Time |
S |
Nick |
Message |
11:06 |
|
thd |
kados: what does 'careers' mean in the merging Argentine code section of LibLime's 2.4 Proposal? |
11:06 |
|
paul |
hello thd |
11:07 |
|
paul |
not sure joshua is there. |
11:07 |
|
paul |
but I know what it means ;-) |
11:07 |
|
thd |
hello paul |
11:07 |
|
paul |
they have added some features to follow what a "borrower" does in the school |
11:07 |
|
thd |
please tell :) |
11:07 |
|
paul |
(remember it's a university) |
11:08 |
|
paul |
so, they know that thomas D was in 1st year in 99-2000, then in 2nd year, option X |
11:08 |
|
paul |
in 2000-2001 |
11:08 |
|
paul |
then was in 3rd year in MIT in 2001-2002 ... |
11:08 |
|
paul |
what about your previous question (the one you wanted to ask me) |
11:08 |
|
thd |
paul: How does that help a library application? |
11:09 |
|
paul |
maybe useless. |
11:09 |
|
paul |
and useless for sure for a public library ;-) |
11:10 |
|
thd |
paul: before the translation question I have for you. Do you know what the Argentine calendars do? |
11:10 |
|
paul |
you can define closed-days |
11:10 |
|
paul |
thus, when you issue a book, the return date will avoid closed days |
11:10 |
|
paul |
(and find the next open day available) |
11:10 |
|
paul |
iirc |
11:12 |
|
thd |
paul: Well that sounds very nice. Book return slots in New York City were closed long ago because people would use them to cause mischief and start fires! |
11:14 |
|
thd |
paul: So from http://www.saas.nsw.edu.au/koh[…]ge=KohaRoadmap3.0 ... |
11:14 |
|
thd |
paul: What are BnF 'notes'? |
11:16 |
|
paul |
(or don't remember at all...) |
11:16 |
|
thd |
paul: I assumed that you or hdl had written that description :) |
11:17 |
|
paul |
if I did, I don't remember at all. and hdl did not either |
11:17 |
|
paul |
(afaik, and, as i'm it's employeer, I should know what he does ;-) ) |
11:18 |
|
thd |
paul: Does the following description make sense to you? |
11:18 |
|
paul |
nope |
11:18 |
|
thd |
? remove existing notes ? Import all the fields ? Bond with acquisitions ?; Require a thorough analysis; PP : Problem of the timeout Apache to be regulated. For the recovery of the local fields, to see Marc ::Record |
11:19 |
|
thd |
paul: I assume that you also do not understand what "Imports Authorities BRANCH" is supposed to do because that is related to BnF notes. |
11:20 |
|
paul |
for this, i'm not sure, but i've an idea : |
11:20 |
|
thd |
yes |
11:20 |
|
paul |
there is a z3950 service in BNF to retrieve authorities |
11:20 |
|
paul |
maybe it's related to this. |
11:20 |
|
paul |
but i'm not sure either. |
11:20 |
|
paul |
maybe this comes from Ineo (you should ask joshua) |
11:21 |
|
thd |
paul: how do I get to that BnF service? |
11:21 |
|
thd |
paul: Is it open access? |
11:21 |
|
paul |
I was writing a mail to ENSMP that was at a meeting last week where BNF showed those new services |
11:23 |
|
thd |
paul: Does the Z39.50 client daemon at demo.koha-fr.org ever run for bibliographic records? |
11:24 |
|
thd |
paul: Who is Ineo? |
11:25 |
|
kados |
thd: I think that was before your time |
11:25 |
|
kados |
thd: INEO is a company in France that is planning to support KOha |
11:25 |
|
paul |
about dmeo.koha-fr.org => i don't know. hdl has the answer |
11:25 |
|
kados |
paul: welcome back! |
11:25 |
|
paul |
thanks, hello joshua |
11:26 |
|
hdl |
what about demo.koha-fr.org ? |
11:26 |
|
paul |
Ineo is actively looking for a perl dev to work full time on Koha |
11:26 |
|
thd |
hdl: Does the Z39.50 client daemon at demo.koha-fr.org ever run for bibliographic records? |
11:26 |
|
paul |
http://fr.lolix.org/search/off[…]ffre.php3?id=4780 |
11:26 |
|
paul |
(in french) |
11:26 |
|
hdl |
Ah ! Not at the moment. |
11:27 |
|
hdl |
But it is sensible to set it on. |
11:27 |
|
hdl |
I should do this... when I have time. |
11:27 |
|
hdl |
Sorry. |
11:27 |
|
paul |
because we may have flood & flood BNF server too. |
11:27 |
|
hdl |
:S |
11:27 |
|
thd |
paul: Is it dangerous? |
11:28 |
|
paul |
so, i think we should not on a public prof interface. |
