Time Nick Message 12:04 thd hdl: there is also the confusing verbose but pretty representation modelled after what LC has for MARCXML. 12:04 thd hdl: http://library.neu.edu.tr/kohanamespace/ 12:07 hdl Is it what they already use ? 12:07 thd hdl?: I am not certain what he has in production 12:08 thd hdl: he certainly has it working on a test basis at least 12:10 thd hdl: the only significant penalty for this design is greater time required for indexing large records if authorities are added 12:10 thd hdl: that is why larger meta-records should be in a supplementary database which is not updated in real time 12:10 hdl But this can be overcome with batch updates I thinf 12:10 thd yes 12:11 thd hdl: tumer now has real time updates set for 2 minutes so that he avoids timeout problems for Zebra 12:16 thd hdl: of curse I want to do more than add authorities 12:17 thd hdl: kados suggested adding FRBR relationships 12:18 thd hdl: I want to go as far as allowing MARC 21, UNIMARC, IBERMARC, Dublin Core, ONIX, OAI, etc. to coexist in the same meta-record for the same work. 12:19 thd hdl: maybe the non-MARC one could be translated to MARC first just to make things a little easier ;) 12:19 thd s/one/ones/ 12:21 thd hdl: I know that everyone will think that is crazy 12:23 hdl This IS a GREAT ideal. But need MUCH time, and I can't think who will be eager to get that stuff. 12:23 thd hdl: I still need a system which can function as a universal OPAC holding hundreds of millions of records. 12:24 thd hdl: as long as the foundation is there the scripting to support it can be built slowly over time 12:24 hdl But if we achieve this, it will be great. 12:25 thd hdl: in the Anglophone world LC already has added ONIX to MARC 21 records 12:26 thd hdl: OCLC, the worlds largest union catalogue has many many must be millions of Dubln Core records but they should if they invented and promoted them. 12:28 thd hdl: It would be great for UNIMARC and MARC 21 libraries to be able to use the alternate format when they acquire material which has never previously been translated. 12:28 thd s/translated/catalogued in the other record syntax/ 12:30 thd hdl: tumer reports that Zebra Proxy has the usemarcon configuration files for using usemarcon to translate between UNIMARC and MARC21 in both directions. 12:31 thd hdl: I just want to be able to search any important type of record consistently at a very exacting level of detail without needing to do repeated searches against different databases. 12:32 thd hdl: I think that consistent searching across different record types is a very general desire. 12:34 thd hdl: it is also something that the European Digital Library addresses. 12:34 dewey okay, thd. 12:34 owen dewey: what is it? 12:34 dewey or often necessary because the user may never successfully guess the authorised heading successfully unless the user is a librarian with years of experience or otherwise especially familiar with the authorised headings needed or something that the European Digital Library addresses. 12:35 owen dewey: forget it 12:35 dewey owen: I forgot it 12:39 thd hdl: http://www.theeuropeanlibrary.org/portal/index.html 12:39 thd hdl: takes me much longer to load now with all the new JavaScript 12:41 thd owen: is there a command for never learn it? 12:42 thd s/it/'it'/ 12:42 owen I don't know 12:42 thd dewey: ignore it 12:42 dewey thd: huh? 12:44 hdl dewey : what is it ? 12:44 dewey i don't know, hdl 12:44 hdl dewey : what is FRBR ? 12:44 dewey FRBR is not a record type 13:09 thd dewey: what is FRBR? 13:09 dewey hmmm... FRBR is Functional Requirements for Bibliographic Records. http://www.ifla.org/VII/s13/frbr/frbr.pdf has the canonical report. http://www.oclc.org/research/presentations/oneill/frbrddb2.ppt has an easily understood presentation. http://www.kaapeli.fi/~fla/frbr05/delseyModeling%20subject%20access.pdf has fun extensions. 13:12 hdl much more complete. 13:14 kados thd: you around? 13:14 thd yes 13:14 kados I just compiled a rule for finding Large Print using MARC 13:15 kados was wondering if you could check it 13:15 kados Large print = 13:15 kados ( 13:15 kados ( ((LDR-07='a' or LDR-07='c' or LDR-07='m') and 008-23 = 'd') or 13:15 kados ((LDR-06='c' or LDR-06='d' or LDR-06='i' or LDR-06='j') and 008-29='d') or 13:15 kados ((LDR-07='s' or LDR-07='b') and 008-23='b') or 13:15 kados (LDR-06='p' and 008-23='b') ) or 13:15 kados ( ((LDR-06='e' or LDR-06='f') and 008-29='d') or 13:15 kados ((LDR-06='g' or LDR-06='k' LDR-06='o' LDR-06='r') and 008-29='d') ) or 13:15 kados ( ((LDR-07='s' or LDR-07='b') and 008-22='d') or 13:15 kados 007-01='tb') 13:15 kados the qualifier names should be fairly obvious 13:16 kados this is why MARC must die :-) 13:16 thd kados: we should be storing things in a supplementary field so that we do not need to have those rules once he supplementary fields have been filled 13:17 thd s/he/the/ 13:18 thd kados: I already see that you left out some possibilities 13:18 kados thd: like what? 13:19 thd kados: there are textual field possibilities 13:19 kados oh goody :-) 13:20 thd kados: 020 $a; 300 $a; 500 $a, and maybe others 13:21 kados thd: they have 'official' values? 