Time |
S |
Nick |
Message |
12:38 |
|
cm |
hi kados |
12:38 |
|
cm |
do you have a minute? |
12:57 |
|
rch |
hi cm |
12:57 |
|
paul |
hi cm & rch |
12:57 |
|
rch |
kados is in NZ... it's still a bit early there |
12:57 |
|
rch |
hey paul |
13:07 |
|
cm |
hi guys. :) |
13:07 |
|
cm |
maybe one of you can answer my question. |
13:08 |
|
cm |
i just ran update_items.pl, and got this error: |
13:09 |
|
cm |
counting first |
13:09 |
|
cm |
no recent items to update |
13:09 |
|
cm |
finding biblionumbers |
13:09 |
|
cm |
fetching marc and items data, updating |
13:09 |
|
cm |
Can't call method "fetchrow" on an undefined value at /usr/local/koha/intranet/cgi-bin/misc/update_items.pl line 93. |
13:10 |
|
cm |
i had just changed my kohaadmin password and was running this from the command line to test it, |
13:10 |
|
cm |
so I don't know if it's related to that or not. |
13:11 |
|
paul |
mmm... if it's your 1st error message, then no, it's not related to the kohaadmin password change |
13:11 |
|
bureado |
I haven't used update_items, but I don't think it's related with the authentication. |
13:11 |
|
bureado |
I think there might be an empty table somewhere. |
13:11 |
|
paul |
you would have something like "connection refused, for XXXXkohaadmin on <mysql_db_name>" |
13:12 |
|
cm |
yeah. but zebra has the password stored in all those kohalis files. that's why I ran it. |
13:13 |
|
paul |
Can't call method "fetchrow" is related to mySQL, NOT zebra |
13:13 |
|
paul |
zebra errors are more weird ;-) |
13:13 |
|
cm |
I know, on both counts. :D |
13:13 |
|
rch |
cm: you ran it with -all ? |
13:14 |
|
cm |
i was just wondering why this fetchrow error would crop up. |
13:14 |
|
cm |
okay, now i feel silly...I forgot to add the option on the end. |
13:15 |
|
cm |
ran it with -today and it worked. |
13:15 |
|
cm |
sorry to bug you all! |
13:15 |
|
paul |
cm ++ |
13:15 |
|
paul |
your welcome. |
13:15 |
|
cm |
yep, thank you. :) |
13:15 |
|
paul |
should I write "you're welcomed", our "your welcome" ??? |
13:16 |
|
bureado |
Don't feel silly. I've been fighting with the Intranet for 30 minutes trying to add a MARC framework until I found that the "OK" button in the spanish translation isn't type="submit" :( |
13:16 |
|
cm |
technically, it should be "you're welcome" |
13:16 |
|
paul |
what do kados in NZ ? |
13:16 |
|
paul |
works with katipo ? |
13:17 |
|
cm |
bureado: yeah, koha can be humbling like that. ;) |
13:21 |
|
cm |
i won't be getting to NZ anytime soon, but at least I'm going to Las Vegas next month. :) |
14:36 |
|
paul |
ok, time to leave for dinner. read you back in 2 yours for the irc meeting. |
15:29 |
|
bureado |
In Koha 2.2.7, when you Add a biblio, why is it so slow to load the MARC tags form? Does it load information about ALL tags? I assume that, from the User's Guide. Is there a way to avoid that? How do you improve the performance in the Add Biblio form. |
15:34 |
|
waylon |
heya all. |
15:40 |
|
kados |
hi all |
15:40 |
|
waylon |
20 minutes till meeting, ya? |
15:41 |
|
kados |
and I think paul said he needed more time |
15:41 |
|
kados |
we maybe need to push it back an hour |
15:41 |
|
waylon |
thats fine.. if it is for the rest... |
15:52 |
|
kados |
chris can't make it either, he's got a conflicting tech meeting |
15:53 |
|
waylon |
-nods- push it back then. hey, we need chris. |
15:54 |
|
hdl |
hi toins |
15:54 |
|
hdl |
hi kados. |
15:54 |
|
toins |
hello |
15:54 |
|
toins |
hi all |
15:55 |
|
hdl |
kados : I can be back in an hour if every one agrees. |
15:55 |
|
kados |
hdl: great, read you then |
15:55 |
|
kados |
mail sent to koha-devel too |
16:00 |
|
thd |
kados: what would an SQL version of release 3 actually mean? |
16:15 |
|
slef |
hi all |
16:16 |
|
bureado |
Hey |
16:24 |
|
paul |
hello world |
16:25 |
|
paul |
baby sleeping, wife reading book to the 2nd boy, 1st reading alone. |
16:25 |
|
paul |
i'm ready ;-) |
16:27 |
|
paul |
thd : a sql version of koha 3.0 would mean a version that don't need zebra. |
16:27 |
|
paul |
it would be interesting for "small" libraries |
16:27 |
|
paul |
(up to something like 30 000 biblios) |
16:27 |
|
thd |
paul: I understood that much but ... |
16:27 |
|
paul |
you want to know how much work it requires to develop it ? |
16:29 |
|
thd |
paul: actually, yes, my point of view without any customers to support is that maintaining multiple versions is too much work for the developers |
16:29 |
|
paul |
the idea would be to have only 1 version. Only a few things would have to be modified. |
16:29 |
|
paul |
2 in fact : |
16:30 |
|
thd |
paul: but what my original question meant is that why would you not continue with 2.X if you wanted to maintain SQL? |
16:30 |
|
paul |
because there are many many new features in 3,0 that would be a pain to backport to 2.2 |
16:31 |
|
paul |
- zebraop function : zebraop stores something to tell zebra that the biblio has to be reindexed. we could check for a systempref, and, if zebra=off, we could use a reverted index to store things in SQL. I already wrote something about this when we started investigating tools for koha 3.0 |
16:31 |
|
thd |
paul: why would anyone not want the advantages of Zebra? |
16:32 |
|
waylon |
what exactly does Zebra do? |
16:32 |
|
paul |
because it is hard to setup & maintain (from a user point of view) |
16:33 |
|
paul |
waylon: when a biblio or an authority is added, we store the "raw" record (XML) in SQL. we (almost) don't have any index on it |
16:33 |
|
paul |
but : |
16:33 |
|
paul |
we send a message to zebra saying "hey, zebra, please index this XML for me" |
16:33 |
|
paul |
then, when someone search in the catalogue, koha ask zebra : |
16:34 |
|
paul |
"hey, retrieve biblios with title=XXX ..., pls" |
16:34 |
|
waylon |
ah.. this is instead of the MARC style indexing... |
16:34 |
|
paul |
zebra does only this, but as fast as lightning |
16:34 |
|
thd |
paul; what about maintaining zebra is complex for the user after installation, and what users are maintaining there own? |
16:34 |
|
paul |
thd : |
16:34 |
|
paul |
1- impossible to dump a zebradatabase |
16:34 |
|
thd |
s/there/their/ |
16:35 |
|
paul |
2- it sometimes crashes & requires full "reindexing" |
16:35 |
|
paul |
3- it's really hard to change anything in the index structure (and unavailable to a librarian I bet) |
16:35 |
|
waylon |
less flexiblity... right. |
16:36 |
|
paul |
so, if the library decides that (in UNIMARC) 502 field is now usefull, they have to activate it in Koha, THE? |
16:36 |
|
paul |
THEN |
16:36 |
|
paul |
going to zebra config file & update them (which is a pain) |
16:36 |
|
paul |
THEN |
16:36 |
|
paul |
reindexing everything |
16:36 |
|
paul |
quite a complex things. |
16:36 |
|
waylon |
how long does an indexing take? |
16:36 |
|
paul |
for a library with a full time computer scientist, it's possible. |
16:37 |
|
paul |
for a small library without, it's impossible |
16:37 |
|
thd |
waylon: actually Zebra is perhaps more flexible for the effort required but alien to the common SQL knowledge people already have. |
16:37 |
|
paul |
waylon: mmm... for example, for a 200 000 biblios database (like SAN-OP), |
16:37 |
|
slef |
dagnabbit, meeting was postponed by an hour |
16:37 |
|
slef |
only just saw that message |
16:37 |
|
slef |
now I've aimed dinner for the wrong time :-/ |
16:37 |
|
paul |
reindexing requires around 1 hour |
16:38 |
|
paul |
(exporting datas from SQL to iso2709 & indexing in zebra |
16:38 |
|
paul |
) |
16:38 |
|
thd |
paul: yes but what small libraries are actually running Koha on their own even with just SQL in version 2 already? |
16:38 |
|
