2008-03-26T00:00:42 <nwp> very motivating. interesting AI/control/game theory opportunities in writing good bots. something cool that they can put their name to when done. potentially complex threading. working through copyright issues. lots of different potentially interesting areas
2008-03-26T00:00:42 <TheSheep> welcome to the world of tomorrow!
2008-03-26T00:00:50 <xorAxAx> hmm
2008-03-26T00:01:00 <xorAxAx> sounds like a lot of piles of mud to work on :)
2008-03-26T00:01:15 <xorAxAx> nwp: i think they had some student projects here that were about writing Q3 bots
2008-03-26T00:01:19 <xorAxAx> and UT
2008-03-26T00:01:21 <TheSheep> nwp: I doubt we have anyone that good right now
2008-03-26T00:01:33 <grzywacz> xorAxAx, no, how many apps for moin do you expect to see this year?
2008-03-26T00:01:37 <grzywacz> s/no/so/
2008-03-26T00:01:44 <xorAxAx> grzywacz: no idea :)
2008-03-26T00:01:49 <grzywacz> guesstimate!
2008-03-26T00:01:52 <xorAxAx> i hope to see a lot
2008-03-26T00:01:54 <dreimark> gizmach: back to your project. just ask!
2008-03-26T00:01:57 <nwp> oh, and currently it needs you to xhost +server on each client. So rewriting to client-server would be a big project
2008-03-26T00:01:58 <xorAxAx> hmm, 20?
2008-03-26T00:02:04 <grzywacz> Hm. ok :)
2008-03-26T00:02:12 <nwp> or at least coming up with a secure way of playing
2008-03-26T00:02:25 <gizmach> dreimark: trying to understand some german :D I'll learn it if I got the project :D
2008-03-26T00:03:07 <nwp> lots of the guys involved originally were at MIT so some good people did some good stuff on it.
2008-03-26T00:03:23 <nwp> main author is now a catholic priest
2008-03-26T00:03:26 <xorAxAx> nwp: hehe
2008-03-26T00:03:31 <xorAxAx> how ironic
2008-03-26T00:04:01 <xorAxAx> but maybe war is something else than tanks shooting at each other
2008-03-26T00:04:35 <TheSheep> xorAxAx: it's definitely different than games
2008-03-26T00:04:48 <xorAxAx> TheSheep: do you know wargames? :-)
2008-03-26T00:05:25 <TheSheep> xorAxAx: do you want to play a game?
2008-03-26T00:05:46 <xorAxAx> hehe
2008-03-26T00:06:10 <TheSheep> s/do/Would/
2008-03-26T00:06:16 <TheSheep> s/want/like/
2008-03-26T00:06:25 <TheSheep> sigh, getting old
2008-03-26T00:08:19 <dreimark> TheSheep: hmm ?
2008-03-26T00:08:47 <TheSheep> dreimark: I'm forgetting exact quotes from obscure movies.
2008-03-26T00:09:43 <xorAxAx> its not obscure
2008-03-26T00:09:49 <xorAxAx> dreimark: did you watch it in the cinema?
2008-03-26T00:10:02 <dennda> xorAxAx: no apps yet?
2008-03-26T00:10:20 <dreimark> xorAxAx: no, I think I have seen it on tv
2008-03-26T00:10:27 <xorAxAx> dennda: indeed
2008-03-26T00:10:33 <xorAxAx> dreimark: ah
2008-03-26T00:10:55 <dennda> ok
2008-03-26T00:11:15 <xorAxAx> so, currently you have a 100% chance of participating :-)
2008-03-26T00:13:53 <dennda> 1 application => 1 slot? Is there any formula for calculating how many slots you get?
2008-03-26T00:14:29 <xorAxAx> no, 1 application => p slots whereas 0 <p <= 1
2008-03-26T00:14:45 <xorAxAx> no, that formula is secretly held in google's computers
2008-03-26T00:15:00 <xorAxAx> and one important factor in it is sympathy for people involved
2008-03-26T00:15:16 <xorAxAx> and their track record. and how they like google.
2008-03-26T00:15:29 <xorAxAx> actually, google removed one open source proejct from SOC this year
2008-03-26T00:15:41 <xorAxAx> even though they participated in the GHOP pilot (together with us and 8 other orgs)
2008-03-26T00:15:43 <grzywacz> Like, after it already got accepted?
2008-03-26T00:16:07 <xorAxAx> no, but usually they would have been accepted because of their participation in the last years
2008-03-26T00:16:31 <xorAxAx> their mistake - they (or better, he) overrated the relationship between the project and google
2008-03-26T00:17:23 <dennda> so if you don't like google you are out of the race?
2008-03-26T00:18:23 <xorAxAx> dennda: did you read the TOS?
2008-03-26T00:18:31 <dennda> partially
2008-03-26T00:18:37 <dennda> honestly, did you?
2008-03-26T00:18:38 <xorAxAx> as you probably already accepted it, you must be in love with google
2008-03-26T00:18:55 <xorAxAx> no, but people were pasting parts of it
2008-03-26T00:19:06 <dennda> is there a section "I admit to love google"?
2008-03-26T00:19:23 <xorAxAx> no, but you agree to do things that you wouldnt do if you werent in love :-)
2008-03-26T00:19:39 <dennda> like what?
2008-03-26T00:19:46 * dennda should reread the TOS
2008-03-26T00:20:36 <xorAxAx> being in love is a nice thing, questions are unimportant. everything will have been better. any other questions?
2008-03-26T00:20:41 <xorAxAx> ;-)
2008-03-26T00:21:00 <dennda> are you actually a hippie?
2008-03-26T00:21:08 <xorAxAx> no
2008-03-26T00:21:11 <grzywacz> haha
2008-03-26T00:21:16 <gizmach> :)
2008-03-26T00:21:42 <xorAxAx> well, i am feeling a bit uncomfortable to reply in a not confusing way because this channel is logged
2008-03-26T00:22:05 * gizmach needs to wake up for 5 hours so nitte all
2008-03-26T00:22:12 <xorAxAx> have a good night, gizmach
2008-03-26T00:22:17 <xorAxAx> erdös only need 4 hours
2008-03-26T00:22:50 <xorAxAx> s/need/&ed/
2008-03-26T00:22:51 <gizmach> dreimark: thx for tutorial :) and to all for help
2008-03-26T00:22:52 <grzywacz> night
2008-03-26T00:23:05 <dreimark> gn grzywacz
2008-03-26T00:23:18 <dreimark> gn gizmach
2008-03-26T00:24:33 <dennda> good night melita
2008-03-26T00:26:07 <dreimark> see you in some hours - good night :)
2008-03-26T00:30:03 <dennda> how precise must the timeline be that I have to deliver?
2008-03-26T00:31:10 <xorAxAx> not very, also we will refine in the next months
2008-03-26T00:31:31 <xorAxAx> its important to communicate the next goals in the timespan while the student is active
2008-03-26T00:31:39 <xorAxAx> and actively discuss whether they can be held
2008-03-26T00:35:09 <dennda> sounds good
2008-03-26T00:36:37 <ThomasWaldmann> cb22: looks like za has a sane minister of public service and administration
2008-03-26T00:36:56 <ThomasWaldmann> http://slashdot.org/articles/08/03/25/1914233.shtml
2008-03-26T00:39:32 <xorAxAx> Microsoft responded aggressively, saying that 'there is no such thing as free software. Nobody develops software for charity.'
2008-03-26T00:42:10 <ThomasWaldmann> "there is no such thing as microsoft. we are all just having a bad daydream."
