IRC log for #koha, 2006-06-18

All times shown according to UTC.

Time S Nick Message
16:07 chris morning
16:47 rach morning
17:07 chris hi rach, hows things?
17:22 rach good thanks
17:22 rach excellent sleepy baby last nite, got through till 7am
22:52 chris anyone about?
02:31 osmoze hello
08:50 owen Good morning dewey
09:01 owen dewey: are you not talking to me?
09:01 dewey owen: i haven't a clue
09:38 kados owen: just posted a note on the npl tech staff about searching
09:38 kados owen: please check to make sure it's not too tech-heavy :-)
09:50 owen Sounds good to me. It just leaves the question hanging: why don't we have authority records and how do we get them?
09:50 kados yea
09:50 kados why don't we have them: because they are expensive
09:51 kados how do we get them:
09:51 kados
09:52 kados the second link there is the 'placing an order' link :-)
09:52 owen Not exactly is it?
09:52 kados heh
09:52 kados yea, somewhere around there is a list of the prices
09:54 kados last time I checked, it was something like $20K for author and subject authorities
09:54 kados it's subscription-based, so that'd be an annual cost
09:54 kados though thd tells me there are cheaper options
09:55 kados owen: just so you know, ID finished the CCL stuff, but when I tried to install it it seg faulted ... (that was this morning)
09:55 kados so I'm waiting to hear from them on what I need to do to run it
09:56 kados (probably some undeclared dependency or something
09:56 kados I think we're still on track for Monday
09:56 owen $20K annually is ridiculously high for a library our size
09:56 kados yep
09:56 kados but it might be possible if we could get 4 or 5 libraries together
09:57 owen Someone should file an anti-trust lawsuit
09:57 kados to spread out the expense
09:57 kados that would be interesting
09:57 kados I think the LOC has a directive to recover expenses for their 'products'
09:59 kados hmmm, can't find them :(
09:59 kados ha!
09:59 kados here we go:
09:59 kados I'm the release manager for the Koha project, an open-source library
09:59 kados automation system. I was wondering whether you could point me in the  
09:59 kados direction of any public domain database that contains cataloging
09:59 kados data that would be useful in the context of a cataloging program's
09:59 kados 'help' feature. Specifically, descriptions of tags and subfields in
09:59 kados MARC, style rules (AACR2, etc.). If possible, I'd like to wrap any
10:00 kados such public domain text into Koha to improve the references within
10:00 kados its interface. Any help you can provide is most appreciated.
10:00 kados here's the response:
10:00 kados --
10:00 kados It's worth noting that none of the information you are looking for is open
10:00 kados +source, per se.  Some of it is available largely without restrictions via the
10:00 kados +web, but all of it is copyrighted and comes with some at least minimal
10:00 kados +acknowledgement requirements.
10:00 kados  A condensed version of the MARC documentation is available at <
10:00 kados + >.  AACR2 is copyrighted by the American Library
10:00 kados +Association and their two co-publishers and is not available anywhere open
10:00 kados +source.  The only place it is available electronically is in Cataloger's
10:00 kados +Desktop, which as you no doubt know is a subscription publication.  Additional
10:00 kados +information about AACR2 is available at <>.  Additional
10:00 kados +information about Cataloger's Desktop is available at
10:00 kados +<>.
10:00 kados which was confusing to me
10:01 kados so then I wrote:
10:01 kados Thanks for the speedy reply. I've considered purchasing a subscription
10:01 kados for Cataloger's Desktop for LibLime's research staff -- when I look
10:01 kados at the terms of use it seems to be pretty restricted as far as using
10:01 kados the electronic text for anything other than reference within the
10:01 kados program (ie, it doesn't look like they would want me to bulk export
10:01 kados the AACR2 and put it into Koha). I do remember vaguely that material
10:01 kados produced by LOC is in fact public domain and use of it in the manner
10:01 kados I've described would be legal, but I'm not a lawyer :-). I'm wondering
10:01 kados whether you would be the one to talk to about this.