paul |
thd ??? |
16:38 |
|
dewey |
it has been said that thd is just dying to tell you but I fear that if I reveal all I will lose the ability to obtain funding for a business myself. or skeptical about that as I would have seen only changes in the DOM that provide enhanced methods of access to the document historically. |
16:38 |
|
slef |
dewey: shut up |
16:38 |
|
dewey |
slef: i'm not following you... |
16:38 |
|
slef |
dewey: silence |
16:38 |
|
dewey |
slef: huh? |
16:38 |
|
thd |
dewey: forget it |
16:38 |
|
dewey |
thd: I forgot it |
16:38 |
|
slef |
dewey: mute |
16:38 |
|
dewey |
slef: huh? |
16:40 |
|
thd |
paul: are any small libraries maintaining their own installation of Koha now? |
16:40 |
|
paul |
yes & no. |
16:40 |
|
paul |
once installed & integrated in the local network, there's nothing to do. |
16:41 |
|
paul |
for example, backups are easy to do & well known. |
16:41 |
|
paul |
most of my customers don't have a computer geek near them & the server is always up. |
16:41 |
|
thd |
paul: could the issues for maintainance not be automated in a script? |
16:42 |
|
thd |
or set of scripts |
16:42 |
|
paul |
not all i'm afraid, because zebra is like unix : it's friendly, but very picky with who it's friends are... |
16:43 |
|
thd |
paul: do you mean it requires authentication for the server? |
16:43 |
|
paul |
no. |
16:44 |
|
paul |
I mean many things can't be automated because they require you to know exactly how to do them. |
16:44 |
|
thd |
paul: what do you mean. Zebra functions are scriptable like any other good Unix application. |
16:45 |
|
paul |
maybe you're right and i'm not confident enough with zebra for instance. |
16:45 |
|
paul |
but I feel that it will be harder to maintain a zebra server than a full SQL one |
16:46 |
|
hdl |
dewey : forget thd |
16:46 |
|
dewey |
hdl: I forgot thd |
16:47 |
|
slef |
can someone put 'installer' under the 2.2.8 heading on http://wiki.koha.org/doku.php?[…]ndandnotes07feb07 please? |
16:48 |
|
paul |
can't you do it yourself slef ? you don't have write access to the wiki ? |
16:48 |
|
slef |
apparently I do |
16:48 |
|
slef |
sorry about that |
16:48 |
|
thd |
paul: I take it as given that Zebra is more difficult because Index Data does not believe in producing or has never been paid enough to produce full and complete documentation of the sort Ross Singer and I like. |
16:48 |
|
slef |
there are many wikis I can't edit - I thought this was another |
16:48 |
|
paul |
thd : indexdata produces long & large docs. |
16:48 |
|
slef |
buh!? "This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong." |
16:49 |
|
paul |
the problem is that it's usually very hard to understand ! |
16:49 |
|
slef |
paul: I don't have write access to the wiki. |
16:50 |
|
thd |
paul: I think the documentation is sufficient but it is not always as detailed and helpful as some MySQL documentation for the corresponding example. |
16:50 |
|
paul |
hdl around ? |
16:51 |
|
hdl |
yes |
16:51 |
|
paul |
you speak of web installer to slef ? |
16:51 |
|
hdl |
not yet. |
16:51 |
|
hdl |
was wiating for a good occasion. |
16:52 |
|
thd |
paul: do your Zebra scripts not work reliably every time? |
16:52 |
|
paul |
thd : i'm not sure, i don't have enough experience |
16:52 |
|
thd |
slef: what wiki page claimed you had no permission? |
16:54 |
|
slef |
thd: http://wiki.koha.org/doku.php (Edit page link from agenda) |
16:54 |
|
thd |
slef: there have been some authentication/permission problems with the Koha wiki and I would appreciate a very complete bug report. |
16:55 |
|
thd |
slef: what was the exact error message. Can you reproduce ti? |
16:55 |
|
thd |
s/ti/it/ |
16:55 |
|
slef |
thd: yes. Exact error was in ""s above. |
16:57 |
|
thd |
slef: Kados had imagined it was a template change which I had introduced but I think it is something else if he has actually reverted completely. |
16:58 |
|
thd |
slef: what browser were you using when permission was denied? |
17:00 |
|
thd |
slef: and what cookie and JavaScript policy do you have set for that browser? |
17:01 |
|
thd |
slef: do you know what the page is locked by another user dokuwiki message is? |
17:05 |
|
thd |
slef: sorry to pester you about this but maybe we could resolve the problem, which I have never seen myself in my own usage, if we had more diagnostic information. Kados said that nothing appears in the error logs. |
17:05 |
|
kados |
hi all |
17:05 |
|
paul |
hi kados |
17:05 |
|
toins |
hello |
17:05 |
|
kados |
sorry I'm late :( |
17:06 |
|
paul |
3mn late is not too much |
17:06 |
|
paul |
;-) |
17:06 |
|
kados |
:-) |
17:06 |
|
kados |
so I suppose we can start |
17:06 |
|
paul |
+ you're maybe still "jetlagged" |
17:06 |
|
kados |
well, I would like to use that as an excuse |
17:06 |
|
kados |
but I've been in NZ for two weeks no :-) |
17:06 |
|
thd |
kados: slef just reported a permissions problem with the wiki. We could discuss what it might be at your convenience. |
17:06 |
|
kados |
thd: so maybe the problem is not with the template |
17:07 |
|
kados |
slef: did you have trouble editing a page after you logged in? |
17:07 |
|
slef |
kados: yep |
17:07 |
|
thd |
kados: I know the template code very well but I do have some ideas for diagnosis if you want to discuss it after the meeting |
17:07 |
|
kados |
shoot |
17:07 |
|
kados |
thd: sure, lets do that |
17:08 |
|
kados |
hehe |
17:08 |
|
kados |
so I guess first a roll call is in order |
17:08 |
|
kados |
who's here? |
17:08 |
|
thd |
kados: maybe you had not completely reverted but we will discuss after |
17:08 |
|
kados |
slef, thd, hdl, toins, and kados so far :-) |
17:08 |
|
slef |
MJ Ray |
17:08 |
|
kados |
hey rch |
17:08 |
|
kados |
rch is Ryan Higgins, he works for LibLime :-) |
17:08 |
|
paul |
waylon was here few minuts ago |
17:09 |
|
tnb |
i'm here :) |
17:09 |
|
paul |
while r*a*ch is Rachel, from katipo. |
17:09 |
|
kados |
chris is still at a tech meeting |
17:09 |
|
waylon |
im here |
17:09 |
|
kados |
great, I'd say that's quorum |
17:10 |
|
kados |
tnb is Tina Burger, also from LibLime |
17:10 |
|
paul |
paul, hdl and toins are all from the same company too. |
17:10 |
|
kados |
paul++ :-) |
17:11 |
|
waylon |
must be cool, getting paid for doing koha work. |
17:11 |
|
kados |
waylon: remind me where you're from? |
17:11 |
|
paul |
all frenchies and probably all listening or looking for france-argentina soccer match maybe |
17:11 |
|
kados |
ahh, cool |
17:11 |
|
kados |
so we have an agenda |
17:12 |
|
kados |
but I assume since slef had trouble editing maybe others didn't get their stuff added? |
17:12 |
|
thd |
kados: you would recognise the former nick of waylon |
17:12 |
|
kados |
anyone have anything to add to the agenda? |
17:12 |
|
kados |
thd: what's that? |
17:12 |
|
paul |
I don't have any problem adding what slef wanted to add |
17:12 |
|
paul |
s/don't/didn't/ |
17:12 |
|
waylon |
Genjimoto |
17:12 |
|
waylon |
genji... |
17:12 |
|
dewey |
genji is waylon_'s username |
17:12 |
|
waylon |
eh.. i abandoned it, liking to use my real name. |
17:12 |
|
kados |
ahh, righ! |
17:12 |
|
kados |
right even :-) |
17:13 |
|
kados |
waylon: been a while since I've seen you, welcome back :-) |
17:13 |
|
kados |
paul: ok, great |
17:13 |
|
paul |
(hdl added "web installer" line too |
17:13 |
|
kados |
so we did the introductions already :-) |
17:13 |
|
paul |
) |
17:13 |
|
paul |
it seems ;-) |
17:13 |
|
kados |
who has some news? :-) |
17:14 |
|
paul |
some business news : we are starting 3 new installs in france. |
17:14 |
|
kados |
(/me got first introduction to rubgy and cricket here in NZ) |
17:14 |
|
