← Previous day | Today | Next day → | Search | Index
All times shown according to UTC.
| Time | Nick | Message |
|---|---|---|
| 12:09 | kados | hi |
| 12:11 | thd: so now you've mastered the command-line utilities :-) | |
| 12:11 | thd: the sky is the limit :-) | |
| 12:11 | thd | kados: yes the world is much faster on the command line and I did my old job on the command line |
| 12:12 | kados: I would not even think of using an MS windows system on anything other than the command line either but I have not done that for some yers | |
| 12:13 | kados | here is your email: |
| 12:13 | "the record editor was removing the | |
| 12:13 | 'a' from leader position 09. That is an easily reproducible if you simply | |
| 12:13 | duplicate a repeatable field. The 'a' will be lost from 000/09 once the | |
| 12:13 | thd | s/yrs/years/ |
| 12:13 | kados | record is reloaded. " |
| 12:13 | I suspect that this is now fixed since duplicated tags nolonger require a page refresh | |
| 12:14 | thd | kados: well that avoids the worst problem when everything was working that you lost your place in the record if you duplicated a tag because there ws no anchor follolwed for the refresh |
| 12:15 | kados | right |
| 12:15 | thd: I just tested duplicating a tag | |
| 12:15 | thd: seems to be working well | |
| 12:15 | owen-away: you around? | |
| 12:17 | thd | kados: the second worst problem when everything was working was the nuissance of pushing subfields around with arrows instead of inserting them at aparticular point from a selection list or having them in the coreect order by default |
| 12:17 | s/coreect/correct/ | |
| 12:19 | kados: there is plenty of time to fix that in future along with adding fields and subfields which are not set to appear by default | |
| 12:22 | kados: do you have any more clever ideas about how to restore my X-windows server so that I can be a perfect speller and test these bugs for myself again. | |
| 12:22 | ? | |
| 12:23 | kados: otherwise I will be reporting the editor as riddled with bugs again because nothing works in lynx. | |
| 12:25 | kados: am I still connected? | |
| 12:26 | kados | thd: yes |
| 12:26 | thd: sorry, had a phone call | |
| 12:26 | thd: can you paste in the error you're getting when you run 'apt-get -f install'? | |
| 12:27 | thd: to paste in screen you go: | |
| 12:27 | C-a C[ | |
| 12:27 | well ... that's to copy | |
| 12:27 | to paste you go: | |
| 12:27 | C-a C-] | |
| 12:27 | (where 'C-' represents the 'control key' | |
| 12:27 | ) | |
| 12:27 | :0( | |
| 12:36 | thd: did it work? | |
| 12:47 | thd | kados: there is something fundamental that I am missing about using screen |
| 12:48 | kados | hehe |
| 12:48 | perhaps | |
| 12:48 | one thing to remember | |
| 12:48 | you can always bail yourself out with C-a " | |
| 12:49 | to pull up a menu | |
| 12:49 | and try not to confuse yourself by creating threaded screens :-) | |
| 12:49 | you can re-name a given terminal with C-a A | |
| 12:50 | I use that to help organize the menu that comes up with C-a " | |
| 12:50 | thd | kados: after I leave a screen terminal with C-a d how do I start a new process in such a way that I can have the new process under the control of the same screen process? |
| 12:50 | kados | thd: think of it as tabbed browsing |
| 12:50 | thd: when you start screen you have one tab | |
| 12:50 | thd: you can create a new tab with C-a c | |
| 12:51 | thd: you can get a list of tabs with C-a " | |
| 12:51 | thd: does that make better sense? | |
| 12:51 | thd | kados: yes that is the problem, I am always opening a new window instead of a new tab or at least think that I am |
| 12:51 | kados | right |
| 12:51 | each 'tab' is it's own terminal | |
| 12:51 | thus, screen is called a 'terminal multiplexor' :-) | |
| 12:51 | thd | C-a c is the answer? |
| 12:51 | kados | yes, that creates a new tab |
| 12:52 | you can rename the tab with C-a A | |
| 12:52 | of course, they don't really appear as tabs | |
| 12:52 | (though you can set them up to, but that's an advanced topic) | |
| 12:53 | thd | kados: do I invoke C-a c after C-a d or directly from the application that I am already using? |