2008-03-26T00:43:44 <ThomasWaldmann> http://science.slashdot.org/science/08/03/21/1529217.shtml?tid=160 < wow
2008-03-26T00:44:39 <ThomasWaldmann> must be really intense if you see gamma rays with your eyes
2008-03-26T00:45:38 <xorAxAx> i seriously doubt that you can see those :)
2008-03-26T00:45:51 <dennda> i thought MS had a new open source strategy
2008-03-26T00:46:12 <xorAxAx> there is only one documented case where direct exposure to gamma rays led to an immediate visible artefact
2008-03-26T00:46:16 <xorAxAx> but the reason wasnt clear
2008-03-26T00:46:25 <xorAxAx> and nobody tried to do any experiments on this
2008-03-26T00:46:28 <xorAxAx> i wonder why :-)
2008-03-26T00:47:17 <xorAxAx> dennda: well, that doesnt mean that they already got the whole picture :)
2008-03-26T00:47:41 <xorAxAx> dennda: many people in open source do not develop "for charity" :)
2008-03-26T00:48:13 <xorAxAx> about 50% of the involved programmers earn money with it. so some small fraction of that doesnt do it "for charity"
2008-03-26T00:48:21 <dennda> xorAxAx: yes, I am aware of the fact :)
2008-03-26T01:17:33 <dennda> the application template says: Project(s) you apply for and your ideas for them
2008-03-26T01:17:55 <dennda> I'll leave that point empty since I don't know if there'll be any other project I'll be applying for
2008-03-26T01:18:07 <dennda> except you tell me my application is bad :)
2008-03-26T01:18:35 <dennda> (I want to focus on moin. I don't have enough time to read several projects source-code)
2008-03-26T01:18:46 <dennda> I hope that is ok
2008-03-26T01:19:43 <dennda> Or isn't that meant as "For which other organizations do you apply"?
2008-03-26T01:40:34 <ThomasWaldmann> projects does not mean "organisations", but your project idea(s)
2008-03-26T01:41:03 <ThomasWaldmann> dennda:
2008-03-26T01:42:02 <dennda> [A[Adfasd
2008-03-26T01:42:34 <dennda> I do really wonder how that happened
2008-03-26T01:42:36 <dennda> sorry
2008-03-26T01:44:14 <dennda> I think I'll finnish the application tomorrow. There're still some things I need to find out / ask :)
2008-03-26T01:49:56 <zenhase> good evening/night
2008-03-26T01:50:06 <dennda> gn zenhase
2008-03-26T01:50:07 <zenhase> or just moin :)
2008-03-26T01:50:18 <zenhase> na, i am coming in right now ;)
2008-03-26T01:50:30 <dennda> heh that's kind of a running gag
2008-03-26T01:50:45 <zenhase> what?
2008-03-26T01:50:50 <zenhase> moin? :)
2008-03-26T01:50:58 <dennda> saying "moin" in this channel
2008-03-26T01:51:26 <zenhase> almost too obvious ;)
2008-03-26T01:51:39 <dennda> hehe
2008-03-26T01:51:45 <dennda> I wonder who came up with the idea
2008-03-26T01:51:58 <kikka> Hi zenhase.
2008-03-26T01:52:04 <zenhase> hi kikkay
2008-03-26T01:52:05 <dennda> it's quite an unusual name :)
2008-03-26T01:52:17 <zenhase> dennda: it's a german product
2008-03-26T01:52:21 <kikka> zenhase: :o kikka here.
2008-03-26T01:52:25 <zenhase> it's really that simple
2008-03-26T01:54:23 <kikka> Tommorow is my last day of holidays :( so I go to bed now.
2008-03-26T01:54:29 <kikka> good night all.
2008-03-26T01:54:43 <dennda> gn kikka
2008-03-26T01:55:49 <zenhase> gn kikka
2008-03-26T01:56:01 <zenhase> kikka: and what happened to your 'y' here?
2008-03-26T02:00:02 * dennda goes to bed as well
2008-03-26T02:00:04 <dennda> good night
2008-03-26T02:00:12 <zenhase> good night dennda
2008-03-26T02:28:51 <ThomasWaldmann> http://moinmo.in/FirstnameLastname hah!
2008-03-26T02:29:07 <HongJun> Hello, I have an idea to add REST support for Moin, what do you think about this idea
2008-03-26T02:30:33 <ThomasWaldmann> HongJun: that "use FirstnameLastname" was rather intended to be JunHong :D
2008-03-26T02:31:03 <ThomasWaldmann> HongJun: you know moin has a rst parser?
2008-03-26T02:31:26 <HongJun> sorry, i don't know
2008-03-26T02:31:39 <zenhase> ThomasWaldmann: not rst
2008-03-26T02:31:47 <zenhase> he is talking about something else
2008-03-26T02:31:47 <ThomasWaldmann> (it could have some more maintenance by someone using it, though)
2008-03-26T02:32:13 <zenhase> i think he means the webservice thingy REST
2008-03-26T02:32:15 <ThomasWaldmann> HongJun: ok, more details please :)
2008-03-26T02:33:20 <ThomasWaldmann> (some abbreviations are overused)
2008-03-26T02:33:26 <HongJun> REST：REpresentational State Transfer
2008-03-26T02:35:02 <HongJun> It is an architectural style, that allow user to use a standard API for content managerment
2008-03-26T02:35:38 <zenhase> the success of web 2.0 rides on it's back :)
2008-03-26T02:36:21 <zenhase> i guess you got the idea also from reading the proposals for the CRUD/action-redesign of moin
2008-03-26T02:37:10 <HongJun> no, which page?
2008-03-26T02:37:50 <zenhase> oh, it was under User Interface enhancements in http://moinmo.in/GoogleSoc2008/InitialProjectIdeas
2008-03-26T02:38:13 <zenhase> but it's a great idea, that is for sure
2008-03-26T02:38:33 <zenhase> though i think it will encompass quite a lot of parts of moin
2008-03-26T02:38:53 <HongJun> Ok, thank you
2008-03-26T02:39:20 <HongJun> I want to applly for this project of SoC
2008-03-26T02:39:23 <zenhase> i suggest you start writing an application with a nice outline of your ideas
2008-03-26T02:39:40 <zenhase> yeah, sounds like a cool idea
2008-03-26T02:39:44 <ThomasWaldmann> i guess i will need some more practical application example than the REST article blah i just read. :)
2008-03-26T02:40:57 <HongJun> Roby on Rail supports REST service
2008-03-26T02:40:58 <ThomasWaldmann> (note that moin is not xml based, so posting xml to it does not do much good)
2008-03-26T02:41:12 <zenhase> hmm
2008-03-26T02:41:16 <zenhase> it's not about xml
2008-03-26T02:41:37 <HongJun> It just an standard version of HTTP
2008-03-26T02:41:45 <zenhase> it's about putting the HTTP verbs back in place again ... and about being quite stateless
2008-03-26T02:42:03 <ThomasWaldmann> and also e.g. using DELETE for deleting items -> no existing client does that
2008-03-26T02:42:12 <zenhase> i think it could be a real good start to get the actions in moin sorted out in a standard way
2008-03-26T02:42:25 <ThomasWaldmann> so we would just create another machine API in addition to xmlrpc
2008-03-26T02:42:36 <HongJun> with four method GET, POST, PUT, DELETE fro CRUD
2008-03-26T02:43:06 <ThomasWaldmann> a user with a browser won't be able to do a DELETE request
2008-03-26T02:43:11 <zenhase> ThomasWaldmann: rich interfaces with JS can make use of that too
2008-03-26T02:43:18 <zenhase> ThomasWaldmann: oh, it perfectly is
2008-03-26T02:43:34 <zenhase> and it gets used already
2008-03-26T02:43:42 <zenhase> especially in ajax and webservice land :)
2008-03-26T02:44:08 <ThomasWaldmann> we don't have ajax stuff yet. and moin should also be usable without js.
2008-03-26T02:44:17 <zenhase> of course
2008-03-26T02:44:26 <HongJun> the is a book "RESTful.Web.Services" about the REST web service
2008-03-26T02:44:31 <zenhase> but you can build added value on top of that
2008-03-26T02:44:41 <zenhase> and it's damn easy
2008-03-26T02:44:51 <zenhase> nobody likes xmlrpc or even soap
2008-03-26T02:45:03 <zenhase> because it's too complex to manage
2008-03-26T02:45:23 <HongJun> REST is a lightweight web service, it may replace soap
2008-03-26T02:45:30 <zenhase> well, i think several java guys like soap ... but then, they like blown-out-of-proportions-xml
2008-03-26T02:47:08 <zenhase> hmm
2008-03-26T02:47:21 <ThomasWaldmann> that "service" needs data to transport
2008-03-26T02:47:22 <zenhase> ThomasWaldmann: which article about REST did you just read?