waylon |
sweet, paul! |
17:14 |
|
paul |
2 on public libraries & one for a high school |
17:14 |
|
kados |
great paul! |
17:14 |
|
kados |
paul: 3.0 installs? |
17:14 |
|
paul |
nope. 2.2.x |
17:15 |
|
kados |
gotcha |
17:15 |
|
cm |
ccfls should be going live with meadville in march... |
17:15 |
|
cm |
though we keep moving our target date. :) |
17:15 |
|
kados |
hi cm |
17:15 |
|
waylon |
Good. im kinda taking on 2.2.x as a pet project. |
17:15 |
|
cm |
hi all |
17:15 |
|
kados |
waylon++ |
17:15 |
|
paul |
pet project ??? |
17:15 |
|
hdl |
(3.0 is in production at SAN WP but you knew.) |
17:16 |
|
kados |
paul: it's an english idiom, means 'on the side project for fun' |
17:16 |
|
waylon |
Ya.. when i don't have anything to do.. i pick up a bug and fix it. |
17:16 |
|
kados |
paul: at least in american english, dunno in NZ english ;-) |
17:16 |
|
paul |
lol |
17:16 |
|
slef |
it's in English too |
17:16 |
|
paul |
so, 2.2.8 ? |
17:17 |
|
hdl |
+ there was a french bug session Monda 22 January ... 108 bugs were raised.) |
17:17 |
|
hdl |
most of them fixed. |
17:17 |
|
paul |
hdl: 149 at the end ;-) |
17:17 |
|
kados |
hdl++ |
17:17 |
|
kados |
wow, that's great! |
17:17 |
|
waylon |
which version, hdl? |
17:17 |
|
hdl |
but there are still some tricky ones with zebra. |
17:17 |
|
hdl |
(3.0) |
17:18 |
|
kados |
so some of you may have heard already |
17:18 |
|
kados |
LibLime and Katipo have signed a 'heads of agreement' for LibLime to take over Katipo's Koha operation in NZ, including hiring Chris, Russ and Mason |
17:19 |
|
hdl |
WoW. |
17:19 |
|
cm |
indeed! |
17:19 |
|
kados |
I'll post again for slef's benefit |
17:19 |
|
kados |
LibLime and Katipo have signed a 'heads of agreement' for LibLime to take over Katipo's Koha operation in NZ, including hiring Chris, Russ and Mason |
17:19 |
|
paul |
hiring chris/russ/mason full time ? |
17:19 |
|
thd |
kados: what is a 'heads of agreement'? |
17:19 |
|
kados |
it's quite big news for us :-) |
17:19 |
|
kados |
paul: correct |
17:19 |
|
slef |
thd: like a memorandum of understanding, I think |
17:19 |
|
paul |
wow ! |
17:20 |
|
kados |
it's an agreement that will help both companies streamline our business workflow |
17:20 |
|
kados |
katipo isn't a Koha company, they are web development company |
17:20 |
|
kados |
and LibLime needs more capacity for Koha development and support |
17:20 |
|
kados |
so the arrangement is good for both of us |
17:20 |
|
waylon |
Will Katipo keep working on Koha, or totally give over to LibLine? |
17:21 |
|
waylon |
s/line/lime/ |
17:21 |
|
toins |
waylon, good question.. |
17:21 |
|
kados |
waylon: not sure 'give over' is the right phrase, but yea, katipo won't be working on Koha any more |
17:21 |
|
thd |
kados: what degree of difference will that likely mean for Koha development in terms of the time that chris, russ, and mason have as compared to the present? |
17:22 |
|
paul |
LibLime Ltd NZ will be a part of LibLime US, or something else ? |
17:22 |
|
waylon |
LibLime Ltd NZ? |
17:22 |
|
kados |
paul: it's a subsidiary of LibLime US |
17:23 |
|
kados |
chris and mason will have more time to devote on koha development |
17:23 |
|
kados |
so it will be especially good for the koha project |
17:23 |
|
toins |
great ! |
17:23 |
|
kados |
waylon: :-) |
17:23 |
|
thd |
kados: that is understood but would you estimate the degree as compared to the time they are currently spending? |
17:23 |
|
paul |
so they leave katipo if iiuc |
17:23 |
|
kados |
and russ will be expanding the market in the austrialasia area |
17:23 |
|
kados |
paul: correct |
17:24 |
|
waylon |
iiuc? |
17:24 |
|
paul |
if i understand correctly |
17:24 |
|
waylon |
ah. |
17:25 |
|
kados |
so that's probably it for news unless anyone has any other items :-) |
17:25 |
|
kados |
or questions, etc. |
17:25 |
|
kados |
so 2.2.8 |
17:25 |
|
waylon |
liblime got a job list? |
17:26 |
|
kados |
waylon: http://liblime.com/about/careers |
17:26 |
|
kados |
2.2.8 |
17:26 |
|
kados |
I know there is at least one feature that still ahs some bugs |
17:27 |
|
kados |
the new corporate serials module |
17:27 |
|
kados |
it will be fixed soon though |
17:27 |
|
rch |
has anyone tested the holidays feature in 2_2? |
17:28 |
|
kados |
rch: was it committed to 2_2? |
17:28 |
|
rch |
yep |
17:28 |
|
kados |
cool |
17:28 |
|
cm |
rch, it seems to work in dev_week. |
17:28 |
|
paul |
rch : yep |
17:28 |
|
rch |
mason added to 2_2 |
17:28 |
|
kados |
excellent |
17:28 |
|
kados |
mason++ |
17:28 |
|
paul |
I have 68 commits to check on rel_2_2 |
17:28 |
|
kados |
paul++ |
17:29 |
|
kados |
paul: we also discussed doing something with templates for 2.2.8 |
17:29 |
|
paul |
(you'll say ++ when it will be done ;-) ) |
17:29 |
|
kados |
hehe |
17:29 |
|
paul |
right |
17:29 |
|
kados |
paul: do we want to eliminate default/css? or is it not appropriate? |
17:30 |
|
paul |
that would be a good thing. but i'm afraid we can't "afford" it here (in France) |
17:30 |
|
kados |
ahh, ok |
17:30 |
|
hdl |
atm I don't favour this idea. |
17:30 |
|
paul |
we already have too many things to do. |
17:30 |
|
kados |
because no french translation + no stylesheets for french libraries |
17:30 |
|
paul |
kados: ++ |
17:30 |
|
kados |
ok, no problem |
17:30 |
|
hdl |
should be a great job to translate all the stuff. |
17:30 |
|
paul |
s/great/huge/ |
17:30 |
|
kados |
maybe for 2.2.9 :-) |
17:30 |
|
kados |
yea, I can understand that |
17:31 |
|
kados |
it's a lot of work |
17:31 |
|
kados |
I have a small feature I created recently |
17:31 |
|
kados |
paul: you can tell me if you want me to commit it |
17:31 |
|
thd |
we need an internationalisation model which is not necessarily template dependent |
17:31 |
|
kados |
paul: http://search.athenscounty.lib.oh.us/bib/16595 |
17:32 |
|
kados |
paul: see the 'save record' feature? |
17:32 |
|
kados |
paul: allows saving record in a variety of formats |
17:32 |
|
paul |
would be great. |
17:33 |
|
kados |
ok, I'll put that on my list then |
17:33 |
|
paul |
although i'm afraid DC schema will not work for UNIMARC |
17:33 |
|
kados |
ahh, good point |
17:33 |
|
kados |
what about MODS? |
17:33 |
|
kados |
probably not either |
17:33 |
|
paul |
right |
17:33 |
|
waylon |
thats interesting. Saving via MARC frameworks? |
17:33 |
|
hdl |
Cool. Are you using XSLT ? |
17:33 |
|
kados |
hdl: yep :-) |
17:33 |
|
hdl |
loc ones ? |
17:33 |
|
kados |
hdl: I just used default LOC stylesheets |
17:33 |
|
hdl |
:P |
17:34 |
|
paul |
you show me the stuff, but did not commit anything yet, right ? |
17:34 |
|
hdl |
Lucky you. |
17:34 |
|
kados |
paul: right |
17:34 |
|
kados |
paul: Record.pm is commited to dev_week and rel_3 |
17:34 |
|
kados |
paul: but not the new version that actually works |
17:34 |
|
thd |
paul: what do you mean by Dublin core not working with UNIMARC? |
17:35 |
|
kados |
paul: and it is completely independent of the rest of the system so it should work fine with rel_2 |
17:35 |
|
hdl |
UNIMARC is way back to have so good standard helps. |
17:35 |
|
paul |
i know for months that someone will have to write the XSLT for UNIMARC => DC if it does not exist |
17:35 |
|
paul |
thd : the DC says "DC.Title" for the title. |
17:35 |
|
paul |
the XSLT must know where is the title. |
17:35 |
|
paul |
in MARC21, it's in 245 |
17:35 |
|
paul |
in UNIMARC, it's in 200$a |
17:35 |
|
kados |
paul: i can hide the MODS + DC based on UNIMARC vs MARC21 |
17:35 |
|
kados |
paul: until you or someone ahs time to create a stylesheet |
17:35 |
|
paul |
right. that was my plan too |
17:36 |
|
kados |
cool |
17:36 |
|
kados |