| 12:53 | kados | thd: you only use C-a d when you are done for the day |
| 12:53 | thd: at the beginning of the day (as early as 4am), you begin work ... log in and issue: | |
| 12:54 | screen -r | |
| 12:54 | now you're in your screen session | |
| 12:54 | if you need a new tab type | |
| 12:54 | thd | kados: that has been my problem I tried to use it before what I had hoped was another tab |
| 12:54 | kados | C-a c |
| 12:54 | right, that won't work | |
| 12:54 | it will disconnect you from the screen session | |
| 12:54 | which is not what you want | |
| 12:55 | thd: you only want to run one screen session at a time | |
| 12:55 | thd: if you type 'screen -list' it will report how many you're running | |
| 12:56 | thd | kados: I realise that I only want one because the advantages of having every application in a separate screen session are small. |
| 12:57 | kados | there are definitely times when you want more than one, but not when you've started learning screen :-) |
| 13:04 | thd | kados: after pressing enter for the default options to configure xserver-xorg I have the following lines |
| 13:04 | (Reading database ... 170488 files and directories currently installed.) | |
| 13:04 | Preparing to replace x11-common 6.8.2.dfsg.1-11 (using .../x11-common_1%3a7.0.22_i386.deb) ... | |
| 13:04 | Unpacking replacement x11-common ... | |
| 13:04 | dpkg: error processing /var/cache/apt/archives/x11-common_1%3a7.0.22_i386.deb (--unpack): trying to overwrite `/usr/X11R6/bin', which is also in package navigator-smotif-477 | |
| 13:05 | Errors were encountered while processing: /var/cache/apt/archives/x11-common_1%3a7.0.22_i386.deb | |
| 13:05 | E: Sub-process /usr/bin/dpkg returned an error code (1) | |
| 13:05 | kados | ok |
| 13:06 | thd | kados: is this my penalty for trying to have navigator-smotif-477 ? |
| 13:06 | kados | thd: to fix this, you need to go: |
| 13:06 | apt-get remove --purge navigator-smotif-477 | |
| 13:07 | thd: you can re-install it later | |
| 13:07 | thd: but first we need to resolve all these conflicts | |
| 13:07 | thd | kados: yes that is what I thought |
| 13:07 | owen | kados: I'm back. You rang? |
| 13:08 | kados | owen: yea ... I was gonna talk briefly about the MARC editor |
| 13:08 | owen: I have some ideas for imporving the user's experience in the npl templates | |
| 13:08 | owen: one of the things that the catalogers told me was that they want 'everything on the page' with no scrolling | |
| 13:09 | owen: which seems reasonable to me :-) | |
| 13:09 | owen: also, i think we need a few new hotkeys for rapid switching between sections | |
| 13:09 | (thanks) | |
| 13:10 | for instance, it should be possible to duplicate a subfield with a simple keystroke combination | |
| 13:10 | owen | Getting into some javascript territory I'm not familiar with, but we can work on it |
| 13:10 | kados | same goes for re-ordering the subfields |
| 13:11 | i think the scrolling bit might be the most challanging | |
| 13:11 | owen | The first snag that comes to mind is how you handle context--how does the system know which field you want to duplicate? |
| 13:12 | How does it know where you are on the page? | |
| 13:12 | kados | thd: it can find where the cursor is |
| 13:12 | toops | |
| 13:12 | owen: it can find where the cursor is currently positioned | |
| 13:12 | owen | Anyway, we can lay out the feature list and see what we can do with it |
| 13:12 | kados | owen: and isn't there an 'e' in js that has some kind of sense of context? |
| 13:13 | if you pass 'e' along it passes some environment vars? | |
| 13:13 | or something :-) | |
| 13:13 | owen | an 'e' ? That's new to me. |
| 13:13 | kados | maybe not |
| 13:14 | owen: maybe, just for the sake of layout, we can remove the left-hand navbar from that screen | |
| 13:14 | owen: that would free up some room | |
| 13:15 | owen: also, quite a lot of the top of the page is currently occupied by the 'Edit MARC Record with Framework' line | |
| 13:15 | owen: i wonder if that could be made teeny-tiny and scooted up to the upper-right hand side of the page parallel to the 'Xataloging Home | Add MARC, etc' links | |
| 13:16 | owen | Certainly |