2008-03-26T02:47:28 <HongJun> restlet is web server written in java
2008-03-26T02:47:42 <zenhase> ThomasWaldmann: the Wikipedia one?
2008-03-26T02:47:50 <ThomasWaldmann> i don't see much of moin that could be used easily in such a way
2008-03-26T02:48:27 <HongJun> here is the origin of REST http://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm
2008-03-26T02:48:30 <ThomasWaldmann> (or better used than with xmlrpc)
2008-03-26T02:49:10 <ThomasWaldmann> so, if you write an application for that, show us the usefulness :)
2008-03-26T02:51:45 <HongJun> Unlike SOAP and XML-RPC, REST does not
2008-03-26T02:51:46 <HongJun> really require a new message format
2008-03-26T02:53:50 <zenhase> well, because it's not about messages like method-calls or so as in rpc ... it's about content transfered
2008-03-26T02:58:30 <ThomasWaldmann> i read misc stuff, including wp and some stuff google returned
2008-03-26T02:59:52 <ThomasWaldmann> it seems mostly describing what the web already is and does. with some more emphasis on rarely used methods (like e.g. DELETE or PUT).
2008-03-26T03:00:28 <zenhase> well, not quite that way
2008-03-26T03:00:29 <ThomasWaldmann> the problem is that nobody has a client that wants to use moin that way.
2008-03-26T03:00:44 <zenhase> oh, everybody does
2008-03-26T03:00:58 <ThomasWaldmann> but people have browsers and some xmlrpc scripts.
2008-03-26T03:00:59 <zenhase> several clients actually
2008-03-26T03:01:11 <zenhase> browsers, rss-readers
2008-03-26T03:01:39 <ThomasWaldmann> so which browser does a DELETE request?
2008-03-26T03:01:52 <HongJun> if moin have a RESR api, the client and application could operate the content on moin easier
2008-03-26T03:02:12 <ThomasWaldmann> which client?
2008-03-26T03:02:29 <zenhase> maintenance scripts and the like
2008-03-26T03:02:44 <ThomasWaldmann> there is none for rest
2008-03-26T03:04:35 <zenhase> actually i think i will withdraw from the discussion here
2008-03-26T03:04:54 <ThomasWaldmann> so implementing rest service would mean starting from zero, for the server part as well as for the client part
2008-03-26T03:05:26 <ThomasWaldmann> of course you can write all that stuff, sooner or later...
2008-03-26T03:05:47 <zenhase> it's quite hard argueing in pro or contra for sth. that more or less is an architectural style of building applications
2008-03-26T03:06:27 <zenhase> the immeadiate effects will be ... hmmm ... more of idealistic nature
2008-03-26T03:07:43 <zenhase> right now there are (as stated by several REST-opponents on the web) very few websites out there, that put REST to use
2008-03-26T03:07:56 <ThomasWaldmann> well, if at least the long terms effects would likely be different from that... :)
2008-03-26T03:09:05 <zenhase> well, there are effects observable around the web, especially in those web2.0 areas
2008-03-26T03:09:30 <zenhase> problem is, it's hard to attribute them to RESTful services
2008-03-26T03:10:35 <zenhase> all those nice mashups, cross-site content exchange from flickr, to del.icio.us, blogs, etc. is largely based on some kind of webservice interaction
2008-03-26T03:10:40 <HongJun> Amazon has provide rest web service
2008-03-26T03:10:49 <ThomasWaldmann> well, anyway. HongJun: be very concrete about what you want to do and for what you want to do that.
2008-03-26T03:11:30 <zenhase> yeah :)
2008-03-26T03:11:36 <ThomasWaldmann> try not to use a lot of buzzwords, but give concrete applications and reasons why it is better than what we have.
2008-03-26T03:11:54 <ThomasWaldmann> also look at the moin code about doability of what you want.
2008-03-26T03:11:58 <HongJun> We don't need to anxious about the client, we could just provide a REST APi, just like we provide RSS/Atom feed
2008-03-26T03:13:47 <zenhase> i need to get some sleep now
2008-03-26T03:13:55 <ThomasWaldmann> gn zenhase :)
2008-03-26T03:13:59 <HongJun> Ok, I will give a concrete later
2008-03-26T03:14:06 <HongJun> bye
2008-03-26T03:14:10 <zenhase> HongJun: fine :)
2008-03-26T03:14:41 <ThomasWaldmann> bye HongJun
2008-03-26T03:15:17 <HongJun> I am still here, i just say bye to zenhase
2008-03-26T03:15:26 <ThomasWaldmann> ah, ok :)
2008-03-26T03:16:19 <zenhase> HongJun: and don't feel initimidated by thomas' criticism ... it actually is very useful. you can measure your ideas on their ability to satisfy other peoples need for concrete examples
2008-03-26T03:17:10 <zenhase> i should have listened better last year myself, perhaps my goals would have been clearer then :)
2008-03-26T03:17:30 <zenhase> (i was a student in the 2007 soc)
2008-03-26T03:17:37 <zenhase> well ... now off to bed
2008-03-26T03:17:45 <HongJun> ok, thank you
2008-03-26T03:17:49 <zenhase> cu guys then :)
2008-03-26T03:18:25 <HongJun> ThomasWaldmann: Is there any document about the moin architecture
2008-03-26T03:20:20 <ThomasWaldmann> there is some stuff on the wiki but not much and not very detailed or current
2008-03-26T03:23:39 <ThomasWaldmann> http://moinmo.in/ReimarBauer/UnderConstruction?action=AttachFile&do=get&target=MoinMoinArchitecture.png
2008-03-26T03:23:54 <ThomasWaldmann> this is a view from 1000km above
2008-03-26T03:29:12 <HongJun_> i can't open it
2008-03-26T03:29:29 <HongJun_> could you send me a email hongjun.bj@gmail.com
2008-03-26T03:30:28 <ThomasWaldmann> it's on the wiki
2008-03-26T03:30:49 <ThomasWaldmann> try do=view
2008-03-26T03:31:59 <ThomasWaldmann> and please rename your name in your account file and your homepage
2008-03-26T03:32:07 <damjan> ThomasWaldmann: do you have some knowledge how many CGI moin users are there in the wild??
2008-03-26T03:32:18 <ThomasWaldmann> no
2008-03-26T03:34:05 <HongJun_> ok
2008-03-26T03:41:09 <HongJun_> how could I change the title of my homepage
2008-03-26T03:43:14 <ThomasWaldmann> rename it (see menu)
2008-03-26T03:43:51 <ThomasWaldmann> use the same target name as you have as username in your profile
2008-03-26T03:49:22 * ThomasWaldmann .sleep()
2008-03-26T03:49:24 <ThomasWaldmann> gn
2008-03-26T03:51:46 <HongJun_> good night
2008-03-26T08:09:23 <HongJun_> Does moin use the MVC model?
2008-03-26T08:10:22 <starshine> HongJun_: expand the acronym please?
2008-03-26T08:11:37 <HongJun_> Model-View-Control
2008-03-26T08:20:39 <starshine> I'm not familiar with this buzzword, it might, or it might not.
2008-03-26T08:21:19 <starshine> (or rather, it doesn't deliberately use the model, I think, but it may nontheless fit it; I can't tell from the name.)
2008-03-26T08:27:32 <dreimark> moin
2008-03-26T08:28:04 <starshine> moin dreimark
2008-03-26T08:28:14 <starshine> you familiar with this buzzword he's askign after?
2008-03-26T08:28:17 <dreimark> HongJun_: it is more Model 2.
2008-03-26T08:28:35 * starshine takes that as yes..
