- Koha - lists.katipo.co.nz

please unsubscribe
by michael collins 15 Dec '00

15 Dec '00

mcollins(a)ihug.co.nz

1 0

MARC Records
by Jonathan Moore 14 Dec '00

14 Dec '00

-------------------------------------------------------- ------------------------ >MARC records have an astonishing amount of detail, both >in terms of data ... >possibility of adding a field to bibliosubject to > >specify a subject > heading system used? The system we are using seems to use the redirection entries that someone mentioned. I think we must have a whole table of entries that say to place a particular marc entry tag into a particular area of the programs database. It also seems to have a similar table for exporting. It would be nice to program this flexibility into Koha. I think it would help down the road in supporting other input formats such as the differences between USMARC and UKMARC. I agree with the idea of storing the marc record en mass in the database. There is a lot of use for being able to export it out at a later time, plus I think a lot of librarians are going to want to be able to modify their collection data from a marc oriented interface. >I also have a request. Could people send me some >samples of their MARC >records so I can do some sanity checking? A lot of my >MARC records are >not from "formal" sources, but were hand entered by >the librarian here. I have posted a couple of marc record files to our ftp server. I am working on a data conversion to get our data out of the system we are currently using to something new, possibly Koha. The old system doesn't place everything in the correct place so keep that in mind. Some of the information is being put in the 900+ range instead of the "correct" location. For example the one file with bar code information has it in the 945 range and it should be a sub field of the 852 tag. One of the files is quite large 30 megs it contains marc records for just about every item in our high school collection. There should be some good examples of multiple records that should when processed have a single shared marc bib record. ftp://usd465.com/pub/marc-samples/ I'm also curious to know how librarians typically use commercial CDs of MARC records. Do they import all of the records into their library system at once, or just import records for books they hold? How are the MARC records typically stored on a CD? Is it just one big file full of MARC records? The way our librarians want to use MARC is done by an add on to the Follet system. It is a yearly cd subscription of common titles/all titles you are likely to have. There are two main uses 1. Run it through your existing holdings data and it will insert correct mark records for the items. 2. Use it when you get new titles. The marc cd is searchable. Search using whatever field you desire: title, author, isbn, etc. Once you find the title click on add copy and put in copy specific data. You are down. I think we order books pre bar coded, so I think they look up the Marc record, click create copy, scan the bar code and enter the price. Some thoughts on determining when to create a new marc bib record in the system. I think in most cases that if you have items with the same title and author then they are probably different copies of the same intellectual property and hence should be link together in the system in some we whether that is by a common marc record or some other method. What if we had 2 configurable threshholds? I list of one or more boolean conditions that must be true for automatic combining and one for queried combining. For example If author and title and publisher are all the same that the items are automatically combine in some manor or cross referenced. If the author and title are the same and nothing else then query the user if he/she wants to combine or cross reference. What are you using to do the Marc processing? I have been looking at the MARC.pm code with the same idea in mind, but I haven't done any coding yet. I have just been looking at the documentation to get an idea of what it can do. It seems pretty powerful and flexible. The website for MARC.pm is http://marcpm.sourceforge.net/ PS I am not sure how this works but I remember reading on the Library of Congress site on marc record format that there are four kinds of marc records 1. Bibliographic 2. Authority 3. Holdings 4. ??? These could be all wrong I am just going by memory. I didn't see any obvious way that these are flagged anywhere. For example Follet seems to treat any marc record without a bar code as a bib record and anything with a bar code can end up creating both a bib record and a holding record that is tied to it. Sorry to go on so long, but while we are talking about MARC record cds, how hard would it be and what legal issues are their on creating a shared marc record library for Koha? Either something that could be accessed live from remote, or perhaps a distributed network where groups of libraries inter publish marc records. Does this already exist somewhere from a non commercial source? Jonathan Moore District Technology Coordinator Winfield Public Schools 316.221.5100 Technology Office 316.221.0508 Office Fax --------------------------------------------------------------- Visit Winfield Public Schools' website at http://www.usd465.com