| 13:17 | I've wished for ages that we could display the actual title of the record being edited instead of just the 'edit marc record' heading | |
| 13:17 | kados | right |
| 13:18 | that should be possible | |
| 13:18 | lemme check the script | |
| 13:20 | unfortunately, it would reqire some additional coding | |
| 13:20 | so I'm not sure paul would be thrilled about having that in rel_2_2 | |
| 13:21 | owen: we seem to have lost the 'economical' vs 'standard' display in the MARC view | |
| 13:22 | owen: http://opac.smfpl.org/cgi-bin/[…]ail.pl?bib=141126 | |
| 13:22 | owen: there are actually two 650s there | |
| 13:22 | owen: LabelMARCView is the name of the syspref | |
| 13:24 | this must be due to paul's new changes :( | |
| 13:24 | owen | What do you mean we lost it? |
| 13:27 | kados | well ... |
| 13:27 | it used to be that if LabelMARCView was set to 'standard' | |
| 13:27 | you'd have: | |
| 13:27 | 650 $asubject $bsubfieldb | |
| 13:27 | 650 $asubject 2 $bsubfieldb 2 | |
| 13:27 | and if it was set to 'economical' you'd have: | |
| 13:28 | 650 $asubject $bsubfieldb $asubject 2 $bsubfieldb 2 | |
| 13:28 | but on SMFPL it's set to 'standard' but displaying as 'economical' :( | |
| 13:28 | does that make sense? | |
| 13:29 | owen | Yes |
| 13:32 | kados | wait a second |
| 13:32 | it's worse than I feared | |
| 13:32 | that's how things are being _saved_ now !!! | |
| 13:32 | holy #$%& | |
| 13:33 | thd | kados: I upgrading now |
| 13:33 | kados | thd: congratulations |
| 13:33 | thd: i found another bug in the MARC editor | |
| 13:33 | thd | kados: how do i scroll bak this IRC client? |
| 13:33 | s/bak/back/ | |
| 13:33 | kados | thd: now, repeated fields aren't saved as repeated fields :( |
| 13:34 | thd: I don't scroll back, I use logs | |
| 13:34 | thd | kados: how do I activate the logs or are they on by default? |
| 13:35 | kados | thd: you have to set them up, it's a complicated process that I've only done once :-) |
| 13:35 | thd: so I can't recall how exactly it works :) | |
| 13:35 | thd: man irssi may be of some help | |
| 13:35 | thd | :) |
| 13:36 | owen | http://koha.org/cgi-bin/logs.p[…]as=&search=Search |
| 13:37 | Man, I'm not keeping up with template updates like I thought I was | |
| 13:38 | kados | owen: I wonder if that is related to my problem |
| 13:42 | owen | That fixes the problem with the subjects display |
| 13:43 | I don't think the LabelMarcView pref affects the OPAC | |
| 13:47 | kados | owen: what fixes the problem? |
| 13:47 | owen: a change you just found? | |
| 13:48 | owen: an the labelmarcview should affect the OPAC as well | |
| 13:48 | owen: that is, it's supposed to | |
| 13:48 | this problem goes deeper than just display though | |
| 13:49 | paul's changes have broken the ability for the system to save repeated fields seperately | |
| 13:54 | thd | kados: how are repeated fields saved if not separately? |
| 13:54 | kados | thd: they are saved under the same field |
| 13:55 | thd: so 650$a$b$c 650$a$b$c becomes 650$a$a$b$b$c$c!! | |
| 13:55 | owen | I just committed a small change to opac-MARCdetail.tmpl that prevents the display from running together like you pointed out |
| 13:56 | kados | owen: thanks |
| 13:56 | owen | But that doesn't have anything to do with the problem you're talking about now |
| 13:56 | kados | owen: right |
| 13:56 | thd | kados: was that the behaviour he was attempting to emulate as his user's have preferred to see MARC records displayed in that less verbose manner? :) |
| 13:57 | kados | thd: I certainly hope not |
| 13:58 | thd | kados: maybe it made the problem harder to identify if the MARC display did not distinguish between what was correct and what was incorrect on his choice of preferences. |
| 14:03 | kados | thd: good point |
| 14:04 | thd: is there a stray 100$9 right above 600$a in your MARC Bibliographic framework? | |
| 14:05 | I think we're also back to saving blank subfields again | |
| 14:05 | sheesh | |
| 14:05 | we've completely reverted | |
| 14:08 | also lost our 'tab between fields' feature | |
| 14:08 | thd | kados: I told you that there were bound to be mistakes in the bibligraphic framework that large |