so what else for 2.2.8? |
17:36 |
|
thd |
paul, kados: we just need a UNIMARC implementation of the XSLT transform |
17:36 |
|
kados |
thd: yep |
17:36 |
|
paul |
thd : yep |
17:36 |
|
kados |
thd: you wanna work on that? ;-) |
17:36 |
|
thd |
yes |
17:36 |
|
kados |
actually it might not be too much work |
17:37 |
|
kados |
taking the LOC marc21 as an example |
17:37 |
|
kados |
thd++ |
17:37 |
|
paul |
+ http://www.ifla.org/IV/ifla64/138-161e.htm |
17:37 |
|
paul |
and maybe : http://www.ukoln.ac.uk/metadat[…]y/dc_unimarc.html |
17:37 |
|
kados |
cool |
17:37 |
|
slef |
when is 2.2.8 aimed for? |
17:39 |
|
paul |
just to point 1 important thing |
17:39 |
|
paul |
(that I should have pointed as "news") |
17:39 |
|
kados |
sure |
17:39 |
|
slef |
hdl: can we get installer-NG ready for it? |
17:39 |
|
paul |
we are waiting for our 4th boy, and he should be here for late feb/early march. |
17:39 |
|
kados |
right |
17:39 |
|
paul |
so in march, don't expect too many things from me |
17:40 |
|
slef |
paul: there's a 'brown paper bag' error in 2.2.7's installer |
17:40 |
|
hdl |
maybe. |
17:40 |
|
paul |
brown paper bag error ??? |
17:40 |
|
slef |
barcorde not barcode |
17:40 |
|
paul |
right slef. |
17:40 |
|
paul |
i also fixed a CRI bug in z3950 search today |
17:40 |
|
kados |
CRI? |
17:40 |
|
paul |
(when searching on title) |
17:40 |
|
paul |
critical |
17:40 |
|
kados |
ahh |
17:40 |
|
kados |
paul++ |
17:41 |
|
slef |
paul: mistake so embarassing that developer should wear a brown paper bag over head in shame |
17:41 |
|
paul |
(probably BLO for you in fact ;-) ) |
17:41 |
|
slef |
aha! paul++ |
17:41 |
|
kados |
hehe |
17:41 |
|
waylon |
paul++ ... was goingto look at that today... had my linux boxen set up, to go ZOOM. |
17:41 |
|
paul |
slef : ok, I understand. in french : "bonnet d'ane" |
17:41 |
|
slef |
2.2.7.1 anyone? |
17:42 |
|
waylon |
huh? guess we've done alot of bugfixing. |
17:42 |
|
paul |
would be better that a poorly tested 2.2.8, with new features like holidays |
17:43 |
|
paul |
in fact, the 68 commits I have to check are related to bug for around 20 of them, and improvements/new features for around 50 |
17:43 |
|
kados |
slef: i worry we've done too much between 2.2.7 and now |
17:43 |
|
slef |
so just 2.2.7 + paul's CRI + barcode + what else? |
17:43 |
|
kados |
slef: to do a mini release |
17:43 |
|
kados |
ahh, I see what you mean |
17:43 |
|
slef |
kados: I'll branch rel_2_2_7 and backport fixes if you want |
17:43 |
|
kados |
paul: that OK with you? |
17:43 |
|
paul |
yep. |
17:43 |
|
kados |
slef++ for offering |
17:44 |
|
kados |
slef+++ when it's done ;-) |
17:44 |
|
paul |
rch : I see you did some commit for mySQL5 fixes. |
17:44 |
|
paul |
is 2.2 cmopliant with mySQL5 now ? |
17:44 |
|
slef |
aye... sort of stuff that's been going on/going wrong this week, you never know ;-) |
17:44 |
|
rch |
haven't tested thoroughly |
17:44 |
|
waylon |
huh? So.... I add bugfixes to rel_2_2_7, new features to rel_2_2? |
17:44 |
|
rch |
but am running 2_2 on mysql5 |
17:45 |
|
slef |
waylon: no, add everything to rel_2_2 and email koha-devel about any blockers/CRIs for 2.2.7.1 please |
17:46 |
|
waylon |
ah. okay. so 2.2.7.1 won't have the new features, that right? |
17:46 |
|
kados |
so before we talk about rel_3_0 and the release of 3.0 |
17:46 |
|
paul |
waylon: right |
17:46 |
|
slef |
waylon: or flag it in bugs.koha.org if you want (as that's something I'll check) |
17:46 |
|
slef |
waylon: right... just fix the worst cockups |
17:46 |
|
kados |
can we talk about the future of 2.2? |
17:46 |
|
kados |
esp as it relates to the improvements that are in rel_3_0? |
17:46 |
|
kados |
we have two options I think: |
17:46 |
|
paul |
it's related to SQL version of rel3 or backport improvements to rel2_2? |
17:47 |
|
kados |
1. backport improvements gradually from rel_3_0 to rel_2_2 |
17:47 |
|
kados |
paul: yes |
17:47 |
|
kados |
2. stop support on rel_2_2 and create a SQL version of rel_3_0 |
17:47 |
|
paul |
I investigated a lot this question, can I explain ? |
17:47 |
|
kados |
paul: sure |
17:48 |
|
slef |
3. create a SQL version of rel_3 and /then/ stop support on rel_2_2 |
17:48 |
|
kados |
slef: yea, that's what I meant |
17:48 |
|
slef |
3. feature-freeze rel_2_2, create a SQL version of rel_3 and /then/ stop support on rel_2_2 |
17:48 |
|
kados |
slef++ for clarifying |
17:48 |
|
paul |
the problem with point 1 is that rel_3_0 has been cleaned a lot by toins. So it will be a pain, to backport things. |
17:49 |
|
paul |
the API changed in almost all .pm packages. |
17:49 |
|
paul |
for example, adding a biblio was : |
17:49 |
|
paul |
(3 parameters) = NEWnewbiblio(3 parameters) in 2.2 |
17:49 |
|
paul |
(2 parameters) = AddBiblio(3 parameters) in 3.0 |
17:49 |
|
paul |
the problem with the other possibility is that : |
17:50 |
|
paul |
- it should be "easy" to store a reverted index of records. |
17:50 |
|
paul |
i already have investigated that before we decided to go to zebra. |
17:50 |
|
kados |
what do you mean by 'reverted index of records'? |
17:51 |
|
paul |
we just had to improve "zebraop" to handle "indexing" of MARC records. |
17:51 |
|
paul |
the mac_words idea |
17:51 |
|
kados |
ahh, right |
17:51 |
|
paul |
marc_words idea |
17:51 |
|
paul |
the BIGGEST problem being : |
17:51 |
|
paul |
how to SEARCH with CCL/CQL/RPN in this db |
17:51 |
|
kados |
yea |
17:51 |
|
paul |
I don't have the answer to the question. |
17:51 |
|
kados |
that's tricky |
17:52 |
|
paul |
maybe we could use only a small subset of CCL for example |
17:52 |
|
paul |
considering that small libraries means smaller features. |
17:52 |
|
paul |
but i'm not 100% happy with this idea. |
17:52 |
|
waylon |
"CCL/CQL/RPN? |
17:52 |
|
kados |
one more problem |
17:52 |
|
kados |
is that almost all koha libraries, hundreds of them, use rel_2 |
17:52 |
|
slef |
waylon: zebra-helpful query languages |
17:52 |
|
kados |
rel_2_2 I mean |
17:53 |
|
paul |
waylon: CCL/CQL/RPN are the 3 languages we can use to query a zebra DB. |
17:53 |
|
kados |
it's critical that we don't leave them hanging |
17:53 |
|
kados |
and the process of upgrading from 2.2.x to 2.2.y works fairly well |
17:53 |
|
thd |
paul: had you not told me that option 2 would be relatively easy before kados appeared? |
17:54 |
|
kados |
I fear upgrading from 2.2.x to 3.0_nozebra could be difficult to implement without a long testing period |
17:54 |
|
toins |
thd, i think so too |
17:54 |
|
paul |
thd : no, I just spoke of the 1st part of the job. |
17:54 |
|
kados |
toins: how long did you work on improving the API in rel_3? |
17:54 |
|
paul |
kados arrived just before I spoke ot the 2nd pard. |
17:54 |
|
kados |
toins: was it about two months? |
17:55 |
|
toins |
humm... yep 2 months i think |
17:55 |
|
paul |
kados: something like that. Maybe 3 months in fact. |
17:55 |
|
thd |
kados: have your version 2 to 3 migrations been going any better lately? |
17:55 |
|
kados |
thd: we've only done one of those |
17:55 |
|
kados |
thd: and its' a real pain! |
17:55 |
|
paul |
2 to 3 or dev_week ? |
17:55 |
|
slef |
the cleaned api seems miles better. no contest to me. option 2 and crack the problems |
17:55 |
|
thd |
kados: you do not have enough practise :) |
17:56 |
|
kados |
sorry, right |
17:56 |
|
kados |
2.2 -> dev_week |
17:56 |
|
paul |
because we added many things to rel_3_0 to ease the process. |
17:56 |
|
kados |
we don't have any rel_3_0 systems in production |
17:56 |
|
kados |
just dev_week |
17:56 |
|
cm |
yeah, please tell me dev_week -> 3_0 won't be too bad! ;) |
17:56 |
|
paul |