2 1

Re: [Koha] MARC Records
by kaye.guidetti＠wesley.com.au 13 Dec '00

13 Dec '00

As a librarian my suggestion would be to combine all subject headings into one field and allow some sort of check against the authority file (ie. do you want to accept it or not if it does not match an existing subject heading). This would allow the librarian to reject sears headings if you use LC etc. Alternatively allow something in set up that allows you to decide at the start if you will accept only the fields which contain sears headings or only the fields that contain LC headings and ignore the others. I know nothing about programming so I don't know how easy this is to do - or if it fits in with what you are trying to do Tonnesen Steve <tonnesen(a)cmsd.bc.ca>@lists.katipo.co.nz on 13/12/2000 07:19:19 Sent by: koha-admin(a)lists.katipo.co.nz To: Koha Open Source Library System <koha(a)lists.katipo.co.nz> cc: Fax to: Subject: [Koha] MARC Records MARC records have an astonishing amount of detail, both in terms of data about the book, as well as ways in which the data can be described. I'm wondering how I should handle the first difficulty: subject headings. MARC records have several different entries for subjects: 650 Subject added entry -- Topical term Subjects like Mammals, Nuclear Warfare, Law, Alcohol, Child Welfare, etc. Appears to be the most common type of subject entry, at least with my data. 651 Subject added entry -- Geographic name Subjects like Canada, Rome, Moon, China, British Columbia, etc. 600 Subject added entry -- Personal name Subjects like Plato, Helen Keller, Charlemagne, etc. In addition each subject entry also specifies a source "system" such as Library of Congress Subject Headings or Sears Subject Headings. Obviously Koha doesn't have this complexity of subject headings, and probably doesn't need it, but I'm suspecting that we probably don't want to mix subject headings from different cataloging systems. Any possibility of adding a field to bibliosubject to specify a subject heading system used? Also, should I be adding any subject headings gathered from MARC records to the catalogueentry table (which I assume is used to deter librarians from adding new, conflicting subject headings like "Cows" and "Bovines"). I also have a request. Could people send me some samples of their MARC records so I can do some sanity checking? A lot of my MARC records are not from "formal" sources, but were hand entered by the librarian here. I'm also curious to know how librarians typically use commercial CDs of MARC records. Do they import all of the records into their library system at once, or just import records for books they hold? How are the MARC records typically stored on a CD? Is it just one big file full of MARC records? Steve. _______________________________________________ Koha mailing list Koha(a)lists.katipo.co.nz http://lists.katipo.co.nz/mailman/listinfo/koha =========================================== This email message has been swept by MIMESweeper for The Wesley Group of Hospitals- Australia Although this email has been scanned for any known virus, please ensure that due care is taken and correct virus management procedures are maintained. If you have any questions contact the Help Desk on x7029 or via email at help(a)wesley.com.au ===========================================

3 2

Re: [Koha] biblio vs. biblioitems
by Tonnesen Steve 13 Dec '00

13 Dec '00

On Wed, 13 Dec 2000, Rosalie Blake wrote: > Chris has already answered, but I'll just add the rationale - which > was that biblio describes the intellectual property. So author title, > copyright date. > > Then any copies we have of this piece of copyright, whether a large > print edition, or a hardback or paperback is attached to the biblio > with the ISBN, publishing info, location in the biblioitem record. If > we've got two large print copies, they'll share one Biblioitem record, > and each copy will have its own item record. That makes sense to me. It looks to me like MARC records don't make that distinction, though. If there is a normal and large print edition, they will have seperate MARC records. I guess the title,author,copyright date combination is always unique and can be used as a key to determine if the two records actually refer to the same "intellectual property" then? If I wanted to import MARC records, and I had a MARC record for a regular David Copperfield book and a MARC record for a David Copperfield video, would they really have the same copyright date? In any case, I would have to be able to recognize somehow when entering the video into the Koha database that a biblio already exists for the item (because the book was entered previously, for example) and then link the biblioitem entry for the video back to the original biblio entry for the book. Not impossible, but I sense some possibility for errors if bibliographic data is being entered automatically from MARC records. Steve.