| 14:09 | kados | thd: of course ... just wondering if it was intentional |
| 14:10 | thd | kados: that sound like most of the easy errors form yanking and pasting similar lines without correcting everything in the new location after that. |
| 14:10 | s/soud/seems/ | |
| 14:10 | s/sound/seems/ | |
| 14:11 | kados | wow |
| 14:11 | the back button is broken too | |
| 14:11 | such that if you save something | |
| 14:11 | then go back | |
| 14:11 | your repeated 650 jumps into the 651!! | |
| 14:11 | this is just nuts | |
| 14:12 | thd | kados: far too much JavaScript in the world breaks the basic navigation functions |
| 14:12 | of web browsers | |
| 14:12 | kados | yep |
| 14:20 | owen | How nice to have my laptop reboot itself without warning. |
| 14:23 | kados | heh |
| 14:25 | owen | I see a reference to LabelMARCView in opac-MARCdetail.pl, but when I turn the preference on and off I don't see any difference in either the npl or css template |
| 14:27 | kados | owen: it's not an on/off preference |
| 14:27 | owen: it should have two values: standard | economical | |
| 14:27 | owen | I just mean when I switch the two options |
| 14:27 | kados | ahh |
| 14:27 | working with rel_2_2 and NPL's data? | |
| 14:27 | owen | Yes |
| 14:28 | kados | on 101? |
| 14:28 | owen | Yes |
| 14:28 | kados | and you've got a record that has repeated fields? |
| 14:28 | ie, two 650s or something? | |
| 14:29 | owen | Sorry--I was thinking I was looking at something else. |
| 14:29 | I was meaning to be testing hide_marc | |
| 14:29 | kados | ahh |
| 14:29 | right | |
| 14:29 | well ... there's another one too | |
| 14:30 | there used to be a way to turn off the labels altogether | |
| 14:31 | yea, here it is: advancedMARCEditor | |
| 14:31 | If set, the MARC editor won't show you tag/subfields description | |
| 14:34 | owen | http://66.213.78.101:8082/cgi-[…]mple/addbiblio.pl |
| 14:34 | Just starting to play around with the layout | |
| 14:35 | kados: when you're using screen, how do you manage editing more than one file at once? Do you open new sessions for each? | |
| 14:35 | kados | yep |
| 14:36 | owen | New screen sessions? Doesn't that mean you might be logged into one machine 15 times? |
| 14:36 | kados | wait |
| 14:36 | no, not new screen _sessions_ | |
| 14:36 | I just open a new terminal within the screen session | |
| 14:36 | C-a c | |
| 14:36 | owen | Yeah, I don't mean sessions |
| 14:36 | kados | right |
| 14:36 | well ... | |
| 14:37 | in some cases | |
| 14:37 | like NPL | |
| 14:37 | I manage several boxes on the same network | |
| 14:37 | and for those instances, I practice threaded screening :-) | |
| 14:37 | owen | I just wondered if there wasn't something I didn't know about vi having multiple buffers or something |
| 14:37 | kados | so my first and primary screen is on a liblime box |
| 14:37 | using that one, I log into an NPL box | |
| 14:38 | and start an instance of screen on that NPL box | |
| 14:38 | to control that screen session I use C-a a X (where X is the command) | |
| 14:39 | I'm pretty sure you can mess with vi buffers too | |
| 14:39 | I've just never had time to figure out how :-) | |
| 14:39 | let me know if you figure it out :-) | |
| 14:44 | in vi you may need to type: | |
| 14:44 | :syntax on | |
| 14:44 | owen | Swank |
| 14:45 | kados | you can save that somehow |
| 14:45 | owen | :) |
| 14:45 | kados | :save or something |
| 14:45 | or edit your vimrc file | |
| 14:45 | thd | colour syntax highlighting is a feature of vim not vi |
| 14:51 | kados | vi is a symlink to vim |
| 14:51 | so you're covered :-) | |
| 14:52 | owen | Nice to know what I'm really using |
| 15:08 | kados | all that mousing :-) |
| 15:09 | heck, I only have one mouse button now :-) | |
| 15:09 | hehe | |
| 15:09 | apple-click does the same thing | |
| 15:10 | and it's right where your thumb is | |
| 15:10 | so not really hard to use | |
| 15:10 | course, you can still use a multi-button external mouse | |
| 15:10 | I've got a 5 button somewhere around here ... but never got around to getting proficient at it | |
| 15:20 | owen: http://browsers.evolt.org/?ie/32bit/standalone | |