and hdl & me already tried on 4 different DB (although not in production), and it's not a pain. |
17:56 |
|
kados |
paul: we're still testing rel_3_0 |
17:56 |
|
paul |
it's just a nightmare now ;-) |
17:57 |
|
kados |
right |
17:57 |
|
kados |
cm: yea, don't worry |
17:57 |
|
cm |
good. :) |
17:57 |
|
kados |
cm: that shouldn't be very hard, the hard part is going to zebra in the first place |
17:57 |
|
cm |
yes, indeed! |
17:57 |
|
paul |
hdl : maybe it's time to show up our new surprise ? |
17:57 |
|
hdl |
rebuildzebra you did ? |
17:58 |
|
thd |
paul: you mean to say that 'a nightmare' is less painful than 'a pain' |
17:58 |
|
thd |
? |
17:58 |
|
paul |
yep. |
17:58 |
|
kados |
hey russ |
17:58 |
|
paul |
no, web installer. but it's not time, wait a minut ;-) |
17:58 |
|
russ |
hello all |
17:58 |
|
kados |
a web installer? |
17:58 |
|
hdl |
hi russ. |
17:58 |
|
waylon |
hiya russ. |
17:58 |
|
paul |
hi new liblimer. |
17:58 |
|
kados |
paul: not just yet :-) |
17:58 |
|
kados |
paul: April 1st if all goes smoothly :-) |
17:59 |
|
paul |
ah, I understand, it's a joke in fact... |
17:59 |
|
kados |
:-) |
17:59 |
|
paul |
I almost thought it was the truth... |
17:59 |
|
kados |
russ: the cat's out of the bat |
17:59 |
|
paul |
libLime conquering the world !!! |
17:59 |
|
kados |
russ: s/bat/bag/ |
17:59 |
|
toins |
hehe |
17:59 |
|
kados |
russ: how was the cricket? |
18:00 |
|
russ |
good fun even though the results didnt got the way we wanted |
18:00 |
|
kados |
bummer |
18:00 |
|
russ |
:-) |
18:00 |
|
slef |
kados: swine |
18:00 |
|
kados |
hehe |
18:00 |
|
paul |
so, back to 2.2.7/8 questions ? |
18:01 |
|
kados |
sure |
18:01 |
|
slef |
kados: rooting used to be funny to aussies |
18:02 |
|
kados |
hehe |
18:02 |
|
kados |
so I guess we don't have any strong direction |
18:02 |
|
kados |
for what to do regarding the 2.2 vs 3.0 prob |
18:02 |
|
slef |
the cleaned api seems miles better. no contest to me. option 2 and crack the problems |
18:02 |
|
rch |
(cleaned api)++ |
18:03 |
|
paul |
so, I may investigate (& write something on the wiki) my idea. |
18:03 |
|
kados |
paul++ |
18:03 |
|
paul |
to see what kind of limits it will have. |
18:03 |
|
paul |
and see if they are acceptable or have to be solved. |
18:03 |
|
kados |
I'm behind a SQL version of 3.0, but we will need to make sure 2.2->3.0_nozebra upgrade is seamless |
18:03 |
|
kados |
excellent |
18:04 |
|
kados |
so the next and last item is 3.0 itself |
18:04 |
|
thd |
paul: given your thoughts for option 2 would the primary record be the XML record stored as whole instead of MARC in SQL? |
18:04 |
|
waylon |
so.. basically.. import all the SQL access/update stuff into 3.0? |
18:04 |
|
waylon |
marc_word tables etc? |
18:04 |
|
hdl |
thd : if you do so, how can you index ? |
18:05 |
|
slef |
thd: can't mysql handle xml yet? |
18:05 |
|
paul |
i'll explain my idea deeper on the wiki. |
18:05 |
|
kados |
slef: dont think so |
18:05 |
|
paul |
slef: I don't think either |
18:05 |
|
kados |
slef: at least not fielded searches |
18:05 |
|
kados |
slef: maybe full text on the whole document |
18:05 |
|
kados |
even that I'm not sure about |
18:05 |
|
kados |
paul++ |
18:05 |
|
waylon |
so marc_word is still the best SQL way of searching. |
18:05 |
|
thd |
hdl: you could still have the nonmarc fields for indexing in SQL |
18:06 |
|
kados |
so the next and last item is 3.0 itself |
18:06 |
|
dewey |
i already had it that way, kados. |
18:06 |
|
slef |
Beginning with MySQL 5.1.5, two functions providing basic XPath (XML Path Language) capabilities are available. |
18:06 |
|
hdl |
thd : it is not enough |
18:06 |
|
kados |
as everyone probably knows, we're about a year behind our original estimates for the release date :-) |
18:06 |
|
thd |
hdl: I was referring to where the record is taken for display and editing not indexing |
18:06 |
|
kados |
slef: ahh, so that's worth looking at |
18:07 |
|
thd |
kados: welcome to the real world of apparent work versus actual work required :) |
18:07 |
|
kados |
slef: esp if we switched to XPath in zebra |
18:07 |
|
kados |
slef: I wonder how basic they are |
18:07 |
|
rch |
xpath is much slower, yes? |
18:07 |
|
rch |
(in zebra) |
18:07 |
|
paul |
so, we speak of 3.0 or still 3.0_nonzebra ? |
18:07 |
|
slef |
kados: I expect the world will be mysql 5.1 by the time koha 3.0sql is ready |
18:07 |
|
slef |
http://dev.mysql.com/doc/refma[…]ml-functions.html |
18:08 |
|
kados |
paul: so far 30=.0_nonzebra |
18:08 |
|
kados |
oops |
18:08 |
|
kados |
3.0_nonzebra I meant |
18:08 |
|
paul |
it seems ;-) |
18:08 |
|
thd |
rch: XPath is much slower for indexing but should be comparably fast to no XPath after indexing |
18:09 |
|
thd |
rch: unless someone has tests showing otherwise |
18:09 |
|
kados |
slef: so xpath might solve part of the problem, but we still will need to re-implement one or all of the query languages (CCL, CQL, PQF/RPN) if we want a consistant search API |
18:10 |
|
rch |
thd: yes probably just indexing |
18:10 |
|
paul |
kados: right |
18:10 |
|
kados |
so can we move to discussing 3.0 itself? |
18:10 |
|
slef |
kados: aye... I think it's more probable someone develops a query language -> XPath mapping for us |
18:10 |
|
slef |
if we're lucky ;-) |
18:10 |
|
kados |
yea, that would rock |
18:11 |
|
paul |
(/me starts training for the 4 voting days we will have soon ;-) ) |
18:11 |
|
hdl |
(Indexdata may have worked on that already. |
18:11 |
|
kados |
so there's a new installer for 3.0? |
18:11 |
|
paul |
yep... |
18:11 |
|
kados |
hdl: added to 3.0 section ? |
18:11 |
|
kados |
good news! |
18:11 |
|
dewey |
good news is that search?q=callnum works now. :) |
18:11 |
|
slef |
I can ask a friend and see whether we can get a grant to fund a student to hack on a query language to xpath converter |
18:11 |
|
waylon |
OS independant web installer? |
18:11 |
|
paul |
slef: ++ |
18:12 |
|
paul |
hdl, throw the URL ;-) |
18:12 |
|
hdl |
i12.hdlaurent.paulpoulain.com/cgi-bin/koha/install/install.pl |
18:12 |
|
hdl |
password and login you can get in your emila boxes. |
18:12 |
|
slef |
waylon: there'll be a system-semi-dependent perl installer for doing the sysadmin install... then hdl's super web installer takes over for web and database |
18:12 |
|
paul |
(you should prefill them in the template for the demo) |
18:13 |
|
hdl |
(slef knows it) |
18:13 |
|
slef |
(I've a problem with the perl installer... it keeps putting the CGI scripts in /usr/lib/site-perl, which isn't ideal) |
18:13 |
|
mason |
hiya everyone |
18:14 |
|
hdl |
So you have the link ? |
18:14 |
|
tnb |
hey mason :) |
18:14 |
|
hdl |
It is a web installer for all the data and datasets populating in Mysql. |
18:14 |
|
slef |
and language selection |
18:15 |
|
kados |
hdl: so the installer is just two steps? :-) |
18:15 |
|
paul |
+ create the DB itself |
18:15 |
|
hdl |
It uses sqldumps user can make. |
18:15 |
|
hdl |
3. |
18:15 |
|
hdl |
1) languageselection + module verification. |
18:15 |
|
hdl |
throws a warning to librarian if not OK. |
18:16 |
|
kados |
cool |
18:16 |
|
paul |
(the step 3 is : http://i12.hdlaurent.paulpoula[…]=selectframeworks and nothttp://i12.hdlaurent.paulpoula[…]ectframeworks%22) |
18:16 |
|
hdl |
2) looks mysqlconnection and database structure. |
18:16 |
|
kados |
is this committed? |
18:16 |
|
hdl |
slef told me. Had no chance to correct this. |
18:16 |
|
hdl |
kados not yet. |
18:16 |
|
hdl |
Still working on it. |
18:16 |
|
kados |
it's really great! |
18:17 |
|
hdl |
BUT |
18:17 |
|
hdl |
kados and all ther is a BUT. |
18:17 |
|
paul |
(on step 3, select "french" to see what it will be done for) |
18:17 |