2008-03-26T08:29:35 <dreimark> starshine: Model 2 is a variant of a server side implemented MVC
2008-03-26T08:29:50 <HongJun_> dreimark: what's do you mean by 'more Model 2"'
2008-03-26T08:30:05 <starshine> when I'm awake sometime someone will have to give me a url for this thingy :)
2008-03-26T08:31:02 <starshine> <- z
2008-03-26T08:31:10 <dreimark> good night starshine
2008-03-26T08:33:15 <dreimark> HongJun_: any action on the webfrontend uses a new request on the server side
2008-03-26T08:34:27 <HongJun_> which modul of moin
2008-03-26T08:35:06 <dreimark> starshine_away: http://e-docs.bea.com/alui/devdoc/docs60/Overview_of_the_Portal_Architecture/Portal_UI/PlumtreeDevDoc_Overview_MVCArchitecture.htm
2008-03-26T08:36:04 <dreimark> HongJun_: http://moinmo.in/ReimarBauer/UnderConstruction
2008-03-26T08:36:21 <dreimark> I have to apologize it is in german
2008-03-26T08:36:37 <dreimark> because it was done for a book
2008-03-26T08:37:16 <dreimark> a rough simple overview gives the picture at
2008-03-26T08:37:19 <dreimark> http://moinmo.in/ReimarBauer/UnderConstruction#head-773b6aa7257d87c24711376e98da2b02e86139ec
2008-03-26T08:43:01 <dreimark> starshine_away: http://en.wikipedia.org/wiki/Model-view-controller
2008-03-26T08:47:06 <HongJun_> Action is the controller?
2008-03-26T08:47:38 <HongJun_> and Page is the View?
2008-03-26T08:48:01 <HongJun_> but i can't get which one is model
2008-03-26T08:56:54 <dreimark> storage of data is currently done by PageEditor, (it is not shown in the picture)
2008-03-26T08:59:04 <dreimark> and the parser is used with the formatter for output
2008-03-26T10:43:24 *** irc.freenode.net sets mode: +o ChanServ
2008-03-26T12:15:11 <dennda> good morning
2008-03-26T12:15:59 <kikka> Hi
2008-03-26T12:23:04 <ThomasWaldmann> moin
2008-03-26T12:30:34 <dreimark> seems we have a problem in the converter#
2008-03-26T12:31:03 <dreimark> however it happens one of my 1.6 wikis gots converted again
2008-03-26T12:36:34 <dreimark> some of the links got broken by that
2008-03-26T12:37:03 <dreimark> e.g. [[attachment: -> [[[attachment:
2008-03-26T12:37:12 <dreimark> s/converter/migration converter/
2008-03-26T12:37:18 * dreimark lunch
2008-03-26T12:41:22 <zenhase> moin
2008-03-26T12:44:03 <ThomasWaldmann> dreimark: what does it mean "got converted again"? without some manual changes in data/meta file, the mig scripts won't do a double conversion as they track the version number.
2008-03-26T12:44:52 <ThomasWaldmann> but it could be that someone used (wrong) [[attachment: in 1.5, and the 1.6 parser isn't too happy with [[[stuff
2008-03-26T12:50:44 <dennda> I see that AttachFile has kind of "special support" for special kinds of files, like zip or vector graphics. Wouldn't it be great if that was easily expandable with an extended storage backend?
2008-03-26T12:51:13 <dennda> One could say, "Hey, there's no support for flash videos yet. Let me add that. ..... Boy, that was easy!"
2008-03-26T12:51:31 <dennda> (ideal case may differ from reality)
2008-03-26T12:51:31 <xorAxAx> dennda: its completly unrelated to the storage backend :)
2008-03-26T12:51:45 <xorAxAx> dennda: but the task you just described is also described on our ideas page
2008-03-26T12:51:58 <dennda> did I confuse things?
2008-03-26T12:52:06 <dennda> just got up, may be possible
2008-03-26T12:52:10 <xorAxAx> yes, layering violation :-)
2008-03-26T12:52:18 <xorAxAx> storage != presentation to the user
2008-03-26T12:52:44 <dennda> yeah, of course
2008-03-26T12:52:58 <dennda> the user doesn't care how it is stored
2008-03-26T12:53:35 <dennda> just trying to find out how things were before hwendell started his task, how they are with his code and how things should be
2008-03-26T12:54:10 <xorAxAx> yes, but i would describe the new UI stuff to be separate from the storage stuff
2008-03-26T12:54:10 <dennda> (the latter being then part of my application)
2008-03-26T12:54:19 <xorAxAx> the latter is more internal, the former is more external
2008-03-26T12:54:31 <xorAxAx> thats why combining this might not make complete sense
2008-03-26T12:54:32 <dennda> yes, I understand the difference
2008-03-26T12:54:45 <xorAxAx> time-wise (because both areas still need some work)
2008-03-26T12:55:23 <dennda> yep, got it
2008-03-26T12:56:17 <dennda> I'm just trying to structure what I know and what I still need to learn on how it works
2008-03-26T12:56:23 <dennda> so
2008-03-26T12:56:27 <dennda> the user wants to attach a file
2008-03-26T12:56:32 <xorAxAx> yes
2008-03-26T12:56:33 <dennda> with AttachFile.py
2008-03-26T12:56:38 <xorAxAx> which is legacy :-)
2008-03-26T12:59:12 <dennda> Ok, what else is responsible for taking the users files and storing them?
2008-03-26T12:59:59 <xorAxAx> only attachfile in 1.6
2008-03-26T13:00:09 <xorAxAx> but in 1.7, that should be moved to the backend
2008-03-26T13:00:32 <dennda> sounds like there has been no move yet?
2008-03-26T13:01:33 <xorAxAx> indeed
2008-03-26T13:02:01 <xorAxAx> i guess heinrich implemented it with a mixture of code that uses the old and new way
2008-03-26T13:02:16 <dennda> I assume that move is part of the Storage-Task?
2008-03-26T13:02:29 <xorAxAx> yes
2008-03-26T13:02:45 <dennda> I was just a bit confused because the ideas page has this as a subtask: make the AttachFile action use new storage items
2008-03-26T13:02:58 <xorAxAx> yes, thats how you can call it
2008-03-26T13:03:31 <dennda> so the AttachFile action doesn't neccessarily corelate to AttachFile.py
2008-03-26T13:03:43 <dennda> or at least it shouldn't in later versions
2008-03-26T13:05:52 <xorAxAx> indeed. now there is some strategy dilemma
2008-03-26T13:06:03 <dennda> hit me
2008-03-26T13:06:32 <xorAxAx> because you would want to have a new UI for the whole item stuff (esp. in the area of attachments) but also use the new backend which might still need some more work for these
2008-03-26T13:06:53 <dennda> well
2008-03-26T13:07:50 <dennda> If I was the person to determine the strategy, I would say: finish the backend before the UI. Having a UI without a working backend is pretty pointless because you'll need to make assumptions on how it will hopefully work
2008-03-26T13:08:01 <xorAxAx> so one has to think about some realisation that does not do too much unrefactorable stuff on the ui side if the task is about the storage itself
2008-03-26T13:08:12 <xorAxAx> yes
2008-03-26T13:08:41 <xorAxAx> also,unittests are a fine way to present a working system as well :)
2008-03-26T13:08:58 <dennda> i learned that the hard way
2008-03-26T13:10:30 <dennda> let me point out what we should have
2008-03-26T13:11:07 <dennda> an abstraction layer that takes items to be stored. depending on the configuration it stores them in a sane way
2008-03-26T13:11:33 <dennda> I saw that hwendell already has some interfaces
2008-03-26T13:12:35 <dennda> but how much has already been realized? does that part already work reliably?
2008-03-26T13:19:47 <dreimark> ThomasWaldmann: will look soon into the copy.
2008-03-26T13:49:44 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3730:9f4fad7e10be 1.7-storage-hwendel/MoinMoin/PageEditor.py: short-cut 'already saved' message like in main branch
2008-03-26T13:49:49 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3741:bb214b348fc5 1.7-storage-hwendel/ (9 files in 9 dirs): merge with main 1.7 branch
2008-03-26T13:49:50 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3742:42d237572c26 1.7-storage-hwendel/MoinMoin/storage/external.py: external storage interface: remove useless and expensive already-exists test
2008-03-26T13:49:51 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3743:98c1a3673b45 1.7-storage-hwendel/MoinMoin/storage/ (backends/common.py external.py): move rename checks up one layer, remove racy 'already exists' check
2008-03-26T13:49:54 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3744:ca130599f7f9 1.7-storage-hwendel/MoinMoin/storage/backends/moin16.py:
2008-03-26T13:49:57 <CIA-39> moin16 backend: make more thread-safe on non-NT, fix rename error case bug
2008-03-26T13:49:59 <CIA-39> Despite claims, the moin16 backend is not thread-safe even with posix
2008-03-26T13:50:01 <CIA-39> filesystem semantics because of the way shutil.rename is implemented,
2008-03-26T13:50:03 <CIA-39> this fixes that.
