MoinMoin: MoinMoinChat/Logs/moin-dev/2008-03-31

2008-03-31T00:21:24  <zenhase> dreimark: you commented on my send_page proposal :)
2008-03-31T00:21:47  <zenhase> dreimark: want to discuss the details tomorrow?
2008-03-31T00:28:59  <ThomasWaldmann> zenhase: he's gone sleeping and it is already "tomorrow" :)
2008-03-31T00:30:42  <zenhase> well, tomorrow in the night-hours is the same as 'after a good nights sleep' in my language/symbol system
2008-03-31T00:30:51  <zenhase> :)
2008-03-31T00:32:42  * zenhase wonders if he should rename the section in his SOC2007 designpaper entitled 'Page.send_page considered unrefactorable'
2008-03-31T00:37:33  <xorAxAx> at least that would remind me at last year
2008-03-31T00:38:33  <zenhase> xorAxAx: hmm?
2008-03-31T00:38:49  <xorAxAx> when you came to similar conclusions :)
2008-03-31T00:39:05  <xorAxAx> btw, one thing i really value nowadays is my whiteboard
2008-03-31T00:39:26  <xorAxAx> i guess a whiteboard is a very important dependency for such a refactoring task :)
2008-03-31T00:39:35  <zenhase> yeah
2008-03-31T00:39:54  <zenhase> i already have a 'pinnwand' (what is it called in englisch?)
2008-03-31T00:40:05  <zenhase> but a whiteboard make a better brainstorming tool
2008-03-31T00:40:17  <zenhase> well, i also have a moleskine notebook
2008-03-31T00:40:38  <xorAxAx> the nice thing about a whiteboard is the possibility to easily clean parts of it
2008-03-31T00:41:04  <zenhase> sometimes i would like that for my brain
2008-03-31T00:41:29  <zenhase> sadly you can't clean it ... you can just start doodling all over it again
2008-03-31T00:47:12  <TheSheep> xorAxAx: that may be a disadvantage
2008-03-31T00:47:26  <xorAxAx> TheSheep: hehe :)
2008-03-31T00:50:26  <xorAxAx> zenhase: which values does the " moleskine notebook" have for you?
2008-03-31T00:51:31  <zenhase> xorAxAx: right now i haven't filled this ____ in the notebook in
2008-03-31T00:51:50  <xorAxAx> zenhase: what do you mean?
2008-03-31T00:52:02  <zenhase> but since i am still experimenting with its potential as a tool for my brain, it's not particular valuable right now
2008-03-31T00:52:13  <xorAxAx> you mean its empty?
2008-03-31T00:52:15  <zenhase> xorAxAx: i thought you meant the first page
2008-03-31T00:52:26  <zenhase> where you can insert your address
2008-03-31T00:52:31  <xorAxAx> no, i meant to ask why you value it
2008-03-31T00:52:41  <zenhase> and state the amount of money you pay to the honest finder
2008-03-31T00:53:12  <xorAxAx> i am lucky not to knew this detail until right now
2008-03-31T00:53:18  <xorAxAx> s/knew/have known/
2008-03-31T00:53:24  <zenhase> hmm?
2008-03-31T00:54:09  <xorAxAx> zenhase: well, by reading up some stuff, i get the impression that every moleskine customer was kidded a bit by a well working marketing dept. ;-)
2008-03-31T00:54:41  <zenhase> xorAxAx: why do you think that?
2008-03-31T00:54:57  <zenhase> it's a notebook
2008-03-31T00:55:00  <zenhase> nothing else
2008-03-31T00:55:02  <xorAxAx> zenhase: because the customers value the brand much higher than the actual product :)
2008-03-31T00:55:12  <zenhase> ah, you mean that
2008-03-31T00:55:29  <xorAxAx> well, then why do you feel urged to say that it is a moleskine notebook?
2008-03-31T00:55:30  <zenhase> i like the feel it has
2008-03-31T00:55:42  <xorAxAx> you mean the cover?
2008-03-31T00:55:48  <zenhase> yeah
2008-03-31T00:55:55  <zenhase> die haptik!
2008-03-31T00:55:57  <zenhase> :)
2008-03-31T00:55:58  <xorAxAx> ok, well :)
2008-03-31T00:56:05  <zenhase> and this rubberband
2008-03-31T00:56:33  <zenhase> also the cover is quite robust and it has a cool small pocket in the back-cover
2008-03-31T00:56:41  <zenhase> i have a sd-card in there
2008-03-31T00:56:57  <xorAxAx> maybe its because i am not a notebook user ...
2008-03-31T00:57:27  <zenhase> well, i am a quite mobile person
2008-03-31T00:57:40  <xorAxAx> but seriously, a notebook as a lifestyle product, conflicts with my beliefs somehow, not sure why
2008-03-31T00:57:48  <zenhase> and also contantly overloaded on information
2008-03-31T00:58:02  <xorAxAx> zenhase: what kind of stuff do you write down there?
2008-03-31T00:58:04  <zenhase> and i don't trust computers
2008-03-31T00:58:14  <zenhase> dates, sudden ideas
2008-03-31T00:58:24  <zenhase> stuff that bothers me
2008-03-31T00:58:36  <xorAxAx> the latter sounds a bit like diary-like usage
2008-03-31T00:58:44  <zenhase> sure does
2008-03-31T00:58:44  <xorAxAx> hmm, often i am angry about myself because i did not write ideas down
2008-03-31T00:58:49  <zenhase> yeah
2008-03-31T00:59:27  <zenhase> also i started writing into it, when i noticed that i act in a stupid way in a situation
2008-03-31T00:59:37  <zenhase> trying to get rid of bad habits
2008-03-31T00:59:54  <xorAxAx> yeah, sounds pretty much like a classic diary :)
2008-03-31T00:59:56  <zenhase> by identifying them and writing them down to think about them later
2008-03-31T01:00:15  <xorAxAx> ... which i havent had yet either
2008-03-31T01:00:17  <zenhase> no, if it was published it would look more like an insane twitter.com-stream
2008-03-31T01:00:23  <xorAxAx> LOL
2008-03-31T01:01:22  <xorAxAx> yeah, i am not a twitter user but i can think of a particular hardcore twitter user :)
2008-03-31T01:01:38  <zenhase> :)
2008-03-31T01:02:34  <xorAxAx> http://tinyurl.com/2gg7nv
2008-03-31T01:02:55  <zenhase> actually i also started using small icons at the border to indicate what an item actually is
2008-03-31T01:03:09  <xorAxAx> reminds me of datebk
2008-03-31T01:03:10  <zenhase> like small light bulbs or big questionmarks or clocks
2008-03-31T01:03:35  <zenhase> but this is only because it started reminding myself of this twitter-allegory
2008-03-31T01:18:27  <xqj37> I use cheap brown a4 paper
2008-03-31T01:19:05  <zenhase> for your diary/notes/whatever?