|
kados |
hdl: what's the but :-) |
18:17 |
|
hdl |
Now sqldatas should be in misc/sqldata/ |
18:18 |
|
hdl |
Now sqldatas should be in misc/sql-data/(en| fr)/ <your level> |
18:18 |
|
kados |
ahh |
18:19 |
|
hdl |
your level beeing required/recommended or additionals or any other name you like. |
18:19 |
|
hdl |
It is not hardcoded. |
18:19 |
|
thd |
paul: what is the login for that page? |
18:19 |
|
kados |
hdl: what is 'level'? |
18:20 |
|
hdl |
required : for datas taht are necessary for Koha to Work. |
18:20 |
|
hdl |
recommended : things that will highly ease the work of librarian. |
18:21 |
|
hdl |
additionals would be any fancy framework or list for instance CD frameworks.. |
18:21 |
|
kados |
right |
18:21 |
|
paul |
or a subset to setup a classic public library |
18:21 |
|
hdl |
in the level directory, you have sql and text files. |
18:21 |
|
paul |
(like patron categories, itemtypes...) |
18:21 |
|
paul |
some authorized values |
18:21 |
|
paul |
... |
18:22 |
|
hdl |
sql to import data to mysql base. |
18:22 |
|
hdl |
and text to describe them. |
18:22 |
|
hdl |
French language version of this page is more explicit than english atm. |
18:23 |
|
hdl |
english I had no data to put in. |
18:23 |
|
hdl |
I didnot know which framework to propose to persons for MARC21. |
18:23 |
|
hdl |
(which was the good one. |
18:23 |
|
hdl |
) |
18:23 |
|
thd |
hdl: all I see is an authentication form. |
18:24 |
|
hdl |
in agogme.com you have login and password. |
18:24 |
|
hdl |
kados ; same for you. |
18:24 |
|
hdl |
Cookies must be enabled. |
18:25 |
|
thd |
hdl: no every hidden directory on my site is open if you know where it is :) |
18:25 |
|
hdl |
(your email box) |
18:25 |
|
hdl |
kohaadmin |
18:25 |
|
kados |
hdl: yep, I got it |
18:26 |
|
thd |
hdl: well, authentication is required there |
18:26 |
|
hdl |
Normally at the end. When install is finished. |
18:26 |
|
kados |
so the installer looks great! |
18:26 |
|
kados |
I'm excited about seeing the code when it's done :-) |
18:26 |
|
kados |
great job hdl |
18:26 |
|
hdl |
I thought about chmodding dirr to 000 and getting to mainpage.pl |
18:27 |
|
paul |
the other part of koha setup being setting config file & virtual host can easily be put in a .deb / .rpm I think |
18:27 |
|
thd |
hdl: I trust that it looks great even if I cannot see for myself? |
18:27 |
|
kados |
paul: great idea |
18:28 |
|
kados |
slef: isn't threre already work on a .deb for 2.2.8? |
18:28 |
|
paul |
that could be slef job + I had a french candidate from Debian to create a .deb |
18:28 |
|
paul |
kados: right |
18:28 |
|
kados |
there remains 'what to do with rel_3 and HEAD' |
18:28 |
|
paul |
kados : no |
18:28 |
|
kados |
my opinion is to delete HEAD and replace with rel_3_0 :-) |
18:29 |
|
paul |
we havent spoke of rel_3_0 status yet ! |
18:29 |
|
kados |
paul: oops, did I jump the gun? :-) |
18:29 |
|
thd |
:) |
18:29 |
|
paul |
here in France, we have SAN-OP going live with 3.0 |
18:29 |
|
kados |
ahh, two items under rel_3_0 status were hdl's installer and what to do with HEAD & rel_3_0 |
18:30 |
|
paul |
but they don't use some features like acquisitions. |
18:30 |
|
kados |
right |
18:30 |
|
paul |
we plan to setup a 3.0 at IPT. |
18:30 |
|
kados |
there is still a lot of testing that must be done before rel_3_0 can become 3.0 IMO |
18:30 |
|
kados |
rch has been testing on our end |
18:30 |
|
kados |
rch: wanna speak to that? |
18:30 |
|
paul |
right. the question being : who works on those tests ;-) |
18:31 |
|
rch |
i've been working mostly on dev_week |
18:31 |
|
rch |
t/ ++ |
18:31 |
|
hdl |
slef++ |
18:31 |
|
kados |
paul: it's the classic prob with OSS library software: librarians don't want to pay for testing software :-) |
18:31 |
|
slef |
kados: yes, but it really needs the web installer and new perl installer |
18:31 |
|
paul |
I have librarians that don't want to pay for testing, but are OK to test if they don't pay ;-) |
18:31 |
|
kados |
I have just resolved myself that 3.0 will be ready when it's ready |
18:31 |
|
kados |
hehe |
18:31 |
|
slef |
kados/paul: can we change installer for 2.2.8? |
18:32 |
|
paul |
slef : maybe. |
18:32 |
|
slef |
if it's ready, of course |
18:32 |
|
kados |
right |
18:32 |
|
paul |
what hdl is doing is not specifically related to 3.0 |
18:32 |
|
waylon |
why do we have two virtual hosts, when they coexist on one server happily? |
18:32 |
|
thd |
rch: what has happened with the general purpose test suite Katipo was developing for company wide use? |
18:32 |
|
slef |
waylon: dunno... would help many people to merge them IMO |
18:32 |
|
kados |
thd: we'd have to ask russ or mason I think |
18:32 |
|
rch |
you mean rach? |
18:33 |
|
hdl |
(some of the modules are 3.0 specific) |
18:33 |
|
toins |
becareful t/ directory can only test module function but not all the scripts individually |
18:33 |
|
waylon |
slef: ive already done that.. |
18:33 |
|
slef |
toins++ |
18:33 |
|
kados |
toins: good point |
18:33 |
|
thd |
rch: sorry, yes I Imagined that you were rach :) |
18:33 |
|
rch |
:) |
18:33 |
|
waylon |
check out wayrob.freehostia.com |
18:33 |
|
waylon |
cardnumber: pntskoha ... password.. the usual default. |
18:34 |
|
thd |
russ: are you still here? |
18:34 |
|
rch |
thd: i hope liblime will be adding to the t/ dir |
18:34 |
|
russ |
in spirit only :-) |
18:35 |
|
slef |
; cvs -d:pserver:anonymouswayrob.freehostia.com:/ checkout koha |
18:35 |
|
slef |
cvs [checkout aborted]: connect to wayrob.freehostia.com(64.72.112.20):2401 failed: Connection refused |
18:35 |
|
thd |
russ: what has happened with the general purpose test suite Katipo was developing for company wide use? |
18:35 |
|
waylon |
actually.. password not theusual default.... |
18:35 |
|
waylon |
slef: cvs? don't have it. |
18:35 |
|
slef |
waylon: so how do I checkout it? |
18:36 |
|
waylon |
slef: oh.... right.. you want a download... |
18:36 |
|
thd |
russ: chris had mentioned that a test suite was being developed at Katipo during devel week |
18:36 |
|
russ |
we trialled a few things, but nothing really worked out |
18:36 |
|
slef |
or just an instruction? is it a simple 'merge the two VirtualHost'? |
18:36 |
|
slef |
<waylon> check out wayrob.freehostia.com |
18:36 |
|
paul |
toins tried to work on /t, but we concluded that we could not do anything really interesting. |
18:37 |
|
hdl |
slef : waylon wanted to say look at |
18:37 |
|
paul |
because we can't test .pl, that are 90% of our code. |
18:37 |
|
paul |
kados : you said that 3.0 will be release when it will be ready. what do you plan to do to have it ready ? |
18:37 |
|
slef |
why can't we test .pl? |
18:38 |
|
toins |
and we can't test with multiple client connected to koha |
18:38 |
|
hdl |
what about selenium ? |
18:38 |
|
toins |
(to discover some mod_perl specific bugs) |
18:38 |
|
paul |
slef : because the result is a html page, that is hard to say wether it succeded or failed. |
18:38 |
|
kados |
paul: well for one, chris will be working on Koha full time soon |
18:38 |
|
waylon |
slef: its basically... put opac/ into the main cgi-bin/koha. put cgi-bin/koha, just the koha directory, into cgi-bin of the server. |
18:38 |
|
hdl |
http://www.openqa.org/selenium/ |
18:39 |
|
slef |
paul: we could perl -c (syntax check) them maybe? |
18:39 |
|
kados |
paul: last time we investigated rel_3_0 it wasn't as stable as dev_week |
18:39 |
|
slef |
but I know much will have to wait until koha is configured |
18:39 |
|
kados |
paul: but a lot has been fixed since then |
18:39 |
|
hdl |
perl -c is a first step. |
18:39 |
|
waylon |