2008-03-26T13:50:05 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3745:51455db20f43 1.7-storage-hwendel/MoinMoin/storage/backends/moin16.py:
2008-03-26T13:50:08 <CIA-39> moin16 backend: fix unsafe temp file use
2008-03-26T13:50:10 <CIA-39> If tmp_dir was configured to a different partition, os.rename
2008-03-26T13:50:12 <CIA-39> would rightfully fail. Before the previous changeset, that would
2008-03-26T13:50:14 <CIA-39> invoke shutil.copy2 via shutil.move which is unsafe because it
2008-03-26T13:50:16 <CIA-39> doesn't atomically create the target file. Hence, put temp files
2008-03-26T13:50:22 <CIA-39> into the directory they will finally be in to fix this problem.
2008-03-26T13:50:24 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3746:219d7cd805bf 1.7-storage-hwendel/BUGS: document a few of the storage design and implementation bugs
2008-03-26T13:50:26 <dennda> johill is on fixing spree
2008-03-26T13:50:27 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3747:1f55ee6c8897 1.7-storage-hwendel/BUGS: storage bugs: #4 triggers #3
2008-03-26T13:50:30 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3748:60531d1581a0 1.7-storage-hwendel/BUGS: another storage bug
2008-03-26T13:50:34 <johill> dennda: just pushing stuff from yesterday
2008-03-26T13:50:35 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3749:dce371badaeb 1.7-storage-hwendel/MoinMoin/ (search/term.py storage/backends/filesystem.py): search terms must be copied before modified
2008-03-26T13:50:37 <johill> and this morning
2008-03-26T13:50:38 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3750:10cec8618045 1.7-storage-hwendel/MoinMoin/search/term.py: search terms: label with costs, make actual deep copies to allow modifying costs
2008-03-26T13:50:45 <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3751:8bbf4b74e5da 1.7-storage-hwendel/MoinMoin/storage/backends/moin16.py: (log message trimmed)
2008-03-26T13:50:48 <CIA-39> moin16 backend: optimise DELETED search
2008-03-26T13:50:50 <CIA-39> The previous commit (sorting search terms by cost) does the bulk,
2008-03-26T13:50:52 <CIA-39> but this still shaves off a bit of the search especially when the
2008-03-26T13:50:54 <CIA-39> search is for a page name expression that matches a lot of pages.
2008-03-26T13:50:58 <CIA-39> Together, search time goes down from ~45 seconds to less than 1
2008-03-26T13:51:00 <CIA-39> for getPageList(exists=1, filter=re.compile('^Category.*').match)
2008-03-26T13:51:36 <dennda> :)
2008-03-26T13:56:40 <johill> I documented 5 major bugs in the storage code
2008-03-26T13:56:45 <dennda> johill: I see you just changed some things with the storage
2008-03-26T13:57:14 <johill> nothing important I'd think, first steps to tackle those bugs
2008-03-26T13:57:16 <dennda> do you know the answer to my question? :)
2008-03-26T13:57:55 <johill> what is the question?
2008-03-26T13:58:50 <dennda> I am trying to structure my knowledge
2008-03-26T13:59:52 <dennda> I tried to point out what we should have: an abstraction layer that takes items to be stored. depending on the configuration it stores them in a sane way using one of the backends
2008-03-26T14:00:45 <dennda> i saw that hwendell already has some interfaces, but how much of the above has already been realized? does that already work reliably?
2008-03-26T14:00:45 <johill> I think that's my complaint about ItemCollection being instantiated all the time
2008-03-26T14:01:03 <johill> i.e. if ItemCollection was only instantiated once, it would be that layer
2008-03-26T14:01:09 <johill> as it is, it is totally buggy and wrong
2008-03-26T14:01:14 <johill> and doesn't present that layer
2008-03-26T14:02:26 <xorAxAx> johill: yes, i sketched already a solution for the instantiation
2008-03-26T14:02:33 <xorAxAx> using a currying method proxy
2008-03-26T14:03:09 <xorAxAx> i.e. a class that delegates every call to another object with the request object added
2008-03-26T14:03:35 <xorAxAx> but heinrich didnt like it
2008-03-26T14:03:58 <johill> I don't see a need for instantiation at all
2008-03-26T14:04:03 <johill> it could be just instantiated once in the config
2008-03-26T14:04:45 <johill> mind you, the fact that it takes a request parameter is imho wrong too
2008-03-26T14:04:56 <johill> it only needs it for some save helper code that does metadata updates
2008-03-26T14:05:00 <johill> which imho shouldn't be at that layer
2008-03-26T14:06:12 <xorAxAx> johill: the translation needs the reqeuest as well, right?
2008-03-26T14:06:19 <johill> there is no translation at that level
2008-03-26T14:06:23 <johill> it only marks strings for translation
2008-03-26T14:06:27 <xorAxAx> johill: well, in my solution, it would be instantiated in the config and the proxy would live on the request
2008-03-26T14:06:29 <johill> the actual translation is done when the exception is caught
2008-03-26T14:06:52 <johill> exception handling is totally fucked too though, another of the bugs I point out in the BUGS file
2008-03-26T14:09:35 <dennda> What is the current storage code capable of doing reliably? I am still reading the code but it would be great if someone could give a rough overview
2008-03-26T14:10:36 <johill> for what value of reliably?
2008-03-26T14:10:51 <johill> does your 'reliably' include proper error detection? if so, probably not much
2008-03-26T14:11:05 <dennda> let me rephrase the question
2008-03-26T14:11:07 <xorAxAx> reliable as in "dennda wont have to change it"
2008-03-26T14:11:14 <dennda> What works?
2008-03-26T14:11:43 <johill> dennda: regular operation works, but any error cases cannot be reliably detected
2008-03-26T14:12:12 <johill> read the BUGS file for now? :)
2008-03-26T14:12:23 <dennda> And regular operation is: It accepts items that need to be stored and stores it with the only working backend?
2008-03-26T14:12:31 <johill> yeah
2008-03-26T14:12:37 <dennda> nothing else?
2008-03-26T14:12:39 <johill> as long as nobody else tries to do the same thing at the same time
2008-03-26T14:12:46 <dennda> oh
2008-03-26T14:12:58 <johill> no, it can do list_items too and stuff like that
2008-03-26T14:13:12 <johill> but e.g. create_item is fundamentally flawed right now
2008-03-26T14:15:18 <johill> or if you want to put it another way, the way locking is done is fundamentally flawed
2008-03-26T14:15:40 <dennda> ok
2008-03-26T14:15:47 <dennda> so it should be made more atomic
2008-03-26T14:17:36 <johill> not necessarily
2008-03-26T14:17:46 <johill> can you read my BUGS file before I explain everything?
2008-03-26T14:17:58 <dennda> sure
2008-03-26T14:34:08 <johill> let me know if you have any questions on that
2008-03-26T14:40:08 <johill> xorAxAx: I was wrong, storage doesn't have any extra conflict resolution bugs, the bug is actually in all versions, see http://moinmo.in/MoinMoinBugs/ConflictDetectionNotWorkingForNewPages
2008-03-26T14:47:35 <dreimark> ThomasWaldmann: meta was wrong, no bug (besided I don't know why I have had it wrong)
2008-03-26T15:11:01 <dreimark> johill: you can use lock instead of warn for edit_locking.
2008-03-26T15:12:02 <johill> and?
2008-03-26T15:12:16 <johill> that doesn't really change anything
2008-03-26T15:12:20 <dreimark> then you cant open a second version of the page
2008-03-26T15:12:29 <johill> oh you mean as a workaround?
2008-03-26T15:12:35 <johill> I'm not sure that actually works in that case
2008-03-26T15:12:40 <dreimark> right, I addred that on the page
2008-03-26T15:12:41 <johill> it might still open it
2008-03-26T15:13:01 <johill> because the page doesn't exist so I'm not sure it properly adds an edit lock
2008-03-26T15:13:10 * dreimark tries
2008-03-26T15:13:24 <johill> in practice though, nobody does that, do they?