| 15:20 | owen: standalone versions of all the important IEs | |
| 15:20 | owen: in case you're interested :-) | |
| 15:20 | owen | Yeah, I've seen that. I'm so lax about testing for older versions of IE, though. |
| 15:21 | kados | hehe |
| 15:21 | I just got a great book | |
| 15:21 | called 'Web Design in a Nutshell' | |
| 15:21 | third edition | |
| 15:21 | it's got a fantastic overview of nearly everything related to web design | |
| 15:21 | and it's published 2006, so very up to date | |
| 15:22 | owen | I've heard of it, but I've never looked at it before |
| 15:22 | kados | you should try to get your hands on it, through MORE or something, it's a good read |
| 15:23 | cool ... I'll put that on my list | |
| 15:23 | if we were going to stick with the current MARC editor, I'd want to do it using the DOM better | |
| 15:23 | owen | It's a fast read, but it really helped me understand standards-compliant scripting better |
| 15:24 | kados | neat |
| 15:30 | owen | kados: how do you see access keys working for navigation in the MARC editor? |
| 15:34 | kados | well ... |
| 15:34 | I'll have to give it some additional thought | |
| 15:34 | but off the top of my head | |
| 15:35 | F1 should provide context-sensitive help | |
| 15:35 | and there should be some way to navigate from field to field and from tab to tab without using the mouse | |
| 15:35 | I started using the tabindex property a bit | |
| 15:36 | but some of that is broken in the latest rel_2_2 | |
| 15:41 | owen | I suppose you could have a 'next tag' hotkey... the tab key works for next subfield |
| 15:41 | It's hard to manage tabindex when so much of the markup is generated by the script | |
| 17:04 | kados | thd: got a question |
| 17:04 | thd: did you mention at one point that there was a better way tto test if something was a fixed field than >10 ? | |
| 17:12 | thd | kados: yes I committed an example in bulkmarcimport.pl |
| 17:12 | kados | thd: could you briefly tell me what it is? |
| 17:26 | thd | kados: if ($fieldNameOrNumber =~ m/^00/) { # you have a fixed field } |
| 17:27 | kados | right ... thanks |
| 17:28 | thd | kados: this works for fields with nonnumeric field names. |
| 17:28 | kados | yep |
| 17:28 | I"m re-writing MARChtml2xml :( | |
| 17:29 | thd | kados: =~ m/^00/ could be sustituted for < 10 throughout Koha with no problem. |
| 17:30 | kados | agreed |
| 17:30 | don't actually need the m though | |
| 17:30 | =~ /^00/ works fine | |
| 17:31 | thd | kados: I change regex systems so often I cannot keep the syntax straight too much of the time between Perl and whatever else :) |
| 17:41 | kados: check the code for fMARC8ToUTF8() in bulkmarcimport.pl in rel_2_2. That code is not perfect but it does work much better than the MARC::File::XML code if the record is MARC 8 and needs conversion to UTF-8. | |
| 09:43 | thd_ | kados: are you there? |
| 10:08 | kados | thd: yes, I"m here |
| 10:11 | thd: here is an example: | |
| 10:11 | 852 1 _aMOUN | |
| 10:11 | _bMOUN | |
| 10:11 | _h398.8 | |
| 10:11 | _iA | |
| 10:11 | _p10002 | |
| 10:11 | _9p14.44usd | |
| 10:11 | 940 _bK-3 | |
| 10:11 | _d05/15/06 | |
| 10:11 | _sSchool Library Journal starred, September 2005 | |
| 10:12 | _sBook Links (A.L.A.), May 2006 | |
| 10:12 | _sPublisher's Weekly, August 2002 | |
| 10:12 | _vFLR | |
| 10:12 | 942 _aMOUN | |
| 10:12 | thd: some vendor uses the above for holdings and local fields :-) | |
| 10:19 | thd: can you explain to me how the call number is supposed to work? | |
| 10:19 | thd: I've never quite understood call numbers | |
| 10:19 | thd: in the 852 scheme above | |
| 10:19 | thd: it looks like $h in the above is the complete call number | |
| 10:20 | thd: I'm not looking for comprehensive coverage ... just the basics :-) | |
| 10:21 | thd: I'm not sure the above vendor uses _standard_ 852 | |
| 10:21 | thd: maybe you can tell by looking at it | |
| 10:21 | thd | kados: so in that example the call number is 398.8 A |
| 10:22 | kados | _a and _b are the 'holdingbranch' and 'homebranch', h and i are the dewey call number, _p looks like the barcode and _9 looks like the price |
| 10:23 | thd | kados: $9 is obviously special and obviously used for the price |
