IRC log for #koha, 2008-02-02

All times shown according to UTC.

Time S Nick Message
12:55 fbcit g'morning koha
13:03 fbcit chris around?
13:07 gmcharlt hi fbcit
13:07 fbcit hey gmcharlt
13:08 fbcit you're here early :)
13:08 fbcit do you know if there is any place in the staff client that patron images are displayed?
13:14 gmcharlt yeah, that would be one place for it
13:15 fbcit ~ line 372
13:15 fbcit if borrowerimages subdir were added to the repo, installer should pick up on it and create it, right?
13:17 gmcharlt looks like patron images are pulled in through Javascript, so wouldn't be much reference to them in code
13:18 gmcharlt creating directory, possibly adding it to map in Makefile.PL and sticking a blank REAME file in it should do it
13:18 gmcharlt issue is that permissions need to be set up so that Apache user (e.g., www-data) can write to the directory
13:18 gmcharlt so would need to put in instructions about setting permissions or ACLs, etc.
13:19 gmcharlt or else sidestep the issue by storing the images as BLOBs in MySQL
13:20 gmcharlt anyway, gotta to start shoveling snow -- bbiab
13:20 fbcit :)
13:59 [K] *** join #koha@FreeNode: Comete
14:48 fbcit hey atz
14:49 atz greets fbcit
14:53 fbcit atz: Have a good vacation?
14:58 hdl hi
14:58 fbcit hi hdl
14:58 hdl anyone working on Fines or Overdues around ?
15:15 slef hi all - should MARC21 be listed in koha-tmpl/intranet-tmpl/prog/en/​modules/admin/z3950servers.tmpl? Is USMARC standing in for it?
15:18 gmcharlt slef: I believe that USMARC is what the record syntax is still called for Z39.50 purposes
15:19 gmcharlt self: but the label could be changed to USMARC/MARC21 or somesuch
15:19 slef gmcharlt: ANSI slow to update?  It would be good if we can change the user-visible label, as I've just had another new koha librarian ask me about it.
15:20 gmcharlt slef: I'll make a patch to change the user-visible label unless you want
15:20 gmcharlt want to, rather
15:20 slef I've all the files open, so I may as well
15:20 gmcharlt ok, thanks
15:22 gmcharlt hdl: question about ISO-5426 -- is it equivalent to just the ANSEL part of MARC-8, or does it (or a related ISO standard) include the MARC-8 encodes for Chinese, Japanese, Korean, Arabic, etc.?
15:23 hdl only ANSEL part of MARC-8
15:24 hdl I got the hash from marc4j.
15:25 hdl I belive we could investiguate and build up a better hash.
15:25 hdl Have you tested that on some data ?
15:25 gmcharlt hdl: thanks.  I wonder if there's a way to get MARC::Charset to convert just ANSEL
15:26 gmcharlt hdl: tried in context of Z39.50 searches of LC -- when encoding is ISO-5426, conversion seems to not handle the diacritics correctly
15:26 gmcharlt setting encoding to MARC-8 (which uses MARC::Charset) seems to work better
15:27 hdl On BNF encoding, it works well for diacritics.
15:27 hdl And they claim to have biblios encoded in ISO5426.
15:29 gmcharlt 'BNF' is the French national library's Z39.50 server?
15:29 hdl Yes.
15:30 hdl And marc8_to_utf8 failed on that.
15:32 gmcharlt hdl: what's a good word for a title search of BNF?
15:32 hdl[…]frame/infopro.htm
15:32 hdl Les misérables
15:33 hdl or ISBN :2020224526
15:39 gmcharlt hdl: alas, that server is not accepting connections at the moment
15:40 gmcharlt hdl: do you know of another French Z39.50 server that works in the same way as BNF?
15:40 hdl SUDOC
15:41 gmcharlt i.e.,
15:41 hdl,439,mode-public.html
15:41 hdl yes
15:49 gmcharlt hdl: today just isn't my lucky day -- that one is not accepting my connections
15:53 gmcharlt hdl: ok, found one I can connect to: Univ. de Rennes
15:53 gmcharlt[…]fr%3A210%2Frennes
15:54 gmcharlt do diacritics look correct for you if encoding is set to ISO-5426 (they don't for me, but MARC-8 does work for me)
16:00 gmcharlt hdl: more info -- for Rennes Z39.50 server (ti=chat) diacritics still look wrong in ISO_5426 mode when code is at currently 3.0 -- i.e., without my charset patch of yesterday
16:00 gmcharlt hdl: do they look correct to you?
16:01 hdl trying now