13:21 thd kados: if it were in 020 it would appear like 1234567890 (large print pbk.) or something like that 13:21 kados thd: or just free text? 13:22 thd kados: free text is tricky 13:22 kados well ... are they consistant? 13:22 kados ie, are there valid and invalid values? 13:23 thd kados: nothing is very consistent about bibliographic records which makes them more fun than the boring consistent records of the world. 13:23 kados :-) 13:24 thd kados: so you should certainly be trying to match anything a lazy cataloguer might have used such as lrg. prt. or whatever 13:24 kados thd: is there a list of those somewher? 13:25 kados thd: do they change from langauge to language? or are they consistant? 13:25 kados thd: AACR2 have official abreviations? 13:25 thd kados: if you inspect every MARC record in existence then you will have the list 13:25 kados hahah 13:26 kados owen: you may get a kick out of this: http://kados.org/marc_must_die_examples.txt 13:26 thd kados: I had a grand time just getting library of congress usage of abbreviations for hardcover and paperback complete in 020 13:26 kados owen: I'm compiling a 'most wanted' list for really bad design decisions in MARC :-) 13:27 owen Good grief! 13:27 kados yea ... show that to lauren and explain why Koha 2.4 isn't ready yet :-) 13:27 thd kados: and LC was just one library with fairly god standards of consistency compared to some 13:27 kados that represents at least 4 hours of digging 13:28 thd s/god/good/ 13:31 thd kados: there is no specific rule for large print in my 1978 copy of AACR. 13:32 thd s/AACR/AACR2/ 13:35 kados so what exactly _is_ 'large print' ... is it a format? 13:35 thd kados: it is a type size 13:37 thd kados: there is no reason the books themselves have to be taller although the preference in printing them is too keep the pagination the same. 13:37 kados OCLC has it listed as a 'subtype' in the 'format' category 13:38 thd kados: OCLC think as carefully as they ought to have when they designed that part of their search form. 13:38 thd kados: consider that they located some extra search possibilities they had indexed out of the way somewhere on the search form. 13:40 kados what extra search possibilities? 13:41 thd kados: that interface was written primarily for librarians to use so they did not usually worry about finding more than one format or type size when they were looking for some place to do an ILL request or copy catalogue a book n hand. 13:43 thd kados: the extra possibilities are merely the less important search restrictions which they included with less thought to their organisation at the bottom of the form. 13:45 thd kados: the top of the form is more carefully organised 13:46 thd kados: have you seen that they are now providing direct access to Open WorldCat with a simple search box at htp://www.worldcat.org . 13:47 kados yes 13:47 thd kados: has that only been a few weeks? 13:48 thd kados: there used to be a username and password dialogue box there for using the good search form that the FirstSearch interface has. 13:49 kados thd: it's only been about a week 13:51 thd kados: screen scraping that may be better than using SOAP with Google for the same content. 13:52 kados thd: that's what I'm going to do :-) 13:54 thd kados: I thought NPL had lost many of its 000, 007, and 008 fields. Have you been filling them from other fields contained in the records? 13:56 kados thd: yes 13:56 kados thd: they only lost 000s 13:56 kados and a few 008s 13:57 kados but it was pretty simple to reconstruct 13:58 thd kados: 008s were originally populated by script at many places form other information contained in the records when 008 was originally introduced. 13:58 thd s/places/big libraries/ 14:26 thd kados: 022 $a and 024 $a should also be considered along with 020 $a for large print. 14:27 thd kados: the problem with using 02X is that you may have more than one format recorded in a repeatable field which does not apply to your item. 14:29 thd kados: your mappings look good but I am not awake enough to see very well. 14:31 thd kados: I almost have a replacement for the wiki CSS that looks like the koha.org CSS. 22:51 thd kados: are you still awake? 01:49 hdl hi 01:55 Strait hi 01:55 dewey hey, Strait 01:55 Strait good morning :) 02:02 Strait bugger 02:02 Strait z39.50 is broken :( 02:03 Strait The requested URL /cgi-bin/koha/z3950/0 was not found on this server. 