| 10:24 | kados: you can check how I mapped them for the framework where I named the mapping explicitly in the extra verbose librarian label for 952 | |
| 10:24 | kados: I give the key based on standard fields and subfields | |
| 10:25 | kados | I propose this scheme: |
| 10:25 | $a => holdingbranch | |
| 10:25 | $a => homebranch | |
| 10:25 | $h => dewey | |
| 10:25 | $i => subclass | |
| 10:25 | $p => barcode | |
| 10:25 | $9 => price | |
| 10:25 | $9 => replacementprice | |
| 10:25 | the main question being, will Koha handle the dewey and subclass correctly | |
| 10:26 | thd | kados: you have $9 used for two different things. |
| 10:26 | kados | yep |
| 10:26 | they are the same on import in this case | |
| 10:26 | it's a brand new library | |
| 10:27 | thd | kados: well then you will not have anything worse than Follet but the library will not always be new |
| 10:28 | kados | right |
| 10:28 | can you think of a solution? | |
| 10:28 | also, can you identify what some of the fields in the 940 are supposed to be? | |
| 10:29 | $s looks like a series title | |
| 10:29 | thd | kados: add more subfields to the framework for 952 and hope that LC never changes 852. |
| 10:29 | kados | $a looks like the homebranch again |
| 10:29 | thd: ? | |
| 10:29 | thd: i don't understand why I need more subfields | |
| 10:29 | thd | s/952/852/ |
| 10:30 | kados: well you certaily need one for items.itemnumber | |
| 10:31 | s/certaily/certainly/ | |
| 10:31 | kados | right |
| 10:32 | thd | kados: having two uses of $9 that will become incompatible seems like something that may become a problem in future |
| 10:33 | kados | yes, I'll remove one of them |
| 10:33 | thd | kados: why would you remove rather than add? |
| 10:34 | kados | thd: the data only has one price field |
| 10:34 | thd: understand that this library will not be editing any records | |
| 10:34 | thd: they come fully-formed from the head of Follett :-) | |
| 10:34 | thd | kados: if they never edit their records in future then they have no problem |
| 10:35 | kados: I see now | |
| 10:35 | kados | my proposal is to create a generalized framework for Follet records |
| 10:35 | that will work for all follet libraries :-) | |
| 10:35 | or libraries that get their records from follett :-) | |
| 10:35 | thd | kados: you should consider the other call number fields that may be used |
| 10:36 | kados | which others? |
| 10:36 | (others that Follett may be using?) | |
| 10:36 | thd | kados: prefix sometimes and suffix hardly ever. |
| 10:37 | kados: $j $h $i $m if I remember | |
| 10:37 | kados | I don't think we have fields in Koha to handle those, right? |
| 10:39 | thd | kados: Koha treats the call number as one field except when the call number sorting code is intended to break DDC class numbers 852 $h at the decimal point to sort the display. |
| 10:40 | kados | I didn't realize we had sorting that was that advanced |
| 10:41 | thd | kados: chris wrote it or at least designed it |
| 10:41 | kados | I wonder if it works |
| 10:41 | I wonder if I added items.prefix and items.suffix whether it would 'just work'? :-) | |
| 10:41 | thd | kados: maybe it does not do anything but he explained at least the intention |
| 10:43 | kados: you would also need to add class number for $h and item or cutter number for $i | |
| 10:44 | kados: then you would need to be certain that all the templates had access to the different parts | |
| 10:45 | kados: From what I have seen you have much experience making additional elements available to the templates. | |
| 10:47 | kados | yes, that seems quite simple |
| 10:47 | so what's this about 'class number' and 'item or cutter number' for $i? | |
| 10:48 | thd: I propose four new columns in items: | |
| 10:48 | dcc_prefix | |
| 10:49 | dcc_class_number | |
| 10:49 | dcc_cutter_number | |
| 10:49 | dcc_suffix | |
| 10:49 | thd: would that be good enough? | |
| 10:49 | thd: is there a resource I can read to learn what the practice of DCC is in libraries? | |
| 10:50 | thd | kados: so 050 $a, $055 $a, 080 $a, $082 $a, 084 $a, 090 $a, 092 $a, etc. are all class numbers. |
| 10:50 | kados | DCC Class numbers? |