16:07 hdl Etude comparée du comptage des éosinophiles chez le chien et chez le chat par un automate à variation d'impédance l'Hemavet 950 (Drew Scientific) et deux méthodes manuelles
16:07 hdl looks right for me.
16:09 hdl But it comes from SUDOC.
16:09 hdl ARTUS seems to be USMARC.
16:09 hdl (No Title)
16:24 gmcharlt hdl: I guess I'll have to try my testing later when BNF or SUDOC start accepting connections from me
16:25 gmcharlt hdl: in meantime, if you could check whether my charset patch breaks anything for your testing of BNF or SUDOC, I'd appreciate it
16:25 hdl can you send it to me ?
16:28 gmcharlt hdl: I've e-mailed it to you
16:30 hdl Patch doesnot apply on
16:33 gmcharlt hdl: sorry missed a preceding patch -- will e-mail it to you
16:35 gmcharlt hdl: sent
16:45 hdl
16:45 hdl It doesnot work.
16:47 hdl :(
16:49 gmcharlt hdl: what happens when you set z3950servers.encoding to 'MARC-8' and search again?
16:58 hdl It Breaks.
16:58 hdl But ISO_5426 works.
17:00 gmcharlt hdl: so if I understsand correctly, with my patch, ISO_5426 still works for the SUDOC server?
17:00 hdl (testing on BNF only)
17:01 hdl But I still have some problems :
17:01 hdl Éd. du Seuil
17:01 hdl Should be Éd. du Seuil
17:02 gmcharlt hdl: the diacritic is not coming across in my IRC channel -- what is the name of it?
17:02 hdl Capital E Acute
17:11 hdl gmcharlt: Have you worked on Authorities ?
17:11 gmcharlt hdl: still in process -- the authorities indexing I posted about on koha-devel is part of it
17:14 hdl I sent a patch which allowed authid to be user defined subfield.
17:14 hdl Have you seen it ?
17:15 gmcharlt hdl: not yet, but sounds useful
17:15 hdl I cannot see this post any more.
17:15 gmcharlt which one -- the authorities indexing one?
17:36 gmcharlt gmcharlt-- # missed that BNF requires a username and password
17:36 gmcharlt hdl: I just sent you another patch -- should fix invocation of ISO_5426 conversion from a Z39.50 search
17:44 gmcharlt hdl: found big difference between ISO-5426 and MARC-8: the combining cedilla is \xD0 in 5426, \xF0 in MARC-8 ANSEL
17:44 gmcharlt hdl: now I know why MARC::Charset doesn't work for it
18:02 atz whoa.... microsoft buys yahoo for $44 Billion ?!
18:05 gmcharlt atz: is the deal complete?
18:06 atz just a bid right now... but a rather persuasive one, I'd say
18:06 gmcharlt nah, Yahoo should hold out for more ;)
18:17 slef US+EU regulators will call it in anyway
18:19 gmcharlt slef: by that you mean kill the deal on antitrust concerns?  hope the EU regulators would do so; I'm not so sure about US
18:19 slef either way, just the delay may mean that if Yahoo agrees then if it is blocked/obstructed too much, Yahoo will tailspin and be bought by someone else
18:20 gmcharlt yep, it's possible
18:20 gmcharlt I haven't been following Yahoo -- has it been on the way down recently?
18:21 slef no, but doesn't seem to have been on the up like google
18:21 slef and microsoft are big anyway
18:22 slef I don't know what analysts think about Wikia Search - it's possible that they reckon it will shift one of the top 4, as Jimmy Wales has media buzz
18:24 slef Yahoo's last results were down, apparently, but even google disappointed
18:24 slef from a libraries point of view, it's sort of disappointing that the big properties are mostly search
18:25 slef then again, if you are a successful non-proprietary search, you won't appear in the list of top properties that the media often quote
18:27 atz anybody know the ppl behind "Alpha-G" consulting that is apparently working to deploy evergreen ?
18:27 atz spinoff from equinox?
18:31 gmcharlt atz: the principal John Craig appears to have been a Dynix; looks like he left it to found Alpha-G before Sirsi bought it
18:31 gmcharlt a Dynix guy, rather
18:31 atz interesting
18:32 gmcharlt he has a pretty detailed profile on LinkedIn
19:42 fbcit is there a quick way to grab a member's cardnumber based on borrowernumber?
08:38 chris evening

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