2008-03-26T15:13:40 <johill> I certainly don't run with actual edit locks
2008-03-26T15:14:25 <dreimark> we do
2008-03-26T15:14:37 <dreimark> at all of our wikis
2008-03-26T15:14:56 <xorAxAx> because your colleagues dislike merging, dreimark? :)
2008-03-26T15:15:14 <dreimark> but you are right this workaround does not help for new pages
2008-03-26T15:15:36 * xorAxAx &
2008-03-26T15:16:38 <dennda> xorAxAx: what's that & supposed to mean? :)
2008-03-26T15:17:40 <johill> dennda: I take it to mean that he re-executes himself in the background ;)
2008-03-26T15:18:05 <dreimark> hmm, I think lock is broken, will check this
2008-03-26T15:18:11 * dreimark needs a coffee
2008-03-26T15:18:17 <dennda> ;)
2008-03-26T15:18:30 <johill> dreimark: I'll remove your workaround again then
2008-03-26T15:18:37 <dreimark> ok
2008-03-26T15:19:10 <dreimark> seems I can edit with lock defined an existing page too
2008-03-26T15:19:19 <johill> huh
2008-03-26T15:19:29 <johill> I think I tested that in -storage and it worked as expected
2008-03-26T15:20:15 <dreimark> edit_locking = 'lock 10'
2008-03-26T15:21:38 <dreimark> and testing with same user on two different browsers
2008-03-26T15:21:54 <dreimark> I get nderen Benutzern wird die Editierung dieser Seite bis 2008-03-26 15:31:09 verweigert.
2008-03-26T15:21:59 <dreimark> but it is not locked
2008-03-26T15:22:17 <dreimark> btw. 1.6.2
2008-03-26T15:23:02 <johill> dunno
2008-03-26T15:23:18 <johill> maybe I didn't test it, I think I saw something like that but not sure
2008-03-26T15:24:14 <zenhase> hmm, friend of mine tries to motivate me and himself into applying for another soc this year
2008-03-26T15:25:52 <johill> with the task to merge the work from last year? ;)
2008-03-26T15:26:18 <zenhase> well, it wasn't quite finished and mergeable imo
2008-03-26T15:27:12 <zenhase> i am not really sure about this ... decisions decisions ... and 5 days to go :o
2008-03-26T15:37:57 <xorAxAx> dennda: i went to the background
2008-03-26T15:39:10 <zenhase> fg xorAxAx
2008-03-26T15:39:11 <zenhase> :)
2008-03-26T15:39:39 <xorAxAx> too late :)
2008-03-26T15:40:12 <zenhase> well, i could still renice you
2008-03-26T15:40:34 <xorAxAx> you are not root
2008-03-26T15:41:54 <zenhase> yeah .. and i don't own the process
2008-03-26T15:41:54 <dennda> try exploiting the vmsplice root exploit. maybe he didn't update :)
2008-03-26T16:19:41 <dreimark> xorAxAx: no the merging in general is not the problem. we have both options sometime ago discussed and most people decided "if one edits others should be blocked to work on the same page"
2008-03-26T16:20:47 <dreimark> and in real live there are only few pages where concurrent edit should be blocked
2008-03-26T16:21:01 <dreimark> but we can only do a global decission.
2008-03-26T16:21:16 <xorAxAx> dreimark: yeah, and i guess its not so bad like with VSS
2008-03-26T16:21:19 <xorAxAx> because of the timeout
2008-03-26T16:24:33 <dreimark> xorAxAx: please expand VSS
2008-03-26T16:24:33 <dennda> def new_revision(self, revno=0): <-- creates a new revision of a storageitem, but you can tell the method which revision number you'd like. I don't see why that is neccessary (or even useful)
2008-03-26T16:25:46 <xorAxAx> dreimark: visual source safe
2008-03-26T16:49:53 <dennda> metadata for storage items is not metadata about the actual file that is stored but provides information on who changed it, when, etc. is that correct?
2008-03-26T16:55:19 <ThomasWaldmann> dennda: well, that also is metadata about the file
2008-03-26T16:56:42 <dennda> like size and such?
2008-03-26T16:57:31 <ThomasWaldmann> not sure if we store the size into metadata. could make sense or not.
2008-03-26T16:58:21 <ThomasWaldmann> everything that was in edit-log in 1.6 should be in item rev's metadata
2008-03-26T16:58:50 <dennda> ok
2008-03-26T16:59:21 <ThomasWaldmann> (but it is not limited to that, I think we could also store categories there, or a hash of the file content, or ...)
2008-03-26T17:15:07 <mvirkkil> Hi
2008-03-26T17:15:19 <mvirkkil> back from easter holidays
2008-03-26T17:15:40 <mvirkkil> I see I forgot to mark my self as being away, so I missed anything you might have written to me :(
2008-03-26T17:17:56 <ThomasWaldmann> wb mvirkkil :)
2008-03-26T17:18:42 <ThomasWaldmann> mvirkkil: i did some stuff on the 1.7-docbook branch
2008-03-26T17:19:48 <napi> Afternoon all
2008-03-26T17:19:50 <mvirkkil> There's a 1.7-docbook branch? Maybe you mean 1.6-docbook?
2008-03-26T17:20:10 <mvirkkil> wow, there is a 1.7-docbook
2008-03-26T17:20:34 <ThomasWaldmann> it's new. i created it because that stuff is not sure to go into 1.7 main branch, until it is tested and documented enough.
2008-03-26T17:20:46 <mvirkkil> What stuff?
2008-03-26T17:20:50 <ThomasWaldmann> (unlike your formatter fixes)
2008-03-26T17:20:55 <mvirkkil> aah, ok
2008-03-26T17:21:04 <ThomasWaldmann> ImportDocBook
2008-03-26T17:21:09 <mvirkkil> Yeah.
2008-03-26T17:21:13 <mvirkkil> I've played with it a bit
2008-03-26T17:21:19 <mvirkkil> Ported some of the xslt over
2008-03-26T17:21:44 <ThomasWaldmann> I copied the action and the xslt from 1.6. I fixed the xslt to generate new wiki link/transclusion markup.
2008-03-26T17:22:14 <mvirkkil> Do I have commit rights to that branch?
2008-03-26T17:22:25 <ThomasWaldmann> also, I made it create {{{#!wiki warning ... }}} for the admonitions
2008-03-26T17:22:46 <ThomasWaldmann> try :)
2008-03-26T17:22:52 <mvirkkil> Wow. Why this sudden interest?
2008-03-26T17:23:18 <ThomasWaldmann> i helped two guys struggling to get that working :)
2008-03-26T17:23:21 <dennda> I understand that storage can store Pages, Items and Users. I expect Pages to be the actual wiki pages, items to be any other items on those pages like images or zip-files, but I don't quite know how you would store a "User". Any simple example or use case?
2008-03-26T17:24:01 <ThomasWaldmann> they said they can test it enough to tell whether it works ok.
2008-03-26T17:25:12 <mvirkkil> I know there are some issues with nested lists sometimes :/
2008-03-26T17:26:00 <mvirkkil> The problem is that list items in moin markup cannot contain block elements like tables, while in docbook they can.
2008-03-26T17:26:33 <mvirkkil> Doing the conversion from a very complex format to a simplified is always hard, but there are some cases which are fixable, but not fixed.
2008-03-26T17:27:42 <mvirkkil> School is ramping up again, so I don't know how much time I'll have with this, but I'll try to have a look.
2008-03-26T17:28:30 <mvirkkil> But more importantly: Is there any issues I should know about with the formatter?
2008-03-26T17:30:41 <ThomasWaldmann> i didnt test it.
2008-03-26T17:31:34 <ThomasWaldmann> btw, 1.7.0 beta/rc/release is soon. so if you could bring the docs into a good state so we can include some stuff, that would be great.
2008-03-26T17:31:53 <ThomasWaldmann> see the MoinMoin:DocBook page for todo
2008-03-26T17:32:18 <ThomasWaldmann> (maybe just needs some update?)