10:01 kados and I got the response:
10:01 kados --
10:01 kados   Thanks very much.  A common misconception is that Library of Congress data is
10:01 kados +in the public domain.  It is not.  It is copyrighted, but that affords LC no
10:01 kados +real protection within the United States.  Without the contractual protections
10:01 kados +we would not be able to keep the cost of the product as low as we do.  (And I
10:01 kados +should point out that we do what we do not with taxpayer money, but rather with
10:01 kados +revenue we generate from the sale of products and services to libraries
10:01 kados +worldwide.)  And of course, the American Library Association is VERY emphatic
10:01 kados +about receiving substantial royalties for use of their data.
10:02 kados so that thread doesn't directly apply to authorities
10:02 kados but it does indicate that at least some of what LOC does isn't sponsored by taxpayer monies
10:02 kados maybe authorities are part of that
10:04 owen By the way, did you ever find anything out about the bugs in addbiblio?
10:04 owen that should be a blocker for the next release
10:05 kados that's on my todo for today
10:45 kados so ...
10:45 kados as of right now
10:45 kados rel_2_2, using the npl templates
10:46 kados and marc21
10:46 kados won't save records
10:46 kados you add a record and it simply vanishes
10:46 kados ARRRRGGGG!
10:48 kados same thing with default
10:48 kados the valies in the record simply vanish
10:50 kados it's times like this when I wonder if open source is the right choice :-)
10:50 kados so the question is, how many hours will it take me to figure it out?
10:59 kados wow
10:59 kados paul just added a bunch of stuff to MARChtml2xml
10:59 kados that seems to be the problem
10:59 kados mine has:
10:59 kados sub MARChtml2xml {
10:59 kados        my ($tags,$subfields,$values,$indicator,$ind_tag) = @_;
11:00 kados pau's has:
11:00 kados sub MARChtml2xml {
11:00 kados        my ($tags,$subfields,$values,$firs​tsubfields,$indicator,$ind_tag) = @_;  
11:00 kados so what the heck is 'firstsubfields'?
11:01 kados rfirstsubfields : a reference to a list, each entry containing 0 or 1. 0 means that the corresponding subfield it NOT the 1st of the tag, 1 meaning it is.
11:01 kados that's a hack if I've ever seen one
11:05 kados it looks to me like paul committed this 'firstsubfield' thing into the script
11:05 kados but not into the template :(
11:05 kados owen: do you see any template related commits related to 'firstsubfield'?
11:06 kados here are some places:
11:06 kados grep -R firstsubfield *
11:06 kados default/en/acqui.simple/addbiblio.tmpl:                <input type="hidden" name="firstsubfield" value="<!-- TMPL_VAR NAME="firstsubfield" -->" size="2" maxlength="1"/>npl/en/acqui.simple/addbiblio.tmpl:                <input type="hidden" name="firstsubfield" value="<!-- TMPL_VAR NAME="firstsubfield" -->" size="2" maxlength="1" />
11:06 kados so they are in the templates
11:06 kados but the feature is broken in the script
11:08 kados when cloning a tag, there was a problem to detect the tag change when rebuilding the XML/MARC record :
11:08 kados        the test was done on an empty field.
11:08 kados        with the new CloneTag sub, this empty field was quite tricky to create on the fly.
11:08 kados        so i've added on each subfield a <input name="firstsubfield"> saying wether the subfield is the 1st or not.
11:08 kados        Thus, when a tag is cloned, it is cloned properly.
11:08 kados        HTML2marc and well as HTML2xml have been modified, but PLEASE DOUBLE CHECK my work
11:19 kados ok ...
11:19 kados I think I've traced the root of the problem
11:19 kados testing now
11:19 kados owen: what's up with this display:[…]
11:20 kados owen: is that due to the record or is there some kind of weirdness in the template?