02:03 Strait that appears after the results have been displayed 02:04 chris ahh i think thats a bug in the npl templates 02:04 Strait yep, it is 02:04 Strait it works with default templates 02:04 chris i saw something about that, either on bugs.koha.org or the devel list 02:05 Strait i tried checking bugs.koha.org already 02:05 chris its fixed in the rel_2_2 branch of cvs 02:05 Strait do you know how to fix it? 02:06 chris its fixed in cvs .. so you could look at viewcvs on savannah that should tell you how it was fixed 02:06 chris whats the name of the tmpl? 02:06 Strait i think it's searchresult.tmpl 02:07 Strait in z3950/ 02:08 chris http://cvs.savannah.nongnu.org/viewcvs/koha/koha-tmpl/intranet-tmpl/npl/en/z3950/searchresult.tmpl?root=koha&only_with_tag=rel_2_2 02:08 hdl Strait: yes it is. 02:08 chris 1.2.2.2 02:08 hdl It returns 0 when no results found. 02:08 chris is where the fix was applied 02:08 hdl But not an HTML page. 02:08 hdl hi chris 02:08 chris http://cvs.savannah.nongnu.org/viewcvs/koha/koha-tmpl/intranet-tmpl/npl/en/z3950/searchresult.tmpl?root=koha&only_with_tag=rel_2_2&r2=1.2.2.2&r1=1.2.2.1 02:08 chris theres a diff 02:09 chris hi hdl :) 02:09 hdl how is the next Cormack ? 02:09 chris all good so far 02:09 hdl great 02:09 chris 22 weeks now .. over half way :) 02:10 Strait great, thanks a million :) 02:16 Strait works like a charm now :) 02:16 Strait i think i'll definitely have to upgrade to 1.2.2 at home 02:18 chris 1.2.2 ??? 02:19 Strait version 1.2.2.2 02:19 Strait well that 02:19 chris ahh of that template 02:19 chris righto 02:21 Strait i think the guys at emilda have done a good job with z39.50 02:21 Strait it's really quite elegant 02:21 chris yeah the next versions of koha handle z3950 much better 02:21 Strait http://demo.emilda.org/ 02:22 chris now there is a perl interface to zebra (what emilda uses) 02:22 Strait although featurewise emilda is nowhere near koha of course ;) 02:22 chris its possible for us to use it also 02:23 Strait yep 02:24 Strait it think the whole consept of adding books in emilda is worth checking out 02:24 Strait i think 02:24 Strait (it's so hard to write english :D) 02:24 chris :) 02:25 chris its all quite different for koha 3.0 02:28 Strait koha 3.0 is available through cvs? 02:28 chris well it wont work 02:28 chris but yep :) 02:28 Strait ah, ok 02:54 Strait hmm, i think we're gonna need a character conversion for the z3950 02:54 Strait from iso-8859-1 to utf-8 02:54 Strait to be able to import voyager records 02:55 Strait (finnish national library uses a voyager system) 02:55 Strait (for the time being ;)) 03:00 chris yep 03:01 chris its probably marc8 03:01 chris theres been tons of work done on conversion in the dev_week branch of cvs 03:01 chris and lots on the wiki 03:01 chris kados, paul and tumer know much more about it than me 03:02 chris http://wiki.koha.org/doku.php?id=encodingscratchpad 04:42 mohamedimran82 hello i have a problem with koha reservation 04:52 mohamedimran82 anybody on this list willing to spare time for errors in koha reservation 05:03 chris sorry mohamed, its friday night here so I wasnt watching my computer 05:03 chris whats the problem? 05:03 dewey the problem is if there is no data in the sort field zebra throws and error and does not even return the unsorted results 05:04 chris dewey: forget the problem 05:04 dewey chris: I forgot problem 05:05 mohamedimran82 thnk u very much but what about the reservation problem 05:05 mohamedimran82 i am able to issue a book to a person inspite of somebody else already reserving it ! 05:06 mohamedimran82 that too with out any warning 05:06 chris hmmm i havent seen that before 05:06 chris what version of koha, and what operating system? 05:07 mohamedimran82 koha 2.2.5 FC3 05:08 chris weird, i havent seen that problem before .. what templates? 05:08 mohamedimran82 default 05:08 chris no idea sorry 05:09 mohamedimran82 but all my reserves r logged in reservation pending list 05:09 mohamedimran82 never mind . but what about the reports 05:09 chris you should get a warning, that says its on a reserve, and do you what to issue it anyway 05:09 chris thats what happens for me 05:10 chris what report? 05:11 mohamedimran82 all my reports 05:12 mohamedimran82 for ex if i try to see borrower stats it throws up a message in error log 05:12 mohamedimran82 [Fri Aug 18 15:49:03 2006] [error] [client 192.168.1.2] DBD::mysql::st fetchrow failed: fetch() without execute() at /usr/local/koha/intranet/cgi-bin/reports/borrowers_stats.pl line 357., referer: http://192.168.1.7:8001/cgi-bin/koha/reports/borrowers_stats.pl 05:12 chris and you choose so options? 