2008-03-26T17:37:46 <HongJun> ThomasWaldmann: I have list some information about what my idea of REST interface of Moin on the wiki
2008-03-26T17:38:25 <ThomasWaldmann> i had a quick look
2008-03-26T17:38:59 <HongJun> ThomasWaldmann: and I wil work out a detailed plan in a few days
2008-03-26T17:39:21 <ThomasWaldmann> it didn't have much concrete stuff about your moin project
2008-03-26T17:39:50 <HongJun> ThomasWaldmann: I want to get you know REST first
2008-03-26T17:41:50 <HongJun> ThomasWaldmann: If you agree that REST will be useful for Moin, I will work out the detail application for the SoC, please give me your oppion
2008-03-26T17:42:17 <ThomasWaldmann> i need details
2008-03-26T17:42:24 <TheSheep> making it restful is going to tocuh practically every plugin in there
2008-03-26T17:42:33 <TheSheep> touch
2008-03-26T17:43:14 <ThomasWaldmann> currently it just looks like a big url change
2008-03-26T17:44:50 <TheSheep> ThomasWaldmann: if you really want it to be REST, it would probably require changing the semantics of some plugins, e.g. replacing 'actions' with 'views'
2008-03-26T17:44:54 <HongJun> TheSheep: I don't think so, there is a Django-Rest-interface that makes the django project restful, It's just an interface API, don't touch the project any more
2008-03-26T17:45:23 <TheSheep> HongJun: moin is not as strictly separated into mvc
2008-03-26T17:47:06 <HongJun> TheSheep: I see. Using restful api to change the actions modul may make the operaction simple and clear
2008-03-26T17:47:41 <TheSheep> HongJun: most links are generated in the theme, for example
2008-03-26T17:48:09 <HongJun> We do't need to change the architecture of Moin, just implement an restful api as a plugin
2008-03-26T17:48:26 <TheSheep> although I think the actual links are made using a method of Page
2008-03-26T17:48:59 <HongJun> theme is the resouce in REST, then it will have a URI
2008-03-26T17:49:35 <TheSheep> HongJun: try looking at the source code of some of the themes
2008-03-26T17:50:23 <HongJun> ok, i will see tomorrow, too later now, I will go bed, bye!
2008-03-26T17:50:35 <TheSheep> good night
2008-03-26T17:50:40 <TheSheep> dream of moin ;)
2008-03-26T17:50:55 <ThomasWaldmann> hehe, gn :)
2008-03-26T17:51:20 <ThomasWaldmann> don't look at send_page before dreaming :P
2008-03-26T17:52:24 <TheSheep> maybe a dispatcher plugin wouldn' be such a bad idea ;)
2008-03-26T17:52:51 <TheSheep> could do both actions and wiki farms
2008-03-26T17:53:50 <TheSheep> naah
2008-03-26T17:54:36 <ThomasWaldmann> i guess we do that after wsgi refactoring :)
2008-03-26T17:55:42 <TheSheep> right, then the actions can be just separate wsgi subapplications
2008-03-26T18:05:35 <dennda> anyone? :)
2008-03-26T18:08:16 <ThomasWaldmann> user is likely the user profile data
2008-03-26T18:08:52 <ThomasWaldmann> item should be the term used for pages and attachments after unification of them
2008-03-26T18:10:51 <dennda> i see, ok
2008-03-26T18:13:01 <dreimark> bbl
2008-03-26T20:54:49 <dennda> The DeveloperApplication Template has two different points: "Project(s) you apply for and your ideas for them" and "What projects do you apply for? Can be some of our suggestions or a suggestion of your own. In any case, tell us YOUR ideas about it." <-- How do they differ? Do they at all?
2008-03-26T20:58:17 <xorAxAx> no, feel free to remove one
2008-03-26T20:58:28 <xorAxAx> also i wonder why its using the plural form
2008-03-26T20:58:54 <dennda> well, you can apply for multiple tasks, can't you?
2008-03-26T20:59:15 <dennda> but I assume it's better to file several applications in that case
2008-03-26T21:06:42 <dennda> xorAxAx: what do you think of this "timeline": a) make sure the current code works, refactor it, fix remaining bugs (e.g. johills BUGS); b) while doing that also improve current documentation and correct it where appropriate; c) write tests for the current code and for the code that will be written; d) make sure new backends can be added easily e) start implementing new features, i.e. start with adding ACL support; f) let attachfile action use new s
2008-03-26T21:07:22 <xorAxAx> yes, you need multiple apps
2008-03-26T21:07:44 <xorAxAx> dennda: your message was truncated
2008-03-26T21:07:53 <dennda> where?
2008-03-26T21:08:01 <xorAxAx> and your ctcp version reply wasnt very enlightening
2008-03-26T21:08:05 <xorAxAx> "let attachfile action use new s"
2008-03-26T21:08:29 <dennda> f) let attachfile action use new storage backend; g) write sql db backend; h) when everything works, write migration scripts for attachment -> storageitem and for switching between backends
2008-03-26T21:09:55 <xorAxAx> nice that you assigned letters. :)
2008-03-26T21:10:23 <dennda> it's easier to swap them if you think this order is completely wrong
2008-03-26T21:10:30 <dennda> (when having letters)
2008-03-26T21:10:31 <dennda> :)
2008-03-26T21:10:44 <xorAxAx> i would go c a b e f h d (not sure what that is about) g
2008-03-26T21:12:43 <dreimark> bb
2008-03-26T21:13:25 <dennda> xorAxAx: yes, d should come before g. they are related
2008-03-26T21:14:21 <napi> ah
2008-03-26T21:14:27 <napi> I was planning on using numbers
2008-03-26T21:14:29 <napi> letters preferable? :p;
2008-03-26T21:15:01 <dreimark> if you dont use I,II,III numbers are ok too ;)
2008-03-26T21:15:25 * napi takes all latin references out
2008-03-26T21:15:25 <napi> hehe
2008-03-26T21:16:17 <dennda> xorAxAx: I am comfortable with your ordering, too
2008-03-26T21:16:46 <dennda> I'll prefer yours since you are more experienced, if you don't mind :)
2008-03-26T21:17:06 <dreimark> dennda: what is the meaning of johills BUGS ?
2008-03-26T21:17:30 <dreimark> dennda: you can't refactor it without proper tests
2008-03-26T21:17:41 <dennda> dreimark: pull in the latest changes from hwendells branch
2008-03-26T21:17:46 <dennda> johill created a BUGS file
2008-03-26T21:18:03 <dennda> dreimark: ok, I understand why that is. c) comes first then
2008-03-26T21:18:30 <dreimark> (ah, it sounded a bit strange, easy to misunderstood by the discussions some days ago)
2008-03-26T21:40:41 <dreimark> http://groups.google.com/group/google-summer-of-code-mentors-list/browse_thread/thread/80272773e20f32ab?hl=en
2008-03-26T22:01:17 * dreimark investigate edit_locking in 1.6
2008-03-26T22:03:00 <gizmach> hi all
2008-03-26T22:03:02 <gizmach> :)
2008-03-26T22:03:17 <xorAxAx> hi melita
2008-03-26T22:03:43 <gizmach> hi Alexander
2008-03-26T22:04:33 <xorAxAx> looks like copy&paste :-)
2008-03-26T22:04:55 <xorAxAx> (because of the capital A :))
2008-03-26T22:05:15 <xorAxAx> gizmach: have you collected any questions? :)
2008-03-26T22:05:54 <napi> xorAxAx, had any other interest in the web based installation and admin idea yet?
2008-03-26T22:06:33 <xorAxAx> no, napi.
2008-03-26T22:07:19 <gizmach> xorAxAx: trying to make some "smart" questions :D
2008-03-26T22:07:41 <napi> hmm
2008-03-26T22:07:54 <xorAxAx> we also accept dumb ones
2008-03-26T22:08:21 <gizmach> xorAxAx: :D that's ok but I've asked so many dumb :)
2008-03-26T22:08:26 <xorAxAx> hehe
2008-03-26T22:13:05 <napi> Want to do the that project but need to know sooo much more about moin than I have time to find out before application deadline :(
2008-03-26T22:14:44 <gizmach> napi: :) join the club ;)
2008-03-26T22:14:58 <TheSheep> napi: try installing one for starters
2008-03-26T22:15:03 <napi> about the web installation? or needing to know more about moin?
