00:00:34 <nikq> Mapnik Trac: Ticket #41 (Provide return value policy in doc strings) updated | http://trac.mapnik.org/ticket/41#comment:3
00:01:15 <nikq> Mapnik Trac: Ticket #246 (python: add function to test for available plugins) updated | http://trac.mapnik.org/ticket/246#comment:3
01:03:11 <ser> hello
01:05:35 *** weizhuo has quit ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
01:12:22 *** springmeyer (n=dane@c-24-19-50-92.hsd1.wa.comcast.net) has joined #mapnik
01:37:25 <ser> on http://wiki.openstreetmap.org/wiki/Mapnik they write osm will take 7GB in postgres
01:38:13 <ser> currently i have 60GB and i am still importing data from few past days
01:38:47 <ser> does it mean there is a chance to vacuum it after an import?
04:18:11 *** aub_ (n=aubrey@216.156.96.6.ptr.us.xo.net) has joined #mapnik
05:24:24 *** aura (n=wolf@cpe-67-49-133-78.hawaii.res.rr.com) has joined #mapnik
05:31:45 *** w0lfie_ has quit (Read error: 110 (Connection timed out))
07:04:56 *** xcacou (n=aga@AToulouse-157-1-87-235.w86-213.abo.wanadoo.fr) has joined #mapnik
09:36:34 <nikq> Mapnik Trac: image_util_cpp_jpeg_quality.diff attached to Ticket #198 | http://trac.mapnik.org/attachment/ticket/198/image_util_cpp_jpeg_quality.diff
09:43:28 <nikq> Mapnik Trac: Ticket #198 (Allow configuration of JPEG quality) updated | http://trac.mapnik.org/ticket/198#comment:2
09:43:49 <nikq> Mapnik Trac: image_util_hpp_jpeg_quality.diff attached to Ticket #198 | http://trac.mapnik.org/attachment/ticket/198/image_util_hpp_jpeg_quality.diff
09:54:23 *** _Dennis_ (n=dennis@irafs1.ira.uni-karlsruhe.de) has joined #mapnik
11:43:33 *** xcacou_ (n=aga@AToulouse-157-1-10-155.w86-201.abo.wanadoo.fr) has joined #mapnik
11:45:38 *** xcacou has quit (Read error: 110 (Connection timed out))
13:04:34 *** aura is now known as w0lfie_
13:04:34 *** ninja_ (n=ninja@cm69.psi132.maxonline.com.sg) has joined #mapnik
14:36:40 *** aub (n=aubrey@216.156.96.6.ptr.us.xo.net) has joined #mapnik
14:37:04 *** aub_ has quit (Read error: 104 (Connection reset by peer))
14:45:11 *** sanjiv (n=chatzill@59.180.129.227) has joined #mapnik
15:38:35 *** xcacou_ has quit (Remote closed the connection)
15:41:35 *** aub has quit (Read error: 104 (Connection reset by peer))
15:41:53 *** aub (n=aubrey@216.156.96.6.ptr.us.xo.net) has joined #mapnik
15:48:39 <nikq> Mapnik Trac: Ticket #198 (Allow configuration of JPEG quality) updated | http://trac.mapnik.org/ticket/198#comment:3
16:01:20 *** aub has quit (Connection timed out)
16:21:48 *** aub (n=aubrey@cpe-72-227-134-148.nyc.res.rr.com) has joined #mapnik
17:06:20 *** sanjiv has quit (Connection timed out)
17:08:44 *** sanjiv (n=chatzill@59.180.158.227) has joined #mapnik
18:36:56 <springmeyer> _Dennis_: i re built with your patch - works nicely
19:07:26 <nikq> Mapnik Trac: Ticket #290 (Demtools feature integration with Mapnik) created | http://trac.mapnik.org/ticket/290
19:20:10 *** synax (n=synax@24.222.57.182) has joined #mapnik
19:20:22 <synax> # Further references: See Artem's email on using the PostGIS from Python
19:20:23 <synax> # Example code at the Mapnik-utils project: http://code.google.com/p/mapnik-utils/source/browse/trunk/tutorials/postgis/postgis_geometry.py
19:20:33 <synax> both of those links from http://trac.mapnik.org/wiki/PostGIS are broken
19:20:43 <synax> :(
19:22:28 *** Spyou (n=spyou@santyrix.in.spyou.org) has joined #mapnik
19:30:07 <jlivni> d/join #geodjango
19:30:36 <springmeyer> synax: http://mapnik-utils.googlecode.com/svn/example_code/postgis/postgis_geometry.py
19:30:47 <synax> yeah, just found that myself
19:30:49 <springmeyer> could you update the wiki?