2008-03-31T01:20:00  <xqj37> notes mostly
2008-03-31T01:20:27  <TheSheep> hehe, they sell these recycled paper blocks clled 'a killogram of paper' here
2008-03-31T01:20:51  <TheSheep> it is also cool for pencil sketches because it has nice texture
2008-03-31T01:21:59  <xqj37> I'm not diciplined enough for keeping a diary
2008-03-31T01:22:17  <TheSheep> xqj37: kep a blog ;)
2008-03-31T01:22:23  <xqj37> jeah but I really specfic about the type of paper
2008-03-31T01:22:24  <TheSheep> keep
2008-03-31T01:22:26  <zenhase> difference?
2008-03-31T01:22:37  <zenhase> except one is on paper, the other online
2008-03-31T01:22:44  <xqj37> nothing interesting to write about
2008-03-31T01:23:13  <TheSheep> zenhase: no leaving blank pages to fill in later
2008-03-31T01:23:35  <xqj37> or to depressing
2008-03-31T01:23:42  <TheSheep> xqj37: then you should start collecting interesting information :)
2008-03-31T01:23:50  <xqj37> "dear blog, today I did not apply for google soc"
2008-03-31T01:23:52  <xqj37> ;)
2008-03-31T01:24:07  <xqj37> jeah but everybody can find interesting information
2008-03-31T01:24:19  <xqj37> it has to be UNIQUE interesting information
2008-03-31T01:24:21  <TheSheep> xqj37: sure, but nobody others unless it's on a blog
2008-03-31T01:24:26  <TheSheep> bothers
2008-03-31T01:24:30  <zenhase> i think that blogging about stuff you find is not the way to go
2008-03-31T01:24:37  <TheSheep> xqj37: selection really can add value
2008-03-31T01:24:43  <zenhase> it should be more creative than that
2008-03-31T01:24:50  <zenhase> i really love those science-blogs
2008-03-31T01:25:04  <TheSheep> zenhase: you think they make up everything themselves? :)
2008-03-31T01:25:39  <xqj37> I like to read blogs where people make up stuff themselves
2008-03-31T01:25:51  <zenhase> TheSheep: well, the ones i read mostly do ... there are first hand experiences in the fields of neuro-medicine
2008-03-31T01:25:55  <TheSheep> real life is boring. you either need to: a) compress long spans of time into short articles, b) compress experience of many people into one article
2008-03-31T01:26:13  <zenhase> TheSheep: and there is often huge discussions of current research
2008-03-31T01:26:28  <zenhase> but writing in an reflective way about something is hard work
2008-03-31T01:26:30  <xqj37> scott adams blog is the only non-science non-tech blog I like
2008-03-31T01:26:56  <TheSheep> zenhase: of course it is
2008-03-31T01:26:59  <zenhase> most people just dump stuff like: today i found this cool how-to on how to make your emacs into a planning tool [link]
2008-03-31T01:27:12  <xqj37> yeah exactly
2008-03-31T01:27:17  <TheSheep> zenhase: but you learn by doing
2008-03-31T01:27:30  <TheSheep> zenhase: you won't get good at it without trying
2008-03-31T01:27:38  <zenhase> TheSheep: good point
2008-03-31T01:27:50  <TheSheep> zenhase: plus, reflecting is very healthy for your mind
2008-03-31T01:27:50  <zenhase> but you can practice offline
2008-03-31T01:28:15  <TheSheep> zenhase: writing for audience reuires different approach
2008-03-31T01:28:19  <zenhase> TheSheep: reflecting is ... yes ... but i was talking about the blind hyping that often takes place in the blogosphere
2008-03-31T01:28:34  <xqj37> TheSheep: well it didn't help kafka  ;)
2008-03-31T01:28:37  <TheSheep> zenhase: 90% of anything is crap :)
2008-03-31T01:28:42  <zenhase> TheSheep: yeah
2008-03-31T01:28:51  <zenhase> TheSheep: that is lifes 10% rule
2008-03-31T01:29:18  <zenhase> still i don't like to contribute to those 90% crap
2008-03-31T01:29:29  <zenhase> not on a worldwide scale
2008-03-31T01:29:35  <zenhase> which the internet certainly is
2008-03-31T01:29:36  <zenhase> ;)
2008-03-31T01:29:38  <TheSheep> zenhase: note that good blogs are either written by people who blogged a lot beofre, or who write by profession
2008-03-31T01:29:45  <zenhase> TheSheep: es
2008-03-31T01:29:47  <zenhase> TheSheep: yes
2008-03-31T01:29:48  <TheSheep> you contribute to both
2008-03-31T01:29:54  <zenhase> oh wait
2008-03-31T01:30:03  <zenhase> you meant profession like as in 'being an author'
2008-03-31T01:30:19  <zenhase> i thought about 'writing about your profession'
2008-03-31T01:30:31  <TheSheep> zenhase: a book writer, journalist, documentation writer, scientific paper writer
2008-03-31T01:31:08  <TheSheep> zenhase: as an additional benefit, keeping a blog will make you write better comments in your code :)
2008-03-31T01:31:09  <zenhase> well, but i plan to join the latter ones one day
2008-03-31T01:31:38  <TheSheep> zenhase: you know this article? http://37days.typepad.com/37days/2006/10/polish_your_mud.html
2008-03-31T01:31:44  <xqj37> well the common scientist is good enough
2008-03-31T01:31:50  <xorAxAx> TheSheep: i dont understand the point of "note that good blogs are either written by people who blogged a lot beofre"
2008-03-31T01:31:51  <xqj37> as a writer
2008-03-31T01:32:18  <zenhase> TheSheep: no
2008-03-31T01:32:20  <TheSheep> xqj37: common scientists often write better than book authors
2008-03-31T01:32:27  <xqj37> true