2008-03-26T22:15:26 <xorAxAx> napi: its really not difficult
2008-03-26T22:15:42 <xorAxAx> napi: just download moinmoin as a tgz file, extract, make sure python is installed, run moin.py
2008-03-26T22:23:33 <gizmach> xorAxAx: I couldn't find moin.py but setup.py hopefully installed it
2008-03-26T22:23:42 <dreimark> gizmach: no
2008-03-26T22:23:42 <gizmach> :D
2008-03-26T22:23:59 <dreimark> just go into the moin-1.6.2 dir
2008-03-26T22:24:05 <dreimark> and call ./moin.py
2008-03-26T22:24:08 <dreimark> that's all
2008-03-26T22:24:13 <xorAxAx> yes, you need at least a moin 1.6.x
2008-03-26T22:24:39 <dreimark> or take the 1.7 version from hg
2008-03-26T22:24:40 <gizmach> oh I got 1.5.8
2008-03-26T22:24:45 <gizmach> :)
2008-03-26T22:24:51 <dreimark> 1.5.8 is too old
2008-03-26T22:24:53 <gizmach> that sounds better
2008-03-26T22:24:54 <gizmach> :D
2008-03-26T22:25:03 <gizmach> I found it on moin website
2008-03-26T22:26:08 <dennda> damn, 7678 words :)
2008-03-26T22:26:36 <dennda> need to cut it
2008-03-26T22:34:10 <xorAxAx> dennda: did it punish/truncate?
2008-03-26T22:34:28 <xorAxAx> i heard about people who didnt have any issues with a long text
2008-03-26T22:35:26 <dennda> I just removed 500 characters
2008-03-26T22:35:32 <dennda> boy, that was more than I expected
2008-03-26T22:37:29 <dennda> no I didn't send it yet
2008-03-26T22:37:38 <dennda> I will in a minute
2008-03-26T22:43:40 <gizmach> dennda: for which project?
2008-03-26T22:52:18 <dennda> 7504 characters
2008-03-26T22:52:19 <dennda> !!!
2008-03-26T22:52:24 <dennda> gizmach: storage backend
2008-03-26T22:53:13 <dreimark> johill: I can't reproduce http://moinmo.in/MoinMoinBugs/ConflictDetectionNotWorkingForNewPages
2008-03-26T22:53:27 <dreimark> not in 1.6.2 localhost or on MM
2008-03-26T22:53:40 <grzywacz> moin
2008-03-26T22:53:49 <xorAxAx> dennda: do it like hebrew, remove a few vocals :)
2008-03-26T22:53:54 <dreimark> I get always to revisions and no edit conflict
2008-03-26T22:54:13 <dreimark> s/to/two/
2008-03-26T22:54:21 <xorAxAx> dreimark: well, thats the bug
2008-03-26T22:54:26 <xorAxAx> there should be a conflict
2008-03-26T22:54:41 <xorAxAx> i think editlocks are not necessary to detect the defect
2008-03-26T22:54:46 <xorAxAx> s/defect/conflict/
2008-03-26T23:00:12 <dreimark> ahh yeah that's true
2008-03-26T23:08:25 <dennda> xorAxAx: You got mail. :)
2008-03-26T23:09:34 <dennda> I hope it's displayed correctly
2008-03-26T23:09:41 <dennda> I used some tabs to structure the text
2008-03-26T23:09:55 <xorAxAx> dennda: the app doesnt send emails in case of applications
2008-03-26T23:10:00 <xorAxAx> we have to poll manually
2008-03-26T23:10:39 <dennda> don't take my "you got mail" too literally :)
2008-03-26T23:17:05 <xorAxAx> ok, got it. might have a peek into it after my exam tomorrow :-)
2008-03-26T23:17:35 <dennda> oh
2008-03-26T23:17:38 <dennda> what kind of exam?
2008-03-26T23:18:01 <xorAxAx> its called an "individualklausur". i will be writing about net neutrality for 4 hours
2008-03-26T23:18:08 <xorAxAx> it belongs to my media studies minor
2008-03-26T23:18:36 <dennda> _("Diplomstudiengang?")
2008-03-26T23:18:40 <xorAxAx> no
2008-03-26T23:19:18 <xorAxAx> bachelor
2008-03-26T23:20:42 <gizmach> xorAxAx: what are you studying ?
2008-03-26T23:20:53 <xorAxAx> gizmach: computer science :-)
2008-03-26T23:21:02 <xorAxAx> with the aforementioned minor
2008-03-26T23:21:12 <dennda> xorAxAx: which semester?
2008-03-26T23:21:19 <xorAxAx> 5th
2008-03-26T23:21:29 <dennda> xorAxAx: may I ask how old you are?
2008-03-26T23:21:36 <xorAxAx> older than you :-)
2008-03-26T23:21:39 <xorAxAx> 23
2008-03-26T23:21:48 <dennda> (I'm always interested in the people I talk to :))
2008-03-26T23:22:00 <dennda> ah ok, so bachelors degree next semester?
2008-03-26T23:22:04 <xorAxAx> yep
2008-03-26T23:22:10 <dennda> which university?
2008-03-26T23:22:16 <xorAxAx> paderborn
2008-03-26T23:23:19 <dreimark> oh that's another magical number
2008-03-26T23:23:25 <dreimark> xorAxAx:
2008-03-26T23:23:34 <dennda> xorAxAx: good luck for your exam!
2008-03-26T23:23:46 * dreimark wishes too
2008-03-26T23:24:32 <xorAxAx> thanks!
2008-03-26T23:24:46 * dennda needs to finish 12432935257235 things for university to monday
2008-03-26T23:25:01 <dennda> I fear that'll be my only application, thus
2008-03-26T23:25:15 <dennda> no time for other code
2008-03-26T23:26:05 <dennda> is my application the first or are there already?
2008-03-26T23:26:22 <xorAxAx> the first one
2008-03-26T23:26:41 <dennda> oh
2008-03-26T23:26:55 <dennda> ok
2008-03-26T23:27:04 <dennda> just noticed I were in your area two weeks ago
2008-03-26T23:27:13 <dennda> visited a friend in goettingen
2008-03-26T23:30:20 <xorAxAx> my sister lives in göttingen. and i think koblenz<->pb is not much more far away than pb<->koblenz :)
2008-03-26T23:30:43 <gizmach> xorAxAx: good luck
2008-03-26T23:30:55 <xorAxAx> ah, 141 vs 300
2008-03-26T23:30:57 <xorAxAx> ok, a bit more
2008-03-26T23:31:08 <xorAxAx> (km by car)
2008-03-26T23:31:11 <xorAxAx> gizmach: thanks
2008-03-26T23:31:36 <dennda> xorAxAx: ?
2008-03-26T23:32:02 <xorAxAx> dennda: göttingen is pretty far away :)
2008-03-26T23:32:12 <dennda> the distance is the same for pb<->koblenz and koblenz<->pb :)
2008-03-26T23:32:19 <dennda> from paderborn?
2008-03-26T23:32:19 <xorAxAx> ah!
2008-03-26T23:32:30 <xorAxAx> i meant pb<->koblenz and pb<->göttingen
2008-03-26T23:32:34 <dennda> looks quite close on the map
2008-03-26T23:32:34 <xorAxAx> yes
2008-03-26T23:33:18 <xorAxAx> but if you visit him another time, you are invited to visit paderborn! :)
2008-03-26T23:34:49 <dennda> thanks, sounds great :)
2008-03-26T23:35:05 <dennda> I'm on a mission to test every cantine
2008-03-26T23:35:42 <dennda> goettingen has the worst canteen I've seen so far :)
2008-03-26T23:36:23 <xorAxAx> hehe
2008-03-26T23:38:58 <xorAxAx> i should invite my sister to let her check our cafeteria
2008-03-26T23:48:34 <dreimark> dennda: the one in wuppertal is quite good :)
2008-03-26T23:48:52 <dennda> havn't been there
2008-03-26T23:49:00 <dennda> ours is close to fantastic
2008-03-26T23:50:21 <dreimark> I have worked some years ago in Katlenburg-Lindau
2008-03-26T23:50:43 <dreimark> that's north from Goettingen MoinMoin: MoinMoinChat/Logs/moin-dev/2008-03-26 (last edited 2008-03-25 23:15:02 by IrcLogImporter)