19:30:58 <synax> can you find the email?
19:31:06 <springmeyer> I also had a updated version sitting around, just committed
19:32:10 <springmeyer> email link works for me
19:32:27 <springmeyer> maybe strip the 's' from https ?
19:33:10 <nikq> Mapnik Trac: PostGIS edited | http://trac.mapnik.org/wiki/PostGIS?version=12
19:37:24 <CIA-6> mapnik-utils: dane.springmeyer * r624 /example_code/postgis/postgis_geometry.py: contruct params dict to send to postgis
19:44:31 <CIA-6> mapnik-utils: dane.springmeyer * r625 /example_code/memory_datasource/style.xml: fix errant xml entry
19:44:31 <CIA-6> mapnik-utils: dane.springmeyer * r626 /example_code/wms/localogc.py: touchup imports
19:44:39 *** synax has quit ()
20:04:48 <CIA-6> mapnik-utils: dane.springmeyer * r627 / (7 files in 6 dirs): random old uncommitted stuff
20:25:56 <ser> on http://wiki.openstreetmap.org/wiki/Mapnik they write osm will take 7GB in postgres
20:25:59 <ser> currently i have 60GB and i am still importing data from few past days
20:26:02 <ser> does it mean there is a chance to vacuum it after an import?
20:35:39 <jburgess_> I don't think so. The extra space is taken by lots of indexes, the wiki page is out of date
20:36:03 <jburgess_> If you don't want to do diff-based updates then you can drop several of the tables and indexes
20:42:24 <ser> i definitely want to do "minutely mapnik"
20:42:31 <ser> jburgess_: thanks :-)
20:42:46 <jburgess_> ok, then I'm afraid you need the extra tables and indexes :(
20:42:47 <ser> i will update the wiki soon
20:43:05 <ser> i;ve already updated it partially
20:44:06 <jburgess_> Thanks. The docs often lag behind the technical developments
20:44:20 <ser> sure :)
20:47:48 <springmeyer> ser: I wonder if you could dig up/produce a little graph of the size of the db over time
20:48:19 <springmeyer> no doubt it is growing fast
20:48:41 *** Spyou has parted #mapnik ()
20:49:49 <springmeyer> jburgess_: I'd love to your feedback/additions to the roadmap: http://trac.mapnik.org/roadmap
20:49:57 <springmeyer> if you have any thoughts
20:51:46 <jburgess_> I can't think of any immediate things which need fixing for 0.6.1 stuff. I guess we need to see what comes along as people start adopting the new 0.6 features
20:52:11 <springmeyer> okay, sounds good
20:52:45 <springmeyer> I guess I'm curious what you might see as goals for 0.7.0 vs 'some later date' ie 1.0
20:53:17 <springmeyer> e.g. I'm thinking handling rtl text, indic fonts might be cool to think about for 0.7.0 in relation to OSM
20:53:18 <jburgess_> Longer term there are a few bits and pieces I might look at like the pistgis-3d/4d support #44 (I have a possible use for measure type geometries for doing things with OSM address interpolation)
20:53:19 <nikq> Ticket #44: add support for 3D/4D postgis geometries, http://trac.mapnik.org/ticket/44
20:53:42 <jburgess_> or adding SVG symbol support, perhaps using some external library, or the cairo svg support
20:54:40 <springmeyer> great, I think we still need to ticket consolidating thoughts on svg support
20:54:58 <jburgess_> improving the RTL support would be useful. There are occasional complaints that things don't work correctly for some Indian, Ethiopian or other far eastern characters
20:56:13 <jburgess_> While I can see it is important, it is tricky for us UK/USA people to know what the problems are and when we have them fixed.
20:56:51 <springmeyer> ya, I'm clueless on #112 :)
20:56:51 <nikq> Ticket #112: Support for Indic OpenType Font rendering with ICU 'Shaping', http://trac.mapnik.org/ticket/112
20:57:11 <springmeyer> but I recognize the value so I filed it :)
20:58:12 <springmeyer> how does that look: http://trac.mapnik.org/milestone/0.7.0 ?
20:58:55 <jburgess_> Another feature which should be fairly easy would be geotiff output
20:59:36 <springmeyer> what would the use be?
21:00:27 <jburgess_> don't know exactly, but some apps can understand geotiff but not the normal OSM/Google style tiles that we make at the moment
21:01:40 <jburgess_> since we know the precise geo co-ords of our output we should be able to render to a tiff file and add the relevant metadata quite easily
21:02:14 <jburgess_> I think other people have done something similar using "world files", but I've not looked at them
21:03:26 <springmeyer> ah ha, surely
21:03:28 <jburgess_> http://www.ductisoft.com/en/gri.html
21:03:37 <springmeyer> but geotiff would be much cleaner
21:03:53 <springmeyer> although worldfiles are perfect for georeferencing the jpeg and png output
21:04:15 <jburgess_> maybe we can add an option to generate a worldfile when saving an image
21:04:30 <springmeyer> yes
21:04:36 <springmeyer> I've actually bolted on world_file generation to the map object
21:04:52 <springmeyer> but it would be a better idea to have it associated with the render_to_file, ya
21:05:07 <springmeyer> with this...
21:05:08 <springmeyer> http://code.google.com/p/mapnik-utils/source/browse/branches/nik2img/mapnik_utils/metaclass_injectors.py#166
21:05:11 <springmeyer> you can do
21:05:21 <springmeyer> >>> m.to_wld()
21:05:28 * springmeyer gets a call...
21:05:57 <jburgess_> on a different topic, we could invite more language bindings. Java, Ruby, ...
21:07:19 <ser> springmeyer: how do you advice me to count MB in particular tables?
21:07:35 <ser> springmeyer: of course i can deploy such a script without any problem
21:08:26 <ser> springmeyer: as I plan to do "minutely mapnik" for at least few another years
21:09:10 <ser> i can do such a graph with probes each minute
21:11:05 <jburgess_> I don't think there is a direct way to read the table size, you need to query some pg_* tables, e.g.
21:11:08 <jburgess_> SELECT relname, reltuples, relpages FROM pg_class WHERE relname like 'planet_osm_%' ORDER BY relpages DESC;
21:13:20 <jburgess_> there might be some postgres contrib scripts to make this easier, or maybe an existing plugin with munin/cactus/pgadmin etc.
21:22:14 *** sanjiv has quit ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032711]")
21:37:22 <nikq> Mapnik Trac: MinScaleDenominator edited | http://trac.mapnik.org/wiki/MinScaleDenominator?version=3
21:39:41 <springmeyer> hey jburgess_: ya would be good to have geotiff as an option for the openstreetmap/export
21:40:10 <springmeyer> jburgess_: do you know anyone with experience writing ruby ->> C++ bindings?
21:40:50 <jburgess_> yes, geotiff would a another useful output format. As would perhpas world files, maybe wrapped with the image in a .zip or .tar.gz
21:40:56 <springmeyer> I think andrew turner mentioned he is going to be working on it
21:41:08 <springmeyer> yes
21:42:21 <jburgess_> I don't know about ruby. It really needs someone that needs this functionality. That way you get test cases and some level of ongoing support / notification if it breaks.
21:42:47 <springmeyer> yes.
21:43:43 <jburgess_> personally I have started experimenting with the python bindings for the mod_tile code. I'm unlikely to try ruby as well.
