Congress being the maintenance agency for the ANSI/NISO Z39.50 standard and not Mike.  Mike cannot change the official documentation but he can compensate for the deficiency in the official documentation with his own documentation.)
19:40 kados koha-bugs moved over to savannah
19:41 kados and bugzilla components updated with new list email
19:42 kados but I bet I still need to tell mailman to allow mail from bugzilla
19:44 chris test came thru ok joshua
19:46 kados great
19:47 chris 2 of them even
19:47 kados yea ... sorry bout that :-)
19:48 kados thd: added you to savannah btw
20:24 thd kados: thank you.  I have a very some bug fixes to start with but I am very behind on everything since I have not had steady access to my system in the past few weeks.
20:28 thd kados: What do you use for cloning a system.  I am recompiling NetBSD to create a fix the author of G4U (Ghost4Unix) has not managed to solve yet.  Norton Ghost does not seem to be a viable option any longer, at least not since Ghost 2003.
20:31 thd kados: Or do you merely reinstall and reconfigure a completely fresh system from base with merely copying configuration files in /etc every time you need a new system.
20:39 thd My question was off topic.  What is the most robust tool for copying a whole system with all installed software and configuration files including Koha to new heterogeneous hardware of the same general hardware architecture?  Now my question is on topic. :)
21:43 kados thd: I use tar :-)
22:16 thd kados: I should have said that G4U uses dd.  Assuming the file system that you are copying with tar is mounted as read only to preserve state perfectly, what happens when tar reaches a bad block on the mounted file system?
22:18 thd s/file system/file system\/partition/
02:06 indradg kados, evening!

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary