IRC log for #koha, 2005-10-27

All times shown according to UTC.

Time S Nick Message
11:23 osmoze bye
17:02 kados chris: any thoughts on Paul's koha-devel email?
17:02 kados rach, russ ?
17:02 russ hi kados
17:03 kados mornin ;-)
17:03 russ can someone forward that to me again please?
17:03 russ what is it evening for you?
17:03 chris thats the template email kados?
17:04 paul I can stay for a few minuts before going to bed if you want to speak of it kados.
17:04 chris forwarded it russ
17:04 russ cheers chris
17:04 paul hello chris & russ & kados
17:04 kados hehe
17:05 kados evening paul
17:05 kados I see you perked right up ;-)
17:05 paul perked ?
17:05 rach help
17:05 rach erg
17:05 kados perked = became alert ;-)
17:05 rach hello :-)
17:05 rach freudian slip there :-)
17:06 kados hehe
17:06 russ does someone have koha up and going with owen's templates somewhere?
17:06 kados owen: still around?
17:06 russ i havent seen them
17:06 kados I don't ...
17:06 kados sorry
17:06 kados I think paul's problem is that they are not user friendly enough for a programmer right paul?
17:06 paul i was just wandering near my computer when I heard the irc noise that occurs when someone type my name !
17:06 kados for testing and such?
17:07 kados hehe
17:07 paul right.
17:07 paul most pages are dead end.
17:07 paul quite unusable.
17:07 kados owen's contention (and mine since I'm dealing with more than just NPL now) is that it's too hard to maintain a template with more complex templates
17:08 kados now katipo just designed a whole new set of templates for rangitiki right chris?
17:08 chris would it be more easy if we used includes more?
17:08 chris designed is a bit strong
17:08 rach yes we did new templates, we did plain ones,
17:08 kados plain?
17:08 chris we took npl templates, ripped all the colours out
17:08 rach but probably not as plain as I think Owen was doing - so they still had some menues
17:08 kados right
17:08 owen And lots of visual formatting (tables, etc)
17:08 chris yes
17:08 rach black and white - no colours, other than links, but still with formating
17:09 chris and then we coloured them back in
17:09 rach but the least formating that we thought we could get away with
17:09 chris one for rangitikei
17:09 kados I see
17:09 chris and one for hlt
17:09 kados yea that's basically what LibLime's client tempaltes are like
17:09 kados NPL with different colors ;-)
17:09 kados but even still, as new things are developed it's a bear to update the templates
17:09 kados it's almost a full-time job
17:09 rach you saw them I think - basically with no colours, and less menus
17:10 kados for each set of templates
17:10 chris http://koha.rangitikei.katipo.[…]/koha/mainpage.pl
17:10 chris thats it coloured back in
17:10 rach yep we certainly found that
17:10 russ i hear that kados - the bulk of our effort for rangitikei and hlt upgrade was updating templates
17:11 rach that it's a big job doing the templates, I can tell you it takes about 40hrs I think to update a set of templates each new version of KOha
17:11 kados yep ... and it will continue as you maintain it ;-)
17:11 rach might actually be more than that it's true
17:11 russ rach i think 40hrs is probably ligt
17:11 kados rach: that's maybe even a bit of a low estimate ...
17:11 russ light
17:11 paul rach said something important I think. maintaining intranet template is a BIG work.
17:11 kados agreed
17:11 paul and I don't see how it can not be.
17:12 chris yes its just like redesigning a medium sized website
17:12 kados yep
17:12 rach yep
17:12 paul (for OPAC it's different as there is less than 20 templates)
17:12 chris yes
17:12 kados cool ... so we all agree mostly about at least the Intranet template
17:12 rach that it's a lot of work :-)
17:12 paul So I think we should clearly define what we need in intranet template.
17:12 kados now the PROG templates were supposed to alleviate the processes of weeding through _changes_
17:12 paul and define what kind of CSS styles we need.
17:13 paul then, everybody can have it's own css
17:13 paul and be happy ;-)
17:13 rach yep the idea of programmer templates, was one set that the programmers could make functionality changes too
17:13 kados because IMO we won't really ever agree on what our clients' want in their tempaltes
17:13 rach that were useful enough to test agains
17:13 rach against
17:13 kados for instance, rangitiki doesn't have the left-hand navbar
17:13 russ kados: and nor would we want to, i see it as a feature of koha
17:13 kados wheras NPL and LibLIme's clients depend on that
17:14 rach it doesn't have your left hand sidebar :-) we tend to favour 2 bars
17:14 kados russ: agreed
17:14 paul I have a solution for such problems with my customers : I explain how much work is needed when a new release is released, and that they would have to pay !
17:14 paul thus, every body in France has default templates for librarian interface ;-)
17:15 kados rach: well ... more than that even ... ranji doesn't have any 'shortcut' keys or quickscan options in their nav bar
17:15 kados paul: makes sense for a small community I think
17:15 rach yep we could easily put those back in
17:16 chris we probably will for hlt
17:16 chris they were quite taken by the idea of quick keys
17:16 rach yep - rangitikei are quite "low tech"
17:16 kados paul: but eventually it might be more expensive for multiple clients to have to have their templates updated than it would be to go with a proprietary system
17:17 kados paul: IMO we need to take advantage of code reuse more in the templates
17:17 kados but I must admin
17:17 kados admit even
17:17 rach yes - so I guess the question is, is there enough in common between the templates that we could actually "agree" on a default set?
17:17 kados I'm not sure what the best way to approach code reuse is
17:17 kados or even, is our whole approach to templates thusfar deficient in some way
17:18 kados for instance,
17:18 kados how do other projects handle multiple templates
17:18 kados mambo for instance, or gallery, or ...
17:18 kados etc.
17:18 owen I think to a certain extent they treat alternate templates as solely the responsibility of the author
17:19 kados one question that I've been interested in is whether we could accomplish all our different designs in CSS only
17:19 owen I think there's only so much you can do with CSS
17:19 owen You reach a point where you really do have to move stuff around on the page.
17:19 rach I think that we have a somewhat different issue - I think that we are adding features (and major ones) at a fearsome rate
17:19 rach compared to other systems
17:20 kados yep
17:20 rach so that while that still continues, we're going to have a problem
17:20 kados makes sense
17:20 rach I would think (although maybe we will continue which would be cool :-) that after the next release we might get a bit more of a hiatus
17:20 kados I think that's where the PROG template idea emerged
17:21 rach on the feature front - or at least less major features being added
17:21 kados PROG might alleviate the process of updating all the templates if it's used
17:22 owen The idea with the prog templates being, a DIFF will more easily show what has been changed if you don't have to sift out the presentational aspects
17:23 rach yep that was the idea
17:23 kados so would it make sense to say that we each will develop new functionality using our own templates
17:23 kados but when we commit code we agree to commit to the PROG templates as well as our templates?
17:23 paul mmm... sounds quite ambitious to me...
17:23 rach and (for me anyway) I wanted something that might be easier to test with
17:24 kados paul: yes it's problematic because it requires coders to commit twice :-/
17:24 paul i'm not sure we would have enough ressources to test enough each release with each set of templates
17:24 rach yep
17:25 kados ok ... so ... other suggestions?
17:25 paul and that would be quite hard to say : if you want featureX you need template katipo, while featureY is with template liblime.
17:25 kados can we modify the role of PROG templates?
17:25 rach well that's pretty much what we do now
17:25 kados exactly
17:25 rach if you want non marc with normal acquisitions, then you go for hte katipo templates
17:26 rach at the moment, it's kinda ok for "us" ie paul/joshua/katipo, because we all have templates we're familiar with
17:26 paul and my problem is that i'm not sure that marc=off and acqui=normal works fine with default templates.
17:26 paul so I don't suggest it to my libraries...
17:26 rach yep - it doesn't :-)
17:27 russ it certainly doesnt :-)
17:28 kados we need a $100,000 USD grant to hire some QA folks ;-)
17:28 rach and none of us are very confident that anything works in windows :-)
17:28 kados ha!
17:28 kados one problem that I can forsee with LibLime
17:29 kados is that the ammount of customization that each library requires
17:29 kados makes Koha not much cheaper than some of the smaller proprietary solutions
17:29 kados (in fact, it's more expensive than the really small ones)
17:29 kados I'm not sure whether that's just the nature of the beast
17:29 chris yep, but it is that ability to customise that makes koha better
17:29 rach yep it pretty much is I think
17:30 chris i think competing on price only gets so far
17:30 kados chris: in some ways I agree ... but in many cases I think it could work better 'out of the box'
17:30 kados not that it's anyone's fault
17:30 kados we're all strapped for time and resources
17:30 rach yep
17:30 chris yep, the installer could set up more stuff
17:31 russ yeah we were talking about that werent we chris?
17:31 chris i think time is the biggest resource we are short of
17:31 chris well me anyway :)
17:32 kados hehe
17:32 kados you are the biggest resources we are short of chris ;-)
17:32 kados resource even ...
17:33 kados so any more ideas about the PROG templates?
17:33 kados paul can you imagine a way that they could be useful?
17:34 paul could we developp a "template compiler" (a XSL stylesheet ?) that transforms PROG into something else ?
17:34 chris that would be cool
17:34 chris we could make the PROG template xml
17:34 kados hehe ...
17:34 paul yep, i'm afraid !
17:35 kados that may be a Koha 4.0 thing
17:35 owen We might as well just chuck the prog templates if the only solution is that complicated (given our lack of time)
17:35 kados go 100% XML
17:35 kados yep
17:36 paul chuck ?
17:36 kados paul: get rid of ;-)
17:36 russ chuck = throw
17:36 owen (sorry paul)
17:36 paul maybe, but the main thing ppl said with koha is : it' not nice/fun/pretty ...
17:37 paul so we need to have a 3.0 really nice.
17:37 kados paul: you say in your email:
17:37 chris i think our aim for 3.0
17:37 paul and that needs rewritting templates, not making default one evolving.
17:37 kados in my idea, the PROG templates should give a framework for template designers
17:37 russ man after going to the lianza conference
17:37 chris is we should consider pretty templates as important as other features
17:37 kados chris: agreed
17:37 russ i dont think there are any ils that can say they are nice/fun/pretty
17:37 chris ie, we should make sure koha looks nice as well as works before a release
17:38 kados <quoting paul>in my idea, the PROG templates should give a framework for template designers</quoting>
17:38 chris as russ says
17:38 kados do we all agree with this?
17:38 chris other library systems are butt ugly, lets not join them :-)
17:38 kados is that the purpose of PROG templates?
17:38 rach indeed
17:39 chris hmm
17:39 kados from the point of view of a template maintainer ... is it useful to have such a framework?
17:39 rach yes - the ideal would be, IMO of course :-) that with the addition/use/changing of some includes and some css files, people could get functional non hideous templates from the programer templates
17:39 rach without having to actually make new .tmpl files
17:39 chris IMHO just one set of templates that are known to work would do
17:40 chris it would make making/maintaining other sets easier
17:40 kados owen: your thoughts?
17:40 chris if there was one that worked
17:40 kados chris: who does the template coding for katipo?
17:40 chris not me
17:40 rach coding or design?
17:40 chris i wave my arms around and say make it pretty
17:40 kados coding
17:40 rach coding generally is bob
17:40 chris and it just happens
17:40 chris :)
17:40 kados is he around?
17:41 owen I could see modifying the programmer templates so that the addition of one or more include files would round it out to a fully functional template.
17:41 kados might be interesting to get his perspective
17:41 owen But a lot more subjective decisions would have to be made regarding formatting and navigation
17:41 rach he would be, but he's on a tight deadline for another job
17:42 owen You can't include from an include file in HTML:Template, can you?
17:42 russ nah is not that tight, he has done some sterling work this morning
17:43 kados morning bob
17:43 bob hi everyone
17:43 chris yep owen
17:43 kados we're brainstorming for how to streamline the process of maintaing Koha tempaltes
17:44 kados and we're hoping you have some ideas ;-)
17:44 chris <!--TMPL_INCLUDE something something
17:45 chris you can even write a routine to html::template can use that parse <!--#include virtual .. and converts them to <!--TMPL_INCLUDE
17:45 kados bob: owen has worked up a set of minimal 'programmer' templates .. dunno how up to speed you are on them
17:46 kados bob: but the idea was that programmers would commit changes to these minimal templates and it would be easier to update a set of 'real' templates using them as a guide
17:46 bob umm, i don't know if i've seen them yet
17:46 kados bob: but there are logistical problems with maintaining two sets of templates ...
17:47 kados bob: any thoughts?
17:47 bob it would be helpful to have at least a list of what variable a template could have
17:48 bob so not exactly a minimal template - could be a text file
17:49 kados interesting
17:49 kados I wonder if there's an automated way of generating that
17:49 bob that way i could know what i could display without altering perl code
17:49 kados owen: would that be useful from your perspective too?
17:51 owen I experimented with doing that with some of the prog templates
17:51 owen moremember.tmpl at least, as I recall
17:51 owen But it wasn't automated at all
17:56 russ would it help if we asked bob to do a bit of a write up to the devel list?
17:56 paul yep.
17:56 paul it's midnight for me, maybe time to go to bed. and maybe i'll dream the best solution ;-)
17:56 russ cos as kados pointed out, he is a little on the spot
17:57 chris sleep well paul
17:57 owen Thanks for staying up, Paul
17:57 rach nice to see you paul :-)
17:58 russ this template thing - is it stopping you from doing stuff paul?
17:58 paul a little bit russ.
17:58 paul I mean i've some features to add, and i was wondering if it was better to wait a little or to write them on default or on PROG
17:59 paul writing them on prog is probably too hard.
17:59 paul maybe PROG would be useable if we just added a standard menu.
17:59 kados paul: do you plan to maintain your 'default' templates in 3.0?
17:59 paul nope.
17:59 kados owen: will you have time to add a standard menu?
17:59 kados paul: would you?
17:59 paul I plan to use a very nice one that someone else will develop ;-)
18:00 paul (I really love katipo "fall" theme)
18:00 paul (& think liblime one is nice too)
18:01 kados owen's gonna jet on us ... he's off as of 1 minute ago
18:01 paul that's another interesting question : should we wait until a decision has been taken & made reality, or continue with default for instance & port it after.
18:01 paul ?
18:01 kados good question
18:01 owen I'm willing to make whatever modifications necessary to make the prog templates workable.  But only insofar as they keep their functionality as programmer templates
18:02 owen ...And yes, gotta go! I'll catch up on the logs later
18:02 paul I agree owen. With a TMPL_INCLUDE, it could be quite easy to do something simple.
18:03 kados lets think it over ... if anyone has any further ideas mail them to koha-devel
18:03 paul ok, so I go to bed. Pls kados, as RM, don't forget to take a decision on this topic.
18:03 paul because it's slowing me & hdl atm.
18:03 kados I'll make a decision before end of week paul ;-)
18:03 paul ok, great.
20:13 JustJoe does anyone have experience setting up koha, i seem to be having a problem getting apache to show the website properly
20:14 JustJoe when i am on the server, i can hit the website fine in lynx, however when i am connecting to it from another machine in a normal web browser i get apache's placeholder page
20:15 JustJoe it just seems funny that localhost forwards me to one place and being on the network forwards me somewhere else
20:19 chris what url do you hit it on when u check it locally
20:19 chris vs over the network
20:20 chris cos if you have <Virtualhost localhost>
20:20 chris ServerName localhost
20:20 chris etc
20:20 chris and then hit it on
20:21 chris 192.172.12.1 (i just made that ip number up)
20:21 chris its not going to match that virtualhost
20:21 chris so you get the placeholder page
20:22 chris what you probably need is
20:22 chris a
20:22 chris ServerAlias whatever.you.access.it.on.the.network
20:22 chris in your apache config
20:25 JustJoe i use localhost when i hit it locally
20:25 chris if you look at you koha-httpd.conf file
20:26 chris it probably says <Virtualhost localhost> ?
20:28 JustJoe <Virtualhost music-lib-svr:80>
20:29 chris and in /etc/hosts
20:29 JustJoe 127.0.0.1       localhost.localdomain   localhost       music-lib-svr
20:30 chris there we go
20:30 chris and when you access it from the network, what url do you use?
20:30 JustJoe i just hit its network IP address
20:30 chris right
20:31 chris in theory
20:31 chris if you take music-lib-svr off that line in /etch/hosts
20:31 chris then put one
20:31 chris its.network.address      music-lib-svr
20:31 chris and restart apache
20:32 chris then lynx http://music-lib-svr/ should work
20:32 chris and from outside http://networkaddress/ should work
20:35 JustJoe thanks, that worked.  Now i just got to figure out the hosts file correctly.  This seems to be where the problem is.
20:35 chris yep
00:03 indradg chris. who uses lynx these days ? ;)
00:03 chris not me, i use links :)
00:03 chris but its generally on most machines :)
00:04 indradg heh
00:06 indradg chris, head over to http://foss.in/2005/ when u r free
00:06 chris links is cool because the mouse works in it
00:06 chris even over ssh
00:06 chris which is really handy for testing stuff on ppls intranets
00:06 indradg i'll be talk about Koha (mostly)
00:06 indradg chris, i agree
00:08 chris wow, in good company :)
00:09 indradg yup!
00:09 chris excellent
00:09 indradg http://foss.in/2005/schedules/
00:10 indradg some really nice sessions in there
00:10 chris sure are
00:10 chris looks like a great conference
00:12 indradg last week at a conference a guy from MSFT Research chatted me up.... we ended up discussing device driver models in FreeBSD vs. Linux... the guy's fav platform is Mac OS X... which he uses at work :)
00:13 chris i like OS X a lot
00:13 chris its a pity that macs cost so much
00:13 chris i wouldnt mind an ibook laptop
00:14 chris but im pretty fond of debian too :)
00:14 indradg yeah... even 2 years back a 128 MB RAM clip used to cost 16000 rupees (~ 3500 USD).... argh!
00:14 chris yikes
00:15 indradg sorry that would be 350 USD... extra zero ;)
00:15 chris ahh thats better... still expensive though
00:15 chris so foss.in is an annual event?
00:16 indradg now things have changed... can get a decent mac laptop with DVD writer and stuff in india for about 1,700 USD in indian currency
00:16 indradg chris, yes... FOSS.IN is an annual event... till last year it used to be called as Linux Bangalore... changed the name this year for an all india flavour
00:17 chris right
00:17 chris and usually in december?
00:17 indradg yes... the best time to visit india :)
00:20 chris darn, india will be in south africa in december
00:20 indradg lol
00:20 chris i was hatching a cunning plan to combine a visit to india with foss.in and a cricket match :-)
00:21 indradg :)
03:48 osmoze hello
09:28 Sylvain hi !
10:13 Sylvain paul ?

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