2008-03-31T01:32:28  <zenhase> TheSheep: i will read it then
2008-03-31T01:32:42  <zenhase> not today, i am quite exhausted
2008-03-31T01:32:45  <TheSheep> xqj37: getting a scientific degree requires you to wirte several "books", and good ones
2008-03-31T01:33:16  <xqj37> and being smart
2008-03-31T01:33:23  <TheSheep> no
2008-03-31T01:33:29  <zenhase> TheSheep: i will think about the points you made
2008-03-31T01:33:32  <TheSheep> unfortunately not
2008-03-31T01:33:50  <CIA-39> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3459:b4b32e3a04af 1.7/MoinMoin/server/server_fastcgi.py: server_fastcgi: add port to the Config, use None as default, so behaviour does not change except when explicitly set
2008-03-31T01:33:53  <zenhase> TheSheep: perhaps it wil get me motivated to start my first blog
2008-03-31T01:34:51  <TheSheep> :)
2008-03-31T01:36:26  <xqj37> TheSheep: well, its not a requirement, but being smart helps
2008-03-31T01:36:27  <zenhase> TheSheep: oh
2008-03-31T01:36:34  <zenhase> TheSheep: i just looked at your blog
2008-03-31T01:36:44  <TheSheep> zenhase: it's not a good blog :)
2008-03-31T01:36:47  <zenhase> you were at the 24c3
2008-03-31T01:37:04  <TheSheep> zenhase: yes, I still need to write about it
2008-03-31T01:37:20  <zenhase> i met thomas and xorAxAx there again
2008-03-31T01:37:36  <zenhase> and also heard that a polish developer was there too
2008-03-31T01:37:42  <CIA-39> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2618:35a3382788e3 1.6/ (MoinMoin/server/server_fastcgi.py docs/CHANGES): server_fastcgi: add port to the Config, use None as default, so behaviour does not change except when explicitly set
2008-03-31T01:37:45  <TheSheep> that's me
2008-03-31T01:37:49  <zenhase> so that was you
2008-03-31T01:37:51  <zenhase> :)
2008-03-31T01:37:56  <zenhase> thought it was karol
2008-03-31T01:38:39  <zenhase> you're going to be a mentor this year then?
2008-03-31T01:38:50  <xqj37> hey thats a bug that bugged me
2008-03-31T01:38:57  <xqj37> and now I witnessed it being fixed
2008-03-31T01:38:57  <xorAxAx> no, it was too far away for karol
2008-03-31T01:39:14  <zenhase> xorAxAx: :/
2008-03-31T01:39:28  <xqj37> a magic moment
2008-03-31T01:39:50  <zenhase> magic?
2008-03-31T01:39:56  <zenhase> did someone sprinkle you with fairy dust?
2008-03-31T01:39:58  <TheSheep> while we chat Thomas works hard
2008-03-31T01:40:09  <zenhase> TheSheep: yeah :o
2008-03-31T01:40:37  <zenhase> i think i will hit the hay soon
2008-03-31T01:40:47  <xqj37> you're right, if it was a microsoft bug it would have been a magic moment
2008-03-31T01:41:24  <zenhase> need a clear head tomorrow
2008-03-31T01:41:40  <xqj37> what day is it anyways?
2008-03-31T01:41:46  <zenhase> monday
2008-03-31T01:42:00  <xqj37> ah ok
2008-03-31T01:42:03  <zenhase> :P
2008-03-31T01:42:22  <xqj37> or: Setting Orange, the 17th day of Discord
2008-03-31T01:42:38  <zenhase> yeah :D
2008-03-31T01:42:52  <zenhase> i never learned the discordian calendar though :/
2008-03-31T01:43:12  <zenhase> shame for someone associated with a discordian society
2008-03-31T01:43:29  <TheSheep> they have a society?
2008-03-31T01:43:52  <zenhase> i meant the CCC
2008-03-31T10:12:51  <dreimark> moin
2008-03-31T10:58:04  <ThomasWaldmann> moin
2008-03-31T11:46:05  <zenhase> moin
2008-03-31T13:02:45  <dreimark> 15
2008-03-31T13:12:39  <zenhase> 42
2008-03-31T13:13:17  <zenhase> moin dreimark btw. :)
2008-03-31T13:14:36  <dreimark> hi zenhase
2008-03-31T13:17:05  <zenhase> i need lunch right now, but i would like to talk with you afterwards :)
2008-03-31T13:18:39  <dreimark> zenhase: I am at work currently and I will be soon at a meeting about scientific visualisations
2008-03-31T13:20:14  <zenhase> ok
2008-03-31T13:20:25  <zenhase> then perhaps later today
2008-03-31T13:20:33  <dreimark> yep
2008-03-31T13:21:25  <zenhase> cu then and good luck or success or so for the meeting :)
2008-03-31T13:22:13  <johill> dreimark: huh ointeresting
2008-03-31T13:23:28  <johill> ThomasWaldmann: I don't think I made that change?
2008-03-31T13:26:28  <johill> xorAxAx: is there any reason that user data is stored as revision metadata rather than item metadata? the latter would be much more consistent and allow us to have item metadata mutable as the only mutable thing throughout
2008-03-31T13:27:48  <xorAxAx> johill: no idea, i think we havent talked a lot about user storage
2008-03-31T13:27:56  <xorAxAx> johill: but i agree to your impression
2008-03-31T13:28:07  <dreimark> ThomasWaldmann: we have had last year this code flow diagram (I remember it was not easy to view in my browser)
2008-03-31T13:28:21  <dennda> Hello :)
2008-03-31T13:28:24  <johill> I was just thinking about that and I noticed that I wrote something about telling the backend which is mutable and which isn't etc
2008-03-31T13:28:33  <dreimark> ThomasWaldmann: is it still somewhere or can you add a new one for 1.7
2008-03-31T13:28:35  <johill> that seems moot if we just make user storage item metadata rather than revision imetadata
2008-03-31T13:28:50  <xorAxAx> johill: why?