| 10:51 | I though 082 was LCC | |
| 10:51 | what's a class number? ;-) | |
| 10:51 | thd | kados: 082 is for DDC (not DCC) |
| 10:52 | kados: as is 092 if I remember | |
| 10:52 | kados: 055 sometimes in Canada | |
| 10:53 | kados: 050 and 090 are for LCC with 055 sometimes in Canada. | |
| 10:54 | kados: the class number is the base number that assigns the material to a place in the classification scheme | |
| 10:54 | kados | is it always subject-based? |
| 10:55 | thd | kados: it is classification base which is not entirely the same concept as subject but a closely related concept |
| 10:56 | kados | ah |
| 10:56 | thd | kados: both classification and subject are concerned with the what the content is about. |
| 10:57 | kados | here's a definition of a "classification scheme" according to itsmarc: |
| 10:57 | A logical system for the arrangement of knowledge. A fully developed classification scheme specifies categories of knowledge and provides the means to relate the categories to each other and to specify in the classification number all or the most important of the aspects and facets of a subject. | |
| 10:58 | thd | kados: classification schemes tend to be much much more hierarchical than subject schemes. Subject thesauri are often more hierarchical than most people imagine but they can be completely flat if they are designed that way. |
| 10:58 | kados | http://en.wikipedia.org/wiki/L[…]ry_classification |
| 10:58 | wikipedia looks like a good reference for this | |
| 10:59 | thd | kados: a fully facetted subject thesaurus would seem to not be heirarchical. |
| 11:00 | kados: then the problem is that you still want a unique call number for all the material that you have in the collection. | |
| 11:01 | kados: so 050 - 098 $b is for the item or cutter number. | |
| 11:02 | kados: the cutter number nothing to do with the content of the material or its place in a classification scheme. | |
| 11:04 | kados | it only has to do with where it is on the shelf? |
| 11:04 | thd | kados: The cutter number is mereley to give an assignement usually based on the author's last name according to a cutter table so that the class number in addition to the cutter number will give the material a unique place on the shelf |
| 11:06 | kados | thd: it looks like some records don't have cutters, is that normal? |
| 11:06 | wait ... I'm perhaps wrong | |
| 11:06 | i guess they all have cutters | |
| 11:06 | thd | kados: sometimes other information has to be added to a cutter number such as a date to distinguish different editions by the same author. |
| 11:08 | kados: prefix and suffix do not appear in 050 - 08X but can sometimes be in additional 09X subfields | |
| 11:10 | prefix commonly designates a special location such as JUV for the juvenile section or FIC for the fiction collection. | |
| 11:12 | kados | thd: for purposes of quickly implementing something, could I substitute subclass for cutter? |
| 11:13 | thd | kados: suffix is uncommon and used in special circumstances the most common of which may be something like 'index table' in the suffix for designating the special suffix location of something very specally within the REF prefix area. |
| 11:14 | kados: subclass was meant by chris to be the part of the decimal number in DDC after the decimal point. | |
| 11:14 | kados | ahh |
| 11:14 | that's unfortunate | |
| 11:15 | thd | kados: dewey was meant to be the part before the decimal number |
| 11:15 | kados | and itemcallnumber is supposed to be 852 $k $h $i $m combined |
| 11:15 | thd | kados: with one addition |
| 11:15 | kados | what an unfortunate design |
| 11:17 | thd | kados: that is very good designe so that you can change the treatment of some colection by only changing an individuated segment rather than having a regex fail on some with everything in one string. |
| 11:17 | s/colection/collection/ | |
| 11:18 | kados: there is an additonal consideration for the complete call number | |
| 11:19 | kados: after the $i on the book label where $t is > 1 would be the copy number when there is more than one copy number abbreviated as c. $t | |
| 11:21 | kados | ? |
| 11:21 | meaning when there is more than one $t? | |