11:21 kados also it looks like the 'economical' vs 'standard' distinction when viewing MARC was either lost in the intranet, or it was never committed
11:21 Anson aloha
11:21 kados hey Anson
11:21 Anson Hi Kados :)
11:21 Anson I'm working on  a centos install of koha
11:21 Anson getting stuck at the yaz install
11:22 Anson is needed by libyaz-2.1.20-1.i386
11:22 Anson is needed by libyaz-2.1.20-1.i386
11:22 Anson is needed by libyaz-2.1.20-1.i386
11:22 Anson is needed by libyaz-2.1.20-1.i386
11:22 Anson is needed by libyaz-2.1.20-1.i386
11:23 Anson probably misusing the rpm commands, or need to add repositories, any suggestions?
11:26 kados hmmm
11:27 kados what about:
11:27 kados yum install yaz
11:27 kados (does centos use yum yet?)
11:27 kados honestly, unless you're deeply tied to using centos, I'd recomment using debian
11:27 kados debian sarge is what index data releases are based on
11:28 kados so you can have the latest stable stuff running in a package management system
11:28 kados it's also what most of the 'supported' koha libraries use
11:28 Anson ok
11:29 kados Anson: plus, I've written up such a nice install for debian :-)
11:29 Anson well charting some new turf'
11:29 kados Anson:[…]Debian_sarge.html
11:30 Anson wish i had an extra box to put debian on :)
11:30 kados i've played around with quite a few distros
11:30 Anson that's ok, diggin through the data tonite + manana should turn in to a library soon enough :)
11:30 kados I've got production systems running fedora, RHEL, gentoo, debian, ubuntu, etc.
11:31 kados and what I've found is
11:31 kados whatever app you're running is best run in the same environment as the developers who are working on that app
11:31 kados it saves you hours and hours of headaches
11:32 kados for koha and yaz that environment is mostly debian
11:32 kados not to say that all the developers use debian
11:32 kados but most of us do
11:32 kados and the index data guys (who maintain yaz) definitely do
11:32 owen kados: not sure on first glance what's going on with that marcdetail page
11:32 kados owen: could it have something to do wtih pauls' new 'one tab' change?
11:33 kados owen: I'll switch to default so we can take a look
11:33 owen Oh, I know--their custom stylesheet needs to be updated
11:33 kados ahh
11:33 kados ok, I'll let you do that then ;-)
11:33 Anson Kados: that makes good sense... which version of debian installs most readily?
11:33 kados owen: or let me know what needs changed
11:34 kados Anson: I'd stick with the latest 'stable' which is 'sarge'
11:34 kados Anson: do a 'netinst'
11:34 kados Anson: or use a 'business card install' off the network
11:34 kados Anson: you basically install a minimal system
11:34 kados Anson: some of this is covered in my installation document
11:34 kados Anson: it's worth a read
11:35 Anson I've read most of it, seems like RHEL is missing a lot of the dependencies....
11:35 kados yep
11:35 kados you can find them
11:35 kados but it's a pain
11:35 Anson :)
11:35 Anson yep
11:36 kados you could be up and running on a sarge system in a day
11:36 kados rather than spend a week getting RHEL running :-)
11:36 kados of course ...
11:36 kados the other alternative is to purchase support from one of the koha vendors :-)
11:36 Anson it wouldn't even take that long, however this could be my miniscule contribution to a very cool project :)
11:36 kados yea, true that
11:37 Anson yup - talked to liblime yesterday, very cool crew
11:37 kados that was me :-)
11:37 Anson :)
11:37 Anson excellent
11:38 Anson whatta trip
11:39 Anson well I think I'm gonna work on finding the rpm servers for the dependencies, it'll be worth it if the documentation gets done and put in the wiki
11:39 Anson besides, i don't really have an extra dev box sitting around right now...
11:40 Anson rock on, ttyl8r :)

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