2008-03-31T13:29:02  <johill> because item metadata is always assumed mutable
2008-03-31T13:30:24  <johill> ThomasWaldmann: oh that's moin 1.7 master, was that transferred from the todo list? I'll have to take another look but I don't remember making that change anyway
2008-03-31T13:30:41  <xorAxAx> johill: i dont understand
2008-03-31T13:31:06  <johill> xorAxAx: what part?
2008-03-31T13:31:13  <xorAxAx> 13:28:35 < johill> that seems moot if we just make user storage item metadata rather than revision imetadata
2008-03-31T13:31:26  <ThomasWaldmann> johill: ok, thanks
2008-03-31T13:31:30  <johill> xorAxAx: if revision metadata is assumed immutable then the backend can e.g. load it on demand
2008-03-31T13:31:38  <johill> if it isn't, it has to load it atomically
2008-03-31T13:31:41  <ThomasWaldmann> dreimark: no idea about the diagram
2008-03-31T13:32:11  <xorAxAx> johill: yes, it needs to be immutable of course
2008-03-31T13:32:23  <johill> xorAxAx: yeah but user storage currently uses revision metadata
2008-03-31T13:32:36  <johill> which makes it mutable, of course, since users aren't actually revisioned
2008-03-31T13:32:51  <xorAxAx> johill: i know and i agreed that this doesnt make sense
2008-03-31T13:33:07  <johill> good :)
2008-03-31T13:33:13  <xorAxAx> lol, http://photos95.nasza-klasa.pl/dev161/0/166/191/0166191318.jpg
2008-03-31T13:33:19  <johill> what are we talking about then?
2008-03-31T13:33:26  <johill> :)
2008-03-31T13:34:01  <xorAxAx> johill: i was trying to understand that sentence
2008-03-31T13:34:28  <johill> ok
2008-03-31T13:34:37  <johill> well, I thought I had talked to you about that, maybe not
2008-03-31T13:34:55  <johill> I had thought we'd need to have a way to tell the backend that we want mutable revision metadata, just for that user storage case
2008-03-31T13:35:23  <johill> but of course, the easier way is to just use item metadata so we don't need such special handling
2008-03-31T13:36:18  <dennda> it sounds easier to do it with immutable data
2008-03-31T13:36:34  <johill> oh and I rememberd why I thought we needed edit-locks, that was just confusion from the current backend implementation
2008-03-31T13:37:51  <johill> eh, I mean read-locks for edit-lock metadata
2008-03-31T13:38:08  <johill> dennda: you don't want to create a new revision of a user object every time the user decides he wants to change his preferences
2008-03-31T13:38:37  <dennda> did I say that?
2008-03-31T13:38:42  <johill> (btw, my connection is rather flaky, if I drop of blame it on the wireless)
2008-03-31T13:38:54  <johill> dennda: how else can you change user data then if you want immutable data?
2008-03-31T13:38:57  <dennda> oh I did say that
2008-03-31T13:39:15  <dennda> I confused it
2008-03-31T13:42:33  <dreimark> ThomasWaldmann: that image had visualize one request call and I remember the size was quite some thousands pixel
2008-03-31T13:42:46  <johill> anyway, I see no reason to not use item metadata and it makes assumptions (only item metadata is mutable) much saner
2008-03-31T13:43:24  <dennda> johill: What exactly does "item metadata" mean in this context? I understand that with revision metadata, well, if there's a change you create a new revision...
2008-03-31T13:43:36  * dennda takes yet another look at the code
2008-03-31T13:45:02  <johill> well each item has
2008-03-31T13:45:07  <dreimark> ThomasWaldmann: found the discussion 2007-06-09T23:19:39  <dreimark> hi ThomasWaldmann, call graph is a nice
2008-03-31T13:45:13  <johill>  * item metadata (basically a dict)
2008-03-31T13:45:16  <johill>  * revisions
2008-03-31T13:45:21  <johill> each revision has
2008-03-31T13:45:26  <johill>  * revision metadata (a dict too)
2008-03-31T13:45:36  <johill>  * actual revision data (for page data, ...)
2008-03-31T13:46:12  <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3772:e0774f11e1f9 1.7-storage-hwendel/docs/CHANGES.storage: add todo about using item metadata for user storage
2008-03-31T13:47:43  <dreimark> ThomasWaldmann: can you update this one http://moinmo.in/CallGraph  (if time)
2008-03-31T13:47:58  <dreimark> bbl
2008-03-31T13:57:14  <ThomasWaldmann> dreimark: not sure that would be helpful for the students :) "vor lauter wald die baeume nicht sehen" X)
2008-03-31T14:06:04  <TheSheep> cool, that page actually crashed my firefox :)
2008-03-31T14:06:50  <xqj37> thats not that easy :P
2008-03-31T14:08:15  <TheSheep> xqj37: on the contrary, my friend
2008-03-31T14:08:34  <TheSheep> especially on a 64bit system :/
2008-03-31T14:08:40  <xqj37> konqueror and opera crash more often for me
2008-03-31T14:09:14  <TheSheep> hmm... I don't think opera ever crashed on me
2008-03-31T14:09:36  <TheSheep> but I never tried the 64bit version
2008-03-31T14:10:03  <xqj37> I have to admit it only crashed with java stuff
2008-03-31T14:10:29  <xqj37> and it constantly crashed compiz when I still used it
2008-03-31T14:10:38  <xqj37> maybe they've fixed it by now
2008-03-31T14:11:41  <TheSheep> compiz crashes by itself without even touching it :)
2008-03-31T14:12:07  <johill> and gnash likes to eat 3.5GiB of memory (i.e. until it is full save for some rest)
2008-03-31T14:12:54  <TheSheep> truly, 'tis the age of reason
2008-03-31T14:35:06  <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3460:90893e33a312 1.7/MoinMoin/search/queryparser.py: QueryParser: use global configuration with text strings (fixes titlesearch)
2008-03-31T14:35:43  <johill> that bot is getting a bit slow
2008-03-31T14:38:20  <xorAxAx> johill: cf. #cia
2008-03-31T14:38:44  <johill> nah, it's ok, I don't need to know everything :)
2008-03-31T14:38:53  <johill> especially not if it comes with joining another irc channel
2008-03-31T14:39:18  <zenhase> :)
2008-03-31T14:50:16  <johill> uh what happened to our application list? :)
2008-03-31T14:55:29  <dennda> why? is something wrong with it?