05:13 chris chose some options even 05:14 chris for eg 05:14 chris click row for category 05:14 chris and column for location 05:14 chris then you should get some results 05:16 chris does that work? 05:17 mohamedimran82 yeah it worked for borrowers. thnx 05:17 mohamedimran82 but why is there problems with wild cards 05:17 chris cool, you need to choose a column or a row 05:17 chris for all the reports 05:18 chris you dont have to choose any values, just what you want as the column or the row 05:18 chris otherwise it doesnt know how to build the table 05:19 mohamedimran82 ya it worked for a few reports. 05:20 mohamedimran82 by the way why dont u complete the documentation on the report section on kohadocs 05:20 mohamedimran82 its empty 05:20 chris cos i dont have time :) 05:21 chris if anyone wants to write it, we'd be happy 05:22 mohamedimran82 one more query 05:22 mohamedimran82 i have managed to authenticate my users from my college LDAP 05:23 mohamedimran82 now the catch. how to map those fields into koha db 05:23 mohamedimran82 when a ldap user logs in, all his personal details in opac are empty 05:24 mohamedimran82 if he is changing that, that a mail is sent to the admin to reflect that on the DB 05:24 mohamedimran82 why this redundant stuff 05:24 mohamedimran82 can we not pull up all the user details from the ldap and fill the same for each authenticated user 05:25 chris yes 05:25 chris thats an option in the ldap module 05:25 mohamedimran82 is it inside LOCAL & /LOCAL 05:26 chris i didnt write the Auth_with_ldap.pm module 05:26 chris but ill go look 05:28 chris yep it is 05:28 mohamedimran82 why this is necessary is that when we take reports we see lots of NULL fields because values are not pulled up from LDAP 05:28 chris yes you will need to edit it 05:29 chris from my %borrower; 05:29 chris on 05:29 chris to match you LDAP structure 05:36 mohamedimran82 i am working on it 05:40 mohamedimran82 my ldap structure says that the display name is gecos 05:41 mohamedimran82 $borrower{firstname} = $memberhash{gecos}; # MANDATORY FIELD 05:41 mohamedimran82 i did this. Is this right? 05:41 chris i dont know sorry 05:42 chris i havent done any work at all on LDAP and Koha 05:43 mohamedimran82 anyways thnx a ton for ur comments on reports 05:44 chris no problem 05:44 chris ok, bedtime for me 05:44 chris good luck :) 05:44 mohamedimran82 sweet dreams 08:04 hdl toins: bonjour 08:04 toins salut hdl 08:05 toins hdl, comment ca va ? 09:23 thd kados: are you there? 09:23 kados thd: yes 09:24 thd kados: I just you a revised adaptation of the koha.org CSS for the wiki. 09:25 thd kados: I am anxious to have a better CSS for the wiki so that I can have confidence in the presentation of wiki content. 09:27 kados ok, I'll try to take care of that asap 09:29 thd kados: i will be much happier about creating wiki content if is more readable and also looks better. 09:54 owen Hi rychi 09:55 rychi hi owen. I was just looking for you. 09:55 rychi I have a question re: npl templates 09:55 owen Shoot 09:55 rychi the title and other links on opac-searchresults is url-encoded. 09:56 rychi this gives me %20's in my searchresults. 09:56 rychi how should this work? 09:56 owen Are you looking at the latest CVS? 09:56 rychi yes 09:57 rychi rel_2_2 09:57 owen There was recently a change in the default template in that regard, which I copied over 09:57 owen I'll check to see what the explanation was 10:00 owen thd: regarding the wiki, did you try applying a user stylesheet, or turning CSS off? 10:01 thd owen: yes, I tested with a user stylesheet 10:03 thd owen: I could not see what I was doing earlier well because I was not using the best menu option for the Firefox Web Developer plugin. 10:05 owen rychi: you're talking about the ESCAPE="URL" 10:05 owen ? 10:05 thd owen: certainly a user stylesheet would fix the wiki for me, however, I want to fix the wiki for anyone I would direct to it. 10:06 rychi owen, right. 10:06 owen thd: I think most people are fine with it 10:06 owen Yeah, rychi, it's the same in the default template. What's the problem it's causing for you? 10:06 thd owen: I realise that but that is most people do not put much content into it. 10:07 rychi owen: i think the escape="url" should be in the <a> tag, but not in the link text? 10:07 rychi owen: the default templates work for me. 10:07 thd owen: if you start to put some significant content into the wiki, the problems become evident. 10:08 thd owen: existing content is simple outlines which are undemanding of the CSS. 10:10 rychi owen: ah, I see in the default tmpl, the escape param is only in the <a> tag. 10:11 thd owen: If you try adding paragraphs, bibliographies with hyperlinks, HTML content, etc. to wiki content you should see problems for readability and poor appearance. 10:11 owen I see, rychi. I'll change it.