| 11:22 | or just when $t is greater than 1? | |
| 11:22 | (is $t always a single-digit number?) | |
| 11:23 | thd | kados: can certainly have more than 9 copies so not necessarily single digit |
| 11:23 | kados | OK |
| 11:23 | hmmm | |
| 11:23 | so then ... | |
| 11:23 | itemcallnumber is supposed to be 852 $k $h $i $t $m combined | |
| 11:24 | if $t is > 1 | |
| 11:24 | otherwise ... | |
| 11:24 | thd | kados: copy number is usually not shown for the first copy |
| 11:24 | kados | itemcallnumber is supposed to be 852 $k $h $i $m combined |
| 11:24 | dewey | i already had it that way, kados. |
| 11:24 | kados | itemcallnumber is supposed to be 852 $k $h $i $t $m combined |
| 11:24 | hehe | |
| 11:24 | dewey is confused even :-) | |
| 11:24 | dewey? | |
| 11:24 | dewey | yes, kados? |
| 11:24 | kados | dewey: are you confused? |
| 11:24 | dewey | bugger all, i dunno, kados |
| 11:25 | kados | hehe |
| 11:25 | thd | :) |
| 11:25 | kados: I have code for all that which I sent you | |
| 11:26 | kados: well my code had some bugs and excluded $k and $m | |
| 11:26 | kados: also one bug used only one digit for the copy number | |
| 11:28 | kados | the code should be quite simple |
| 11:28 | provided I understand what must be done correctluy | |
| 11:28 | correctly even | |
| 11:28 | thd | kados: the code I sent was designed for filling 952 from 050 - 09X. |
| 11:28 | kados: yes, very simple code | |
| 11:29 | kados | there is one small issue remaining |
| 11:29 | in the data, I see: | |
| 11:29 | 050 00 _aPZ8.3 | |
| 11:29 | _b.A5567 2005 | |
| 11:29 | 082 04 _a398.8 | |
| 11:29 | _222 | |
| 11:29 | 082 00 _a[E] | |
| 11:29 | _222 | |
| 11:29 | then ... | |
| 11:29 | 852 1 _aMOUN | |
| 11:29 | _bMOUN | |
| 11:29 | _h398.8 | |
| 11:29 | _iA | |
| 11:29 | _p10002 | |
| 11:29 | _9p14.44usd | |
| 11:29 | thd | kados: you need spaces between elements to avoid muching them into an inditinguishable mass |
| 11:30 | what kind of DDC number is the repeated 0982? | |
| 11:31 | s/0982/082/ | |
| 11:31 | kados | is there any scheme by which we can map LCC and DDC to the Koha tables, without changing the database and withhout losing data? |
| 11:31 | I don't know what kind it is | |
| 11:31 | how can I tell? | |
| 11:33 | thd | kados: yes, my code only covered LCC but you just read the correct fields and read the indicators if they have been set correctly. |
| 11:34 | kados: my Z39.50 client has code for finding LCC, DDC, UDC, and NLM no matter where people try to hide the number. | |
| 11:34 | kados | heh |
| 11:35 | mainly, I'm interested in ensuring that all follett records will work out of the box | |
| 11:35 | thd | kados: In the real world indicators are often not set correctly |
| 11:36 | kados: at least too many small libraries or careless big ones will ignore indicators specifying the classification scheme | |
| 11:37 | kados: are you only concerned with DDC follet records? | |
| 11:44 | kados: you should use generalised names in the items table such as items.call_no_prefix instead of items.ddc_prefix as you had proposed. | |
| 11:46 | kados | only DDC follet records |
| 11:47 | what I don't understand | |
| 11:47 | thd | kados: the names of the columns should still be classification agnostic |
| 11:47 | kados | is ... if the info is already in the 082, why would it be in 852 also? |
| 11:47 | that's confusing | |
| 11:48 | thd | kados: 082 only has what would appear in 852 $h $i |
| 11:48 | kados: as for all 050-08X | |
| 11:49 | kados: 082 applies for all items | |
| 11:49 | kados: 852 is item or copy specific | |
| 11:53 | kados: it may be partly redundant but originally 050 and 082 were reserved exclusively for assingment by LC and other libraries had to use 09X and then had 852 for item specific numbers | |
| 11:53 | kados | I see |
| 11:54 | thd | kados: when copy cataloguing 852 $h $i should be filled from 050 - 09X automatically |
| 11:55 | kados: then the library can adjust the number as needed for local practise which should also be automated | |
| 11:55 | kados: there are several options for how cuttering is done for example |
← Previous day | Today | Next day → | Search | Index