2 1

Re: [Koha] biblio vs. biblioitems
by Simon Blake 13 Dec '00

13 Dec '00

On Wed, Dec 13, 2000 at 12:13:02PM +1300, Rosalie Blake said: > Hi again > > Chris has already answered, but I'll just add the rationale - which > was that biblio describes the intellectual property. So author title, > copyright date. > > Then any copies we have of this piece of copyright, whether a large > print edition, or a hardback or paperback is attached to the biblio > with the ISBN, publishing info, location in the biblioitem record. If > we've got two large print copies, they'll share one Biblioitem record, > and each copy will have its own item record. At some stage after the table name was well buried in lines of code, we decided that a better name for biblioitem would be "formats". And you never know, sometime in the future we may get around to a search and replace and change it in the code. Cheers Si -- Simon Blake simon(a)katipo.co.nz Katipo Communications +64 25 300 825

1 0

biblio vs. biblioitems
by Tonnesen Steve 12 Dec '00

12 Dec '00

Why is there a biblio table and a biblioitems table? Is there a one-to-one correspondence between items in the two tables, linked by the biblionumber fields? If this is the case, what is the purpose of the biblioitemnumber field? Steve.

3 2

MARC Records
by Tonnesen Steve 12 Dec '00

12 Dec '00

MARC records have an astonishing amount of detail, both in terms of data about the book, as well as ways in which the data can be described. I'm wondering how I should handle the first difficulty: subject headings. MARC records have several different entries for subjects: 650 Subject added entry -- Topical term Subjects like Mammals, Nuclear Warfare, Law, Alcohol, Child Welfare, etc. Appears to be the most common type of subject entry, at least with my data. 651 Subject added entry -- Geographic name Subjects like Canada, Rome, Moon, China, British Columbia, etc. 600 Subject added entry -- Personal name Subjects like Plato, Helen Keller, Charlemagne, etc. In addition each subject entry also specifies a source "system" such as Library of Congress Subject Headings or Sears Subject Headings. Obviously Koha doesn't have this complexity of subject headings, and probably doesn't need it, but I'm suspecting that we probably don't want to mix subject headings from different cataloging systems. Any possibility of adding a field to bibliosubject to specify a subject heading system used? Also, should I be adding any subject headings gathered from MARC records to the catalogueentry table (which I assume is used to deter librarians from adding new, conflicting subject headings like "Cows" and "Bovines"). I also have a request. Could people send me some samples of their MARC records so I can do some sanity checking? A lot of my MARC records are not from "formal" sources, but were hand entered by the librarian here. I'm also curious to know how librarians typically use commercial CDs of MARC records. Do they import all of the records into their library system at once, or just import records for books they hold? How are the MARC records typically stored on a CD? Is it just one big file full of MARC records? Steve.

3 2

MARC Records
by Tonnesen Steve 12 Dec '00

12 Dec '00

Is anybody working on supporting Marc records at all? I've written a short perl script that will parse the data in a Marc records file, and could probably write something up that would insert records into the Mysql database, if work isn't being done on this already. How are Marc records normally handled? Is biblio data entered in only for books that the library holds, or is an entire CD full of biblio data loaded into the library database? I'm thinking that a new table should be added that will hold the entire Marc record for a bibliography, as the Marc record will certainly contain more info than the Koha Mysql tables currently will contain. Storing the Marc records would mean that the record itself could be parsed and displayed on the screen for each biblio if more info was desired. Steve Tonnesen.