2008-03-31T14:56:12  <zenhase> perhaps it exploded?
2008-03-31T14:56:24  <zenhase> yesterday there were 14 iirc
2008-03-31T14:59:49  <johill>  I hadn't looked since the day before that
2008-03-31T15:01:05  <zenhase> :)
2008-03-31T15:04:24  <dennda> How many are there now?
2008-03-31T15:05:19  <johill> 17-1
2008-03-31T15:05:29  <dreimark> cool
2008-03-31T15:06:12  <TheSheep> some are pretty weird
2008-03-31T15:06:28  <johill> some?
2008-03-31T15:06:39  <zenhase> weird?
2008-03-31T15:07:15  <dennda> 17-1? Why -1?
2008-03-31T15:07:20  <TheSheep> pretty?
2008-03-31T15:07:25  <dennda> are?
2008-03-31T15:07:51  <johill> dennda: -1 because one is completely useless and whoever put it in even changed the title to 'ignore me' or something like that
2008-03-31T15:08:02  <zenhase> hmm
2008-03-31T15:08:15  <dennda> ah
2008-03-31T15:08:17  <dennda> so 16
2008-03-31T15:08:26  <zenhase> perhaps he did not know how to withdraw
2008-03-31T15:08:44  <zenhase> 16, quite nice
2008-03-31T15:08:55  <zenhase> and whats weird about some of them? :)
2008-03-31T15:08:59  <xorAxAx> its nice that he didnt withdraw
2008-03-31T15:09:02  <xorAxAx> johill: i hope you realise that
2008-03-31T15:10:44  <johill> xorAxAx: to be honest, I'm confused by it. why put it in in the first place? I suppose not withdrawing it helps with getting more slots or something?
2008-03-31T15:11:01  <johill> (seems easy to exploit if that's the case)
2008-03-31T15:11:26  <zenhase> if this exploit gets overused, it will get noticed quite quick i suppose
2008-03-31T15:12:01  <dennda> I had the very same thought
2008-03-31T15:12:17  <johill> yeah, I suppose it'll be rather visible if you do some bigger-picture stats and compare
2008-03-31T15:12:39  <zenhase> i think i will update my WSGI-proposal
2008-03-31T15:16:27  <johill> I hate the app. somehow the fixed font screws up on my browser and I get a really ugly font
2008-03-31T15:17:17  <zenhase> this is the 4th installment of SoC right?
2008-03-31T15:17:56  <johill> sounds about right, not entirely sure
2008-03-31T15:18:00  <zenhase> how did the app evolve over the years?
2008-03-31T15:18:03  <johill> zenhase: what do you want to change in your application?
2008-03-31T15:18:19  <dreimark> wifi sucks here
2008-03-31T15:18:31  <dreimark> it is the 4th
2008-03-31T15:18:54  <zenhase> johill: thomas wanted that i clarify on how the transition will effect existing users and perhaps some ideas on how to make install easy for those not adept at WSGI
2008-03-31T15:19:50  <johill> zenhase: oh ok
2008-03-31T15:21:06  <zenhase> any objections? ;)
2008-03-31T15:21:09  <johill> zenhase: wsgiref works in older python too, it's just not shipped, the problem I had was finding a way to mount the static file serving part
2008-03-31T15:21:34  <zenhase> johill: what do you mean?
2008-03-31T15:21:47  <johill> well we need to serve static files under /moin-static-170 or whatever it is
2008-03-31T15:21:52  <zenhase> that there is none or that you did not get it mountet properly?
2008-03-31T15:22:28  <johill> there is a static file app but I couldn't find a way to mount it under that dir in wsgiref
2008-03-31T15:22:41  <johill> (well, the static file app is third-party I think, but hey, not much code anyway)
2008-03-31T15:22:43  <zenhase> wsgiref is only a server for wsgi
2008-03-31T15:22:52  <zenhase> mounting in WSGI is done by middlewares
2008-03-31T15:23:07  <zenhase> so you will have to write one yourself
2008-03-31T15:23:14  <johill> yeah, I couldn't find a mount middleware
2008-03-31T15:23:20  <zenhase> or better: let me do it in the process of this proposal ;)
2008-03-31T15:23:33  <johill> although supposedly some stacks contain one
2008-03-31T15:23:41  <johill> yup :)
2008-03-31T15:23:43  <zenhase> it's only a few lines
2008-03-31T15:24:16  <kikka1> Moin
2008-03-31T15:24:20  <zenhase> basically it's looking at script_name and pathinfo, popping of the next part on pathinfo unto scriptname and dispatch
2008-03-31T15:24:25  <johill> yeah, I didn't want to learn more about wsgi details at that point ;)
2008-03-31T15:24:54  <TheSheep> zenhase: maybe some collaboration with the rest guy would be good
2008-03-31T15:26:01  <zenhase> TheSheep: well, the request-stuff is definitely coupled with something like REST
2008-03-31T15:26:28  <zenhase> but only on a low level
2008-03-31T15:26:30  <johill> not too much though since wsgi request code exists already
2008-03-31T15:27:17  <zenhase> johill: what's you general opinion on my proposal then?
2008-03-31T15:27:26  <zenhase> i mean yes. there is a WSGI adapter
2008-03-31T15:27:59  <zenhase> but it's one of many and moin doesn't make use of WSGIs advantages
2008-03-31T15:28:30  <zenhase> it's merely a convenience adapter
2008-03-31T15:30:43  <johill> zenhase: yes, I think it's very useful to do wsgi
2008-03-31T15:30:47  <johill> and only that
2008-03-31T15:30:58  <zenhase> yeah
2008-03-31T15:31:00  <johill> I'm not entirely sure about session/auth code at that level because openid doesn't seem to fit in well
2008-03-31T15:31:21  <zenhase> explain why do you think that?