11:29 |
|
thd |
paul: What does 'inscription' mean in "Messages of blocking/releasing of the inscription" |
11:30 |
|
paul |
?? |
11:31 |
|
thd |
paul: Does the following make sense to you? |
11:31 |
|
thd |
To make explicit the messages of blocking and releasing of inscription compared to the rules of management of the media library. Management of the inscription for readers "except library". |
11:32 |
|
thd |
paul: Or is that another puzzle from INEO? |
11:33 |
|
paul |
that is, but I remember a little this one. |
11:33 |
|
paul |
it's not Ineo related, in fact. It's SAN (Ouest-Provence) that want those features. |
11:33 |
|
thd |
paul: Is that one of your customers? |
11:33 |
|
paul |
they are related to issuing rules, that are a little different in SAN |
11:34 |
|
paul |
nope. that is Ineo one. |
03:16 |
|
osmoze |
hello |
03:23 |
|
chris |
hi osmoze |
03:56 |
|
paul |
hello chris |
03:56 |
|
paul |
quit quite today on #koha ;-) |
03:57 |
|
paul |
(quite quite I mean) |
03:57 |
|
hdl |
hi all |
03:59 |
|
paul |
(will deploy it on IPT server, for 2 days of tests before official release) |
04:24 |
|
chris |
cool, good news about 2.2.4 paul |
06:53 |
|
Sylvain |
hi all |
09:10 |
|
thd |
paul: are you there? |
09:12 |
|
thd |
paul: Tell me about the problem of leading or trailing articles in Arabic. I have offered to fix many character set issues for Koha. |
09:24 |
|
thd |
hdl: perhaps you could answer a question about sorting algorithms for UNIMARC titles. |
09:25 |
|
hdl |
If it not Too time consuming. |
09:26 |
|
thd |
hdl: In MARC 21 there is an indicator for defining the number of characters occupied by the leading non-sorting word and trailing space. I can find no equivalent for UNIMARC. |
09:28 |
|
thd |
hdl: Does one have to simply identify the leading articles with string matches for the language to sort UNIMARC titles without the leading articles? |
09:31 |
|
hdl |
Leading articles : is this not related to stop words ? |
09:31 |
|
thd |
hdl: Leading articles would be: le, la, l', les, un, une, des, de, d' for French. |
09:32 |
|
hdl |
They are all included in french stop words list. |
09:32 |
|
hdl |
I ordered items on a branch field. |
09:33 |
|
hdl |
Titles were already ordered. |
09:33 |
|
thd |
hdl: Stop words are usually words too common to index because of index size problems or lack of rarity for usefulness in distinguishing texts in a search. |
09:34 |
|
thd |
hdl: Do you remove leading articles when ordering titles? |
09:35 |
|
hdl |
I order items in the result lists. |
09:35 |
|
hdl |
Titles were already ordered. |
09:36 |
|
hdl |
But Answer is no. |
09:36 |
|
thd |
hdl: What do you mean by titles were already ordered? |
09:36 |
|
hdl |
I didn't design any peculiar process for order. |
09:37 |
|
hdl |
Testing your base... you could have seen the order on your own. |
09:37 |
|
hdl |
;) |
09:37 |
|
thd |
hdl: Therefore titles appear in letter by letter order. |
09:39 |
|
thd |
hdl: I suspected I would know what to expect form Koha at this point in its development, such that I did not need to test. Testing would merely have led me to ask the same question :) |
09:41 |
|
hdl |
Maybe, but you wouldnot have searched especially for me. :) |
09:41 |
|
thd |
hdl: Do your libraries not have enough titles for this question to be asked before? |
09:43 |
|
thd |
hdl: The reason I search for you or paul is that this is not a problem in MARC 21 because there is an indicator position setting which characters from the title should not be used for sorting. Yet I find nothing similar for UNIMARC. I had hoped that you would know what the standard sorting practise not followed by Koha yet might be. |
09:44 |
|
hdl |
Sorry. |
09:45 |
|
hdl |
I don't know how it is managed. |
09:45 |
|
hdl |
Each library has its parctise. |
09:45 |
|
hdl |
s/parctise/practise/ |
09:46 |
|
hdl |
putting leading article behind between squared brackets. |
09:46 |
|
thd |
hdl: There should be standards :) paul posed a related question to koha-devel for Arabic. |
09:47 |
|
thd |