2 1

Installation Log
by Tonnesen Steve 10 Dec '00

10 Dec '00

I'd like to post my log regarding installing Koha both to document my experience and to get some feedback on snags I'm hitting: Steve Tonnesen Network Administrator Coast Mountains School District Terrace, BC CANADA ---- 1. Installed required Debian packages: install mysql-base, mysql-server, mysql-client, libmysqlclient6 packages install dbd and dbi packages libdbi-perl libdbd-mysql-perl install libdate-manip-perl package 2. downloaded and untarred koha-1.06.tar.gz databaseinstall.sh mysqlrootpass Koha kohausername kohauserpass This script created the database, tables, and access account perfectly. 3. Edited scripts/C4/Database.pm with kohausername and kohauserpass Also had to set name of database to 'Koha' <-- wasn't documented in INSTALL file 4. Edited scripts/C4/Output.pm set $path to location of includes I used /usr/local/koha/librarian/htdocs/includes 5. Copied C4 directory to /usr/lib/perl5/5.005/i386-linux 6. Created virtual servers in apache's httpd.conf: ------------------------------------------------ <VirtualHost 192.168.10.230> ServerName opac.cmsd.bc.ca DocumentRoot /usr/local/koha/opac/htdocs/ ScriptAlias /cgi-bin/ /usr/local/koha/opac/cgi-bin/ </VirtualHost> <VirtualHost 192.168.10.231> ServerName koha.cmsd.bc.ca DocumentRoot /usr/local/koha/librarian/htdocs/ ScriptAlias /usr/local/koha/librarian/cgi-bin/ </VirtualHost> ------------------------------------------------ 7. Created koha directories in /usr/local/ /usr/local/koha/ /usr/local/koha/opac /usr/local/koha/opac/htdocs/ /usr/local/koha/opac/cgi-bin/ /usr/local/koha/opac/cgi-bin/koha/ /usr/local/koha/librarian/ /usr/local/koha/librarian/htdocs/ /usr/local/koha/librarian/cgi-bin/ /usr/local/koha/librarian/cgi-bin/koha/ 8. Copied files cp -a koha-1.06/opac-html/* /usr/local/koha/opac/htdocs/ cp -a koha-1.06/scripts/*.pl /usr/local/koha/opac/cgi-bin/koha/ cp -a koha-1.06/intranet-html/* /usr/local/koha/opac/htdocs/ cp -a koha-1.06/scripts/*.pl /usr/local/koha/librarian/cgi-bin/koha/ cp -a koha-1.06/scripts/acqui /usr/local/koha/librarian/cgi-bin/koha/ 9. Populated Database Tables: insert into aqbookfund values ('1', 'Annual Book Budget', 'DIST'); insert into aqbookfund values ('2', 'Fund Raising', 'SCH'); insert into aqbookfund values ('3', 'Textbooks', 'SCH'); insert into aqbudget values ('1', '2000-12-01', '2001-12-01', 5500); insert into aqbudget values ('2', '2000-12-01', '', 1200); insert into aqbudget values ('3', '2000-12-01', '2001-12-01', 8000); insert into branches values ('STWE', 'Stewart Elementary School', '4523 Sea Way', 'Stewart, BC', 'V0T 1W0', '250 636 2241', '250 636 2544', 'jsnair(a)cmsd.bc.ca', 0); insert into branches values ('SHE', 'South Hazelton Elementary School', '4521 21st Ave.', 'South Hazelton, BC', 'V0J 2R0', '250 842 6688', '250 842 5153', 'stonnesen(a)cmsd.bc.ca', 0); Some problems that I have encountered thus far: 1. When adding entries to basket, link to supplier points to whitcouls.html which is supposed to be a demo page, I gather, but doesn't exist in any case. 2. When viewing basket, total is always $16.95 (hard coded in basket.pl) and subtotal, gst and total are always blank. 3. When receiving an order, I can select the book that was received from the order and enter the appropriate information, but clicking on the "Save Changes" link on the next page returns an error (tries to load /cgi-bin/koha/acqui/, which is a directory, not a script).

2 1

RE: [Koha] Card Numbers
by Williams, Olwen - SAL 06 Dec '00

06 Dec '00

Hi there, I was involved a while ago in some programming on Koha. It's a standard check digit routine. Base 11 from memory. Horowhenua are supplied with the barcodes, but writing a program to generate numbers is an easy task (or you can change the algorithm seeing as you have access to the source). More years ago than I'll own up to it was the first task I was given as trainee programmer. -----Original Message----- From: Tonnesen Steve [mailto:tonnesen@cmsd.bc.ca] Sent: Thursday, 7 December 2000 06:26 To: koha(a)lists.katipo.co.nz Subject: [Koha] Card Numbers Here's my first question about Koha. How are the card numbers generated? I looked through the code and found the algorithm for checking validity of card numbers and determined that 123456789 was a valid card number, but how does one normally generate these? :) Steve. _______________________________________________ Koha mailing list Koha(a)lists.katipo.co.nz http://lists.katipo.co.nz/mailman/listinfo/koha

1 0