2008-03-31T15:31:38  <johill> not sure really, more of a feeling I got when writing the openid RP code
2008-03-31T15:31:44  <zenhase> hmm
2008-03-31T15:31:50  <zenhase> i will take a look at this code
2008-03-31T15:31:50  <johill> when I refactored auth I briefly looked at wsgi auth middlewares
2008-03-31T15:32:02  <zenhase> i did too yesterday again
2008-03-31T15:32:08  <zenhase> actually there is only one i think
2008-03-31T15:32:14  <zenhase> and thats authkit
2008-03-31T15:32:16  <johill> and mostly decided that none was really suitable
2008-03-31T15:32:18  <zenhase> which looks a bit strange
2008-03-31T15:32:21  <johill> yeah maybe, it's been a while
2008-03-31T15:32:36  <zenhase> but does not mean it has to stay the only one
2008-03-31T15:32:38  <johill> of course, a decent wsgi auth jmiddleware that could be a spin-off project ;)
2008-03-31T15:32:45  <zenhase> yes
2008-03-31T15:32:49  <zenhase> that is what i am thinking too
2008-03-31T15:32:54  <johill> s/j//, s/that //
2008-03-31T15:33:29  <zenhase> since i am quite interested in python webapps and wsgi is the only way i would build one nowadays, i myself see a strong need for an auth middleware
2008-03-31T15:33:38  <johill> so don't get me wrong, I'm not against that, quite the contrary
2008-03-31T15:34:08  <johill> maybe I just got off it because authkit was so weird :)
2008-03-31T15:34:09  <zenhase> it's like the unloved bastard son of web applications .... lots of frameworks for wsgi, but not for this really important part
2008-03-31T15:34:38  <johill> heh
2008-03-31T15:42:02  <zenhase> i would also like to know which of the 2 proposals i made is seen as being more beneficial?
2008-03-31T15:42:13  <zenhase> i mean, send_page is quite a central point
2008-03-31T15:42:17  <johill> tough call, I can't decide really
2008-03-31T15:42:22  <zenhase> but then
2008-03-31T15:42:23  <johill> but I think wsgi enables more
2008-03-31T15:42:55  <zenhase> there are other proposals like DOM formatter/parser or WSGI for example which can lay a better foundation for the send_page refactoring
2008-03-31T15:43:03  <zenhase> yeah, it enables more
2008-03-31T15:43:18  <zenhase> but then it's not always about features, is it? :)
2008-03-31T15:44:03  <johill> well, not just features, also wrt. testing and internal code separation
2008-03-31T15:44:49  <zenhase> yeah
2008-03-31T15:45:35  <zenhase> well, both proposals are kind of about internals :)
2008-03-31T15:45:54  <xorAxAx> zenhase: well, the reason for the missing auth middlewares is easy
2008-03-31T15:46:02  <xorAxAx> zenhase: it would be a large layering violation
2008-03-31T15:46:15  <xorAxAx> currently, some middleware already do minor such violations
2008-03-31T15:47:37  <zenhase> xorAxAx: why do you think that it is a violation?
2008-03-31T15:48:23  <zenhase> establishing an identity of the user of an application can in my opinion be separate from the functionality of the application itself
2008-03-31T15:48:36  <xorAxAx> zenhase: how would the middleware communicate with the real application without passing data on side channels thereby violating the whole design?
2008-03-31T15:48:45  <zenhase> xorAxAx: sidechannels?
2008-03-31T15:48:54  <zenhase> xorAxAx: it passes its information via environment
2008-03-31T15:48:57  <johill> xorAxAx: ?
2008-03-31T15:49:04  <zenhase> as it's intended in WSGI
2008-03-31T15:49:07  <xorAxAx> yep, the environment is such an equal sidechannel
2008-03-31T15:49:09  <johill> xorAxAx: so ssl authentication is a layering violation by design?
2008-03-31T15:49:14  <xorAxAx> s/equal/evil/
2008-03-31T15:49:17  <zenhase> no it isn't
2008-03-31T15:49:20  <xorAxAx> johill: no
2008-03-31T15:49:26  <xorAxAx> zenhase: let me phrase from the specs ...
2008-03-31T15:49:28  <zenhase> it's intended for such use
2008-03-31T15:49:31  <xorAxAx> s/phrase/quote/
2008-03-31T15:49:36  <johill> xorAxAx: but it's done by the web server and it passes info about the cert in the environment
2008-03-31T15:49:59  <xorAxAx> johill: the problem is that the auth mw needs to query the credentials source
2008-03-31T15:50:06  <johill> xorAxAx: not if it comes with it
2008-03-31T15:50:09  <xorAxAx> while the normal app needs to do that as well
2008-03-31T15:50:23  <zenhase> xorAxAx: it needs to establish an identity
2008-03-31T15:50:28  <zenhase> xorAxAx: that's all it does
2008-03-31T15:51:02  <zenhase> if this identity is sourced from the same source as later on the application draws information, that is not really a layering violation
2008-03-31T15:51:20  <zenhase> it just happens to use a common storage
2008-03-31T15:51:31  <xorAxAx> zenhase: true
2008-03-31T15:51:32  <zenhase> just think about ldap auth
2008-03-31T15:51:44  <zenhase> moin has no ldap server builtin
2008-03-31T15:51:49  <xorAxAx> well, one example is that some mws use environ keys that are static - making it non-composable
2008-03-31T15:52:01  <xorAxAx> that was the largest issue i had until right now with some mws
2008-03-31T15:52:11  <zenhase> if you make ldap auth, auth-mw establishes the identity from the server and passes everything it got on to moin
2008-03-31T15:52:29  <zenhase> what do you mean, static keys?
2008-03-31T15:52:42  <zenhase> like in 'beaker.session'?
2008-03-31T15:52:48  <xorAxAx> yes, for example
2008-03-31T15:52:53  <zenhase> that is not the problem
2008-03-31T15:52:57  <zenhase> really
2008-03-31T15:52:58  <xorAxAx> it is
2008-03-31T15:53:15  <xorAxAx> if you need two session apps, they will conflict
2008-03-31T15:53:26  <xorAxAx> its a bug in beaker if it doesnt allow you to configure the key
2008-03-31T15:53:26  <zenhase> well
2008-03-31T15:53:35  <zenhase> it does not?