hdl: Is that a French bibliographic rules standard? |
09:48 |
|
hdl |
But it is done at the BNF |
09:49 |
|
thd |
hdl: I assume they follow whatever standard there is :) |
09:52 |
|
thd |
hdl: Do you know what paul meant about Arabic usage in his message to koha-devel. It does not seem to be a question of script directionality. |
09:52 |
|
paul |
thd, i'm here |
09:53 |
|
paul |
it's not a question of directionality |
09:53 |
|
paul |
let me write some english words in "arabic" syntax : |
09:53 |
|
paul |
"this is samplea" |
09:53 |
|
paul |
"what do youmean" ? |
09:53 |
|
paul |
"i'm tall mana" |
09:54 |
|
paul |
do you understand now ? |
09:54 |
|
thd |
paul: So what did you mean about stop words in Arabic? Did you mean leading articles? |
09:54 |
|
paul |
I was wondering if zebra can handle such constructions |
09:54 |
|
paul |
yes. |
09:55 |
|
thd |
paul: Did you mean to close the space between the word and the article in your examples? |
09:55 |
|
paul |
this is an english example. |
09:56 |
|
paul |
if I understand well what 'ive been told, there are more complex questions. |
09:56 |
|
paul |
because, sometimes "a" at the end of a word is important ("a" english example as usual) |
09:58 |
|
thd |
paul: I see now this is for every article throughout the text not just the leading article that I had been asking hdl about. |
09:59 |
|
thd |
paul: Are you certain that this is not related to script directionality? |
10:00 |
|
thd |
paul: What software was being used for encoding where you were? |
10:02 |
|
thd |
paul: One problem with Arabic, Hebrew, and related languages, none of which I can read, is that they have a distinction between logical character ordering for encoding and readable character ordering for display. |
10:03 |
|
thd |
paul: This has to be managed by software. |
10:05 |
|
thd |
paul: What I imagine that you want to ask is how can Arabic queries be correctly handled so that they retrieve the proper results as those results are encoded. |
10:09 |
|
thd |
paul: Is your question not equivalent to French conversion from 8859 in the user's browser to UTF-8 for the database and then back to 8859 for display? Directionality in Arabic is merely an additional complexity with which a complete query and display conversion has to work. |
10:11 |
|
thd |
paul: What software were they using to manage Arabic input and display? |
10:13 |
|
paul |
thd : my contact uses winISIS, that has an arab version. |
10:14 |
|
thd |
paul: I have offered to fix these character set issues in Koha for some major languages and character sets. I would at least like to know how to think about the problem for Koha. |
10:14 |
|
thd |
paul: Is this the UNESCO software? |
10:14 |
|
paul |
yep |
10:16 |
|
thd |
paul: Does the software display correctly for them under Windows? |
10:16 |
|
paul |
it seems |
10:17 |
|
thd |
paul: Then, what circumstance lead you to notice the article ordering problem? |
10:21 |
|
thd |
paul: Do you know if they using Unicode or 8859-6 for Arabic WinISIS? |
10:29 |
|
thd |
paul: I am downloading the manual for Arabic ISIS now. However, if the manual is in Arabic I will not be able to read it. |
10:29 |
|
thd |
:) |
10:32 |
|
thd |
paul: This question is important. If Arabic WinISIS worked fine for them as you had said, what circumstance lead you to notice the article ordering problem? |
10:33 |
|
paul |
thd : i have no idea at all. I was just asking a question to indexdata after a question a tunisian asked me last week. |
10:33 |
|
paul |
Otherwise, i don't nothing with arab ;-) |
10:38 |
|
thd |
paul: Your question might be rephrased in relation to language character ordering rules. However, I expect that is something that the application software, namely Koha, would have to solve for converting queries and cataloguing data back and forth between display forms, editing forms, and storage encoding forms. |
10:40 |
|
thd |
paul: There is software that addresses these issues for some contexts. Yudit is an example for Arabic, Hebrew, and related languages. |