slef: Determine if you can set your PERL5LIB.. if not, you'll have to do a find and xarg, to set 'use lib' |
18:39 |
|
paul |
guys... we are speaking of 4 things at the same time, and it's quite confusing... |
18:39 |
|
hdl |
But it is usually OK. |
18:39 |
|
waylon |
oh.. sorry paul.. |
18:40 |
|
slef |
paul: flame the chairman ;-) |
18:40 |
|
kados |
hehe |
18:40 |
|
kados |
paul: so the question was 'what will be done for rel_3_0 to be ready for 3.0', right? |
18:40 |
|
paul |
kados, what do we speak about now ? |
18:40 |
|
paul |
yep. |
18:41 |
|
paul |
I see 3 differents things to work on : |
18:41 |
|
paul |
- installer |
18:41 |
|
kados |
as I see it we need 1. to test every component; 2. to make the install easier; 3. make migration scripts work better |
18:41 |
|
paul |
- testing & bugfixing (if we consider we are feature freeze, which is not official) |
18:41 |
|
paul |
- default CSS for OPAC & intranet. |
18:41 |
|
paul |
kados : right |
18:41 |
|
paul |
point 4 : migration |
18:42 |
|
kados |
hehe |
18:42 |
|
kados |
I'd say that's what needs to be done |
18:42 |
|
hdl |
kados: point 1 we have done to a certain extent with our day in Paris Jan 22. |
18:42 |
|
slef |
2. should be solved by hdl's great work and forward-porting whatever we do for 2.2.8 |
18:42 |
|
kados |
yep |
18:43 |
|
hdl |
3 paul worked on It. |
18:43 |
|
hdl |
and provided good bases. |
18:43 |
|
hdl |
That still need some work on. |
18:43 |
|
kados |
so I will set some time for LibLime staff to do internal testing of rel_3_0 and we will post bugs to bugs.koha.org |
18:43 |
|
paul |
(which 3 ? mine or kados one ?) |
18:43 |
|
hdl |
kados' one |
18:44 |
|
paul |
ok, then i'm OK. |
18:44 |
|
kados |
we will await hdl's new installer with great anticipation :-) |
18:44 |
|
paul |
does anyone have any suggestion for point 4 (stylesheets) ? |
18:44 |
|
hdl |
But default css would be a great thing too. |
18:44 |
|
kados |
doesn't san have some? |
18:44 |
|
waylon |
suppose i could start installing rel_3_0.... can it run under windows as well as rel_2_2? |
18:44 |
|
paul |
yep, but I think they are quite basic & could be improved. |
18:45 |
|
kados |
owen is on maternity leave |
18:45 |
|
paul |
maternity leave ? |
18:45 |
|
kados |
but I could nominate him to work on it in his absense :-) |
18:45 |
|
kados |
paul: new baby |
18:45 |
|
paul |
a baby too ? |
18:45 |
|
kados |
yep :-) |
18:45 |
|
paul |
already born ? |
18:45 |
|
kados |
yep |
18:45 |
|
kados |
about a week ago I think |
18:45 |
|
kados |
i was in NZ, so don't know exact date |
18:45 |
|
hdl |
i think so |
18:45 |
|
paul |
wow, there are many babies in koha team... |
18:46 |
|
slef |
paul: stop assigning bugs to them |
18:46 |
|
paul |
lol |
18:46 |
|
paul |
when I see liblime demo opac, I think he could do a great job at designing "official" OPAC. |
18:47 |
|
kados |
so I will talk to him about that when I return on the 15th |
18:47 |
|
paul |
I don't expect koha 3.0 before this date :-D |
18:47 |
|
slef |
aw |
18:48 |
|
kados |
hehe |
18:48 |
|
kados |
anyone have any other points to discuss? |
18:48 |
|
waylon |
kados: any reason why we have two virtual hosts? |
18:48 |
|
paul |
yes, a last one : do we declare feature freeze on rel_3_0 branch ? |
18:48 |
|
kados |
waylon: lets discuss that after the meeting |
18:49 |
|
paul |
waylon: reminder : it's 11PM here in france ;-) |
18:49 |
|
rch |
are we still hitting rel_3 v HEAD |
18:49 |
|
kados |
rch: good point |
18:49 |
|
rch |
if there's a feature freeze, we need a head |
18:50 |
|
paul |
tumer is not here, but I think branching head => tumer and rel_3_0 => head is a good point. |
18:50 |
|
paul |
tumer is not here, but I think branching head => tumer and rel_3_0 => head is a good idea. |
18:50 |
|
rch |
that makes sense to me |
18:50 |
|
waylon |
Tumer? |
18:50 |
|
dewey |
i heard Tumer was having lots of authentication problems |
18:50 |
|
paul |
waylon: tumer is a guy from Cyprus |
18:50 |
|
thd |
kados: why would you ever delete head? |
18:51 |
|
paul |
thd : not delete head. |
18:51 |
|
slef |
paul: france needs to move to english time ;-) |
18:51 |
|
kados |
paul: yea |
18:51 |
|
paul |
but say that tumer version is not head, but a specific one. |
18:51 |
|
kados |
yep |
18:51 |
|
paul |
and say rel_3_0 is the official one, so in head. |
18:51 |
|
kados |
paul++ |
18:52 |
|
thd |
paul: Was branch head what was meant by kados? |
18:52 |
|
waylon |
ahh.. like Debian freezes? |
18:52 |
|
paul |
thd : I think so |
18:52 |
|
thd |
paul: you understand English better than I do :) |
18:53 |
|
hdl |
tumer could not commit on head for quite a long time now. |
18:53 |
|
hdl |
But there are killer fetures in it. |
18:53 |
|
paul |
thd : I already spoke of this with kados, so it's not my english :-) |
18:53 |
|
hdl |
He has problems with mod_perl under Windows. |
18:54 |
|
thd |
hdl: what is preventing him from committing? |
18:54 |
|
hdl |
he wnated to ask some more information to kados. |
18:54 |
|
hdl |
Maybe an auhtentication problem. |
18:55 |
|
hdl |
on savannah |
18:55 |
|
hdl |
or is it windows related. |
18:56 |
|
hdl |
tumer said he still waited for us in Cyprus :P |
18:56 |
|
paul |
(in my bedroom :-D ) |
18:56 |
|
kados |
hehe |
18:56 |
|
kados |
so lets close the meeting |
18:56 |
|
hdl |
So do I. |
18:56 |
|
kados |
and I'll make sure to schedule another one soon |
18:56 |
|
kados |
within a couple of weeks |
18:57 |
|
hdl |
Thx |
18:57 |
|
thd |
kados; so is it to late to discuss wiki authentication problems today? |
18:57 |
|
paul |
I'll be here (even if baby is here) |
18:57 |
|
rch |
perhaps more discussion of HEAD on devel list? |
18:57 |
|
paul |
rch ++ |
18:57 |
|
kados |
thd: unfortunatly yes, today is super busy |
18:57 |
|
kados |
thd: maybe write a mail to koha-devel? or to me privately? |
18:57 |
|
waylon |
someone ableto tell me why we use two virtual hosts? |
18:57 |
|
kados |
thd: I won't have time till probably fri or sat to work on it though :( |
18:58 |
|
paul |
rch / kados i'll write something on koha-devel about tumer/head/rel_3_0 branches |
18:58 |
|
paul |
(tomorrow morning) |
18:58 |
|
thd |
kados: what is happening with the Copyright assignment for Koha with respect to Katipo and LibLime Ltd? |
18:59 |
|
rch |
paul: great |
18:59 |
|
paul |
kados : you'll write something on koha-devel about other topics ? |
18:59 |
|
kados |
thd: info is forthcoming on that |
18:59 |
|
kados |
paul: yes |
18:59 |
|
paul |
ok, great |
18:59 |
|
paul |
oh !!!! |
18:59 |
|
paul |
I missed a VERY IMPORTANT NEWS !!! |
19:00 |
|
thd |
kados: I need more deailed diagnostic information for the wiki problem because I have not seen it myself |
19:00 |
|
paul |
the french NPO should be born next month. |
19:00 |
|
thd |
slef: are you here? |
19:00 |
|
toins |
NPO ? |
19:00 |
|
paul |
ENSMP + sociology + me had the 1st meeting, we have signed the papers. |
19:00 |
|
paul |
(Non Profit Organization) |
19:00 |
|
toins |
ok |
19:01 |
|
paul |
ENSMP should have transmitted them to governement |
19:01 |
|
paul |
we should have the feedback in something like 1 month. |
19:01 |
|
kados |
wow, that's great paul! |
19:01 |
|
paul |
with the feedback, the "kohala" will be a reality. |
19:01 |
|
kados |
w00t |
19:01 |
|
paul |
kohala = "Koha Libre Association" |
19:01 |
|
thd |
paul: if johnb does not do anything I nominate the French organisation to run the international association |
19:02 |
|
russ |
great news paul |
19:02 |
|
russ |
thd: small steps small steps |
19:02 |
|
paul |
i'll let you know whent it will be official, of course. |
19:02 |
|
thd |
russ: US grants are wasting away :( |
19:03 |
|
paul |