2008-03-31T15:53:39  <xorAxAx> no idea
2008-03-31T15:53:44  <zenhase> iirc it is
2008-03-31T15:53:44  <xorAxAx> it was your example :)
2008-03-31T15:54:08  <zenhase> i thought your problem is, that every app has it's own ideas about standard keys
2008-03-31T15:54:13  <zenhase> take paste for example
2008-03-31T15:54:18  <zenhase> 'paste.config'
2008-03-31T15:54:22  <xorAxAx> so, i dont have any general advice that would point out design issues ... i guess designing a sane authkit thingie might indeed workout
2008-03-31T15:54:28  <zenhase> i dunno if this is configurable
2008-03-31T15:54:37  <zenhase> but actually it doesn't matter
2008-03-31T15:54:42  <zenhase> not that hard
2008-03-31T15:55:06  <zenhase> because it is not about 'automatic full composability without any negative sideeffects'
2008-03-31T15:55:20  <zenhase> it's about an easy framework for developers
2008-03-31T15:55:31  <xorAxAx> gah, its not a framework :)
2008-03-31T15:55:32  <zenhase> which makes it easy to separate concerns
2008-03-31T15:55:40  <xorAxAx> its a spec that implements the decorator pattern
2008-03-31T15:55:47  <xorAxAx> s/implements/uses/
2008-03-31T15:56:03  <xorAxAx> zenhase: yes, thats the primary advantage
2008-03-31T15:56:06  <zenhase> perhaps you misunderstand me about what i mean with framework
2008-03-31T15:56:23  <zenhase> i mean framework like: a set of ideas and constraints to work in
2008-03-31T15:56:30  <xorAxAx> yes, you mean it literally
2008-03-31T15:56:34  <zenhase> yeah :)
2008-03-31T15:56:49  <zenhase> not in the software engineering pattern kind of meaning ;)
2008-03-31T15:57:27  <zenhase> so again: what do you think is the problem with auth as middleware?
2008-03-31T15:57:37  <zenhase> i still don't really see the violation of layers
2008-03-31T15:58:40  <xorAxAx> there is none, as i said :)
2008-03-31T15:58:47  <zenhase> uh ok :)
2008-03-31T15:59:10  <zenhase> then i am sorry for all this fuzz :)
2008-03-31T15:59:34  <xorAxAx> no, it was me who started the collaborative thinking about this aspect
2008-03-31T16:00:57  <zenhase> :)
2008-03-31T16:29:07  <zenhase> i have an english tongue issue, perhaps someone can help me
2008-03-31T16:29:26  <zenhase> how do i refer to a female professor
2008-03-31T16:29:33  <zenhase> ?
2008-03-31T16:29:47  <xorAxAx> in german?
2008-03-31T16:29:54  <zenhase> no, in english
2008-03-31T16:30:01  <xorAxAx> like you just did
2008-03-31T16:30:23  <zenhase> no, i mean like talking to or about a person
2008-03-31T16:30:40  <CIA-39> Johannes Berg <johannes AT sipsolutions DOT net> default * 3773:03da76ab88a9 1.7-storage-hwendel/LOCKFREE: remove some text about mutable stuff, better solution to just use item metadata
2008-03-31T16:30:41  <zenhase> "Sehr geehrte Frau Prof. Dr. Schultz"
2008-03-31T16:31:07  <zenhase> bzw. "Wie ich bereits Frau Prof. Dr. Schultz mitteilte ..."
2008-03-31T16:31:20  <xorAxAx> no idea
2008-03-31T16:36:13  <zenhase> hmm
2008-03-31T16:36:17  <zenhase> quite a problem
2008-03-31T17:31:00  <xorAxAx> "The new deadline for student applications is Monday, April 7, 2008.
2008-03-31T17:31:01  <xorAxAx> "
2008-03-31T17:32:16  <xorAxAx> "As of about 5 minutes ago, we had 4200 applications, as compared with
2008-03-31T17:32:16  <xorAxAx> 6300 last year over at 2.5 week period. "
2008-03-31T17:34:33  <xorAxAx> 17:34:10 <@lh> code.google.com/soc/2008/ and the timeline will be updated in a few hours, people are working on it
2008-03-31T17:35:01  <xorAxAx> i always have to grin when reading "will be updated in a few hours" regarding trivial website changes :)
2008-03-31T17:51:15  <xorAxAx> xqj37: so, you dropped out of soc, but maybe you want to participate in this project nevertheless? :)
2008-03-31T18:28:58  <dennda> xorAxAx: is he the one who "withdrew" his application?
2008-03-31T18:29:13  <xorAxAx> dennda: yes
2008-03-31T18:50:47  <gizmo_> how many applications are sent for Moin projects?
2008-03-31T18:50:57  <gizmo_> btw hi all
2008-03-31T19:02:29  <dreimark> hi gizmo
2008-03-31T19:02:41  <dreimark> 17
2008-03-31T19:07:50  <dreimark> 18:42 < anirudh> !extension
2008-03-31T19:07:50  <dreimark> 18:42 <+socinfo> "extension" is student application deadline has been extended to April 7th
2008-03-31T19:08:44  <xorAxAx> dreimark: old :)
2008-03-31T19:08:57  <gizmo> dreimark: thanks
2008-03-31T19:11:29  <dreimark> xorAxAx: I should have scrolled more lines up
2008-03-31T19:12:00  <xorAxAx> reading everything that happens in this channel makes a lot of sense, generally, even if you just skim it
2008-03-31T19:13:43  <dreimark> yeah, i read the line "updated in a few hours," and thought  the same as you did
2008-03-31T19:14:26  <xorAxAx> hehe
2008-03-31T19:14:56  <xorAxAx> dreimark: i recommend trackbar.pl
2008-03-31T19:14:58  <xorAxAx> for irssi
2008-03-31T19:15:05  <xorAxAx> that eases reading channels
2008-03-31T19:16:22  <dreimark> cool
2008-03-31T19:43:54  <dennda> xorAxAx: what does it do? show a line where you last read like xchat?
2008-03-31T19:44:24  <xorAxAx> yes
2008-03-31T19:45:04  <zenhase> re
2008-03-31T19:45:34  <zenhase> i just met up with xqj37
2008-03-31T19:46:29  <zenhase> we had a several-hour-discussion about programming, artificial intelligence and the fabric of human society :)
2008-03-31T19:48:49  <xorAxAx> do you believe in the singularity, zenhase?
2008-03-31T19:51:17  <zenhase> hahhaha
2008-03-31T19:51:46  <xorAxAx> does that mean yes or no?
2008-03-31T19:52:14  <zenhase> i am a bit skeptic ... i told him that if at all AI reaches this point, it would be perhaps very far in the future like 500 years
2008-03-31T19:52:38  <zenhase> it's too early to make assumptions imo
2008-03-31T19:52:57  <zenhase> because we don't know anything about ourselves right now ;)
2008-03-31T19:53:05  <zenhase> not on this scale at least
2008-03-31T19:54:35  <xorAxAx> IMHO there is no indication (yet) that this might ever happen
2008-03-31T19:55:01  <xorAxAx> esp. regarding the few successes of the AI community
2008-03-31T19:55:05  <zenhase> yeah
2008-03-31T19:55:19  <zenhase> as said ... too early to make assumptions
2008-03-31T19:57:33  <zenhase> did you talk with him too about singularity?
2008-03-31T19:57:56  <zenhase> or did that just come up, when i said 'AI'?
2008-03-31T19:58:17  <xorAxAx> no, i havent talked to him yet
2008-03-31T19:58:40  <zenhase> oh
2008-03-31T19:58:42  <xorAxAx> yes, thats usually my first connotation these days
2008-03-31T19:59:04  <zenhase> i also just realized, that the 3 topics i mentioned can be all related to singularity :)
2008-03-31T19:59:24  <zenhase> still they were separate discussions
2008-03-31T19:59:59  <zenhase> and some hours ago i actually did not know about this singularity summit
2008-03-31T20:00:03  <xorAxAx> what did the "fabric of human society" discussion touch?
2008-03-31T20:00:10  <xorAxAx> singularity summit?
2008-03-31T20:00:17  <zenhase> the conference
2008-03-31T20:00:29  <zenhase> regarding this idea
2008-03-31T20:01:19  <zenhase> well, that discussion was more about role of the state in supporting social wellbeing
2008-03-31T20:01:41  <xorAxAx> peter norvig, heh
2008-03-31T20:01:42  <zenhase> i won't say anything about the details here :)
2008-03-31T20:01:55  <zenhase> hmm?
2008-03-31T20:02:22  <xorAxAx> was one of the keynote speakers last year. it seems like there is no such event in 2008 (yet planned)
2008-03-31T20:02:41  <zenhase> we'll see
2008-03-31T20:09:43  <ThomasWaldmann> re
2008-03-31T20:13:07  <kikka1> rere
2008-03-31T20:45:39  <waldi> hrm, ElementTree lacks comments
2008-03-31T20:49:13  <zenhase> documentation in general, iirc
2008-03-31T20:51:08  <waldi> no, comment nodes
2008-03-31T20:51:16  <waldi> the comments are removed during parsing
2008-03-31T20:51:42  <waldi> and text is included into the elements themself instead of textnodes
2008-03-31T20:54:15  <zenhase> well
2008-03-31T20:54:31  <zenhase> thought that is what elementrees implementation is about
2008-03-31T20:54:33  <grzywacz> moin
2008-03-31T20:54:39  <waldi> lightweight
2008-03-31T20:54:43  <zenhase> yep
2008-03-31T20:54:43  <waldi> sure
2008-03-31T20:54:56  <zenhase> what do you need comments for?
2008-03-31T20:55:29  <waldi> if I parse an file and write it again, the contents should be equivalent
2008-03-31T20:56:12  <zenhase> hmm
2008-03-31T20:56:27  <zenhase> what does the xml-spec say about this?
2008-03-31T20:56:49  <zenhase> is there anything about preserving the exact byte-equality of documents?
2008-03-31T20:56:54  <waldi> no
2008-03-31T20:57:07  <waldi> i said equivalent, not equal
2008-03-31T20:57:30  <zenhase> well, comments have no meaning in XML
2008-03-31T20:57:49  <waldi> only to the file, yes
2008-03-31T20:57:50  <starshine> <!-- if you say so -->
2008-03-31T20:57:58  <zenhase> starshine: :D
2008-03-31T20:57:58  <waldi> but it looks that i need some modifications to elementtree anyway
2008-03-31T20:58:08  <zenhase> hmm
2008-03-31T20:58:40  <zenhase> what do you need?
2008-03-31T21:00:10  <waldi> i think the current implementation with leaf text makes some tree modification really hard. the text belongs to the element one level up
2008-03-31T21:00:51  <waldi> it is the easy implementation, sure
2008-03-31T21:01:29  <xorAxAx> waldi: did you answer my compat question?
2008-03-31T21:02:10  <waldi> which one?
2008-03-31T21:04:15  <xorAxAx> """"Oldstyle parser should be also able to use it." sounds like the parsers need to be modified. Will you make sure that the system will still work with legacy plugins out of the box? Do you think that the DWIM transformation will be a complex task?"""
2008-03-31T21:04:55  <waldi> hmm, DWIM should be NP hard
2008-03-31T21:05:04  <xorAxAx> ?
2008-03-31T21:05:34  <waldi> DWIM have to compute ever possible output and then decide which one to use *hide*
2008-03-31T21:06:04  <xorAxAx> by sending it to random email adresses and seeing which user clicks which link in the spam?
2008-03-31T21:06:34  <waldi> well, i would prefer the implementation using IMPS ;)
2008-03-31T21:06:45  <xorAxAx> i would certainly prefer that you elaborate a bit on these issues
2008-03-31T21:06:54  <xorAxAx> in comments or an edited application
2008-03-31T21:07:47  <waldi> well, i thought i answered that before
2008-03-31T21:08:45  <xorAxAx> when and where?
2008-03-31T21:09:25  <waldi> comments to the application, 2 days ago or so
2008-03-31T21:10:28  <xorAxAx> no, see the comment that i just quoted, you werent very clear and i asked you to clarify
2008-03-31T22:42:08  <zenhase> what is DWIM?
2008-03-31T22:52:52  <xorAxAx> do what i mean
2008-03-31T23:04:56  <dreimark> hehe anyway good night