once the association exist, we will have our 1st public meeting, with everybody that want to join the association. |
19:03 |
|
paul |
ENSMP is very very experienced in such structures. |
19:04 |
|
paul |
so I'm confident we made something very sure for the future |
19:04 |
|
rch |
sounds good, paul! |
19:04 |
|
paul |
now, I can go to bed ;-) |
19:04 |
|
thd |
slef: are you still awake? |
19:04 |
|
paul |
bye world |
19:04 |
|
toins |
bye all |
19:05 |
|
thd |
good night France |
19:05 |
|
rch |
g'night, France |
19:05 |
|
cm |
bye all |
19:05 |
|
thd |
goodbye Ohio |
19:06 |
|
thd |
slef: I am still hoping you have not gone to sleep |
19:06 |
|
waylon |
so thats it? I can ask my two virtualhost question now? |
19:06 |
|
thd |
waylon: yes but your audience is now much smaller |
19:08 |
|
thd |
waylon: what are your virtual host questions? |
19:08 |
|
waylon |
anyway... ive managed, without support from the sysadmin at all, to put koha into a userspace enviroment, with no CPAN access, no PERL5LIB access, no virtualhosts (except for the subdomain that they gave me) .. no logs (CGI:Carp is your friend) ... |
19:09 |
|
waylon |
So.. in other words... a extremely constricted enviroment. |
19:09 |
|
waylon |
no ScriptAlias. |
19:10 |
|
waylon |
The question is.. why do we use two virtualhosts in the default koha? |
19:11 |
|
thd |
waylon: one is for the OPAC which library readers use and the other is for administration which only librarians use. |
19:12 |
|
rch |
waylon: i think it just makes sense. two interfaces, two interfaces |
19:12 |
|
chris |
thats not enough for most places |
19:12 |
|
waylon |
no? |
19:12 |
|
chris |
most places have the intranet on another ipnumber entirely |
19:13 |
|
chris |
with firewall rules in place |
19:13 |
|
chris |
at least ones ive been involved with do |
19:14 |
|
chris |
not really |
19:14 |
|
thd |
waylon: if the address of the librarian interface is maximally secure it is probably also not widely distributed as an extra measure |
19:14 |
|
chris |
you might want ot put the intranet under https too |
19:15 |
|
chris |
s/intranet/librarian interface/ |
19:15 |
|
slef |
thd: no. phone again. |
19:15 |
|
thd |
slef: ping me when you are off phone |
19:17 |
|
waylon |
what could someone do, with a intranet, without a password, to comprimise it? |
19:18 |
|
thd |
waylon: they could attempt to guess a password for one |
19:20 |
|
waylon |
thd: hmm.. secure passwords, not less than x characters.... adding of some password rules, like having compulsory requirement to have a few numbers.... mixed case... |
19:22 |
|
rch |
then the sysadmin constantly gets called to reset people's passwords :) |
19:22 |
|
rch |
people are bad with passwords. |
19:23 |
|
rch |
always better to have more security in place. |
19:23 |
|
chris |
waylon: there is no reason to open the librarian interface up to the whole internet |
19:24 |
|
thd |
waylon: they could try devious means to gain access to the server. Security does not require obscurity but it does help. There is no reason that Koha should be a real target for a cracking attempt but that is no reason to be complacent and therefore the most visible target to some cracking system wanting to make mischief in libraries |
19:25 |
|
slef |
thd: ping |
19:25 |
|
thd |
waylon: storing backups in a safe place is probably more important than passwords |
19:26 |
|
slef |
waylon: constraining the password reduces the search space for attackers... remember enigma? |
19:26 |
|
thd |
slef: you may have missed a couple of questions I asked when you were tripping over things |
19:26 |
|
thd |
slef: I am trying to diagnose the wiki authentication problem |
19:26 |
|
waylon |
ah.. adding rules, means the attackers have more rules to enter into their password generators? |
19:27 |
|
thd |
slef: what browser were you using when you had the permission denied message? |
19:27 |
|
slef |
thd: I suspect it's browser config. It's an iceweasel, default policy is to deny everything (and blame the french). |
19:28 |
|
slef |
thd: I don't know what settings are needed by the wiki - could you list them on the login screen? |
19:28 |
|
thd |
slef: so you do not accept cookies by default |
19:28 |
|
waylon |
slef: Okay. in essense, having koha in such a constricted enviroment, as a production basis, is not good. |
19:29 |
|
slef |
thd: no, nor execute scripts, allow new windows, load images ... |
19:29 |
|
thd |
slef: I am not an admin of that system but I have modified dokuwiki templates to good effect in the past |
19:30 |
|
waylon |
So.... I just put the intranet host, in another subdomain, and obscurity of the subdomain plus passwords are better than just passwords alone? |
19:30 |
|
slef |
waylon: I have no qualms with it, if it's all that we have. I think I'd want SSL on librarian if we can but may still be neater to merge the VirtualHost sections and SSLRequireSSL the librarian interface. |
19:31 |
|
thd |
slef: I noticed to my disappointment that dokuwiki uses some buggy javaScript in the editor |
19:31 |
|
waylon |
slef:who you setting up with koha? |
19:32 |
|
slef |
waylon: obscurity of the subdomain is probably worse than having both on one host... if someone discovers the hostname for the librarian pages, http://librarianhostname/ redirects them to the admin login... if it's on the same hostname as the OPAC, it's less obvious IMO |
19:32 |
|
slef |
waylon: some academics |
19:33 |
|
chris |
it wasnt obscurity i was suggesting, just a different ip range for a start, one not externally routed |
19:33 |
|
slef |
that has reason |
19:34 |
|
waylon_ |
btw... you don't even need any virtualhost entries, if you want koha running from the main hostname. |
19:34 |
|
thd |
slef: my suspicion is that the wiki authentication problems others have had on the wiki are browser, JavaScript, and cookie configuration specific problems. I have made many posts to a small set of pages without any problem so I have been left out of the fun. |
19:35 |
|
slef |
waylon_: just a SetEnv PERL5LIB in .htaccess files? |
19:36 |
|
waylon_ |
slef: Dont even need that. A find | xarg line can put use lib '<path' statements into all your pl pm files. |
19:36 |
|
slef |
waylon_: SSLRequireSSL in .htaccess... as long as your web host gives you a modssl server configured by default. |
19:37 |
|
slef |
waylon_: modding all the files seems a little messy. |
19:38 |
|
waylon_ |
as far as moddile allthe files.. not messy. One, well.. two lines. Doesn't take long to process, on my p3 box. 192 mb ram. |
19:39 |
|
waylon_ |
450mhz i think... |
19:45 |
|
waylon |
if you assume that all p* files have use strict; in them.. then this line works. |
19:46 |
|
waylon |
find -name "*.p*" | xargs perl -pi -e "s/(use strict\;)/use strict\;\nuse lib \'<pathtoC4>'\;\n/gi"; |
19:46 |
|
waylon |
or is that "*.p?".. |
19:51 |
|
slef |
asleep at the prompt, sorry |
19:51 |
|
slef |
bbl |
19:51 |
|
waylon |
eh.. night then, slef. |
20:45 |
|
tnb |
chris: you still around? |
20:50 |
|
chris |
yep |
22:58 |
|
tnb |
chris: ? |
22:58 |
|
chris |
yo |
22:58 |
|
tnb |
oh good :) |
22:58 |
|
tnb |
hi :) |
22:58 |
|
tnb |
I need to get ahold of josh ASAP... you seen him? |
22:58 |
|
tnb |
he's not answering his phone |
22:59 |
|
chris |
hmmm ill ask russ |
22:59 |
|
tnb |
thanks, sorry for the bother |
22:59 |
|
chris |
no worries |
23:00 |
|
chris |
he's probably in starbucks :) |
23:00 |
|
chris |
ahh no, russ tells me he's on the bus to my house |
23:01 |
|
tnb |
ok, thanks :) |
23:01 |
|
chris |
so when he gets here ill let him know you are after him |
23:02 |
|
tnb |
ok, thanks |
23:02 |
|
tnb |
:) |