21:46:34 <nikq> Mapnik Trac: Ticket #281 (Get together OGCServer 'wms.py' examples for cgi/fastcgi/wsgi/mod_python  ...) updated | http://trac.mapnik.org/ticket/281#comment:2
21:48:16 <nikq> Mapnik Trac: Ticket #97 (Add an option to not allow mapnik cut labels) updated | http://trac.mapnik.org/ticket/97#comment:4
21:48:32 <nikq> Mapnik Trac: Ticket #97 (Add an option to not allow mapnik cut labels) closed | http://trac.mapnik.org/ticket/97#comment:5
21:49:17 <nikq> Mapnik Trac: Ticket #98 ([SOLUTION] Text labels cut on render edge) closed | http://trac.mapnik.org/ticket/98#comment:5
21:50:20 <ser> jburgess_: i've jst installed phppgadmin and it does not count size tables, but i will find it out the way, thanks for suggestions
21:50:50 <jburgess_> ser: I found a longer query which might give you "real" number in MB/GB etc:
21:51:04 <jburgess_> SELECT relname, pg_size_pretty(relpages::bigint * 8 *1024) AS size, CASE WHEN relkind = 't' THEN (SELECT pgd.relname FROM pg_class pgd WHERE pgd.relfilenode::text = SUBSTRING(pg.relname FROM 10)) ELSE (SELECT pgc.relname FROM pg_class pgc WHERE pg.reltoastrelid = pgc.relfilenode) END AS refrelname, relfilenode, relkind, reltuples::bigint, relpages FROM pg_class pg ORDER BY relpages DESC;
21:52:43 <jburgess_> that should give you something like: http://rafb.net/p/UkY1ei61.html
21:52:51 <ser> jburgess_: wow, thanks
22:08:57 <ser> i will try to deploy a script which run this SELECT after each commit to the database
22:09:47 <ser> but i am still importing last two days of osm, tomorrow i probably will have an actual database
22:10:10 <ser> each day is importing about 300 minutes
22:25:35 <nikq> Mapnik Trac: UsingScons edited | http://trac.mapnik.org/wiki/UsingScons?version=15
22:28:18 *** ninja_ has quit ()
22:39:10 <ser> jburgess_: why do you work on mod_tile if there is pretty tilecache?
22:40:23 <ser> jburgess_: especially when apache is not very efficient server as we know for such a purposes
22:43:56 <nikq> Mapnik Trac: Ticket #290 (Demtools feature integration with Mapnik) updated | http://trac.mapnik.org/ticket/290#comment:1
22:45:39 <springmeyer> ser: what you just said does not make any sense to me :)
22:51:57 <ser> it was a question, not a sentence
22:54:09 <ser> apache is process-based server, nginx or lighttpd are asynchronous servers. imho for serving partially static content is better to use asynchronous one.
22:55:19 <ser> as i can understand, mod_tile is generating a tile and serving it from cache later, and it is static, right?
22:56:21 <ser> sorry for misundarstanding, i do not want to attack any ideas or solutions, i'd just like to know the reason why do you write mod_tile
22:57:39 <ser> i am learning the topic and i am still far far far away from your knowledge
22:59:55 *** orls (n=chatzill@94-194-100-108.zone8.bethere.co.uk) has joined #mapnik
23:00:01 <springmeyer> mod_tile has a sophisticated separation of request handling and threaded render daemon
23:00:34 <springmeyer> ser to better understand maybe see jon's email here:
23:00:35 <springmeyer> http://www.nabble.com/Re%3A-Improvement-of-tile-rendering-p22634233.html
23:01:52 <ser> thanks, i am reading
23:02:05 <orls> Hi folks, congrats on 0.6 release
23:02:27 <springmeyer> orls: hey there, and thanks
23:03:23 <orls> Sorry to ask the question you've probably been asked a thousand times tonight; is there something wrong with the win32 binary release?
23:04:09 <springmeyer> no worries, have not had any feedback yet :)
23:04:16 <springmeyer> what are you experiencing?
23:05:14 <orls> The download link seems not to be functioning at all. Getting a 'document contains no data' message.
23:05:37 * springmeyer goes to check...
23:07:01 <springmeyer> ya berlios is down
23:07:07 <springmeyer> that's lame
23:07:16 <springmeyer> orls: alternate spot to download:
23:07:17 <springmeyer> http://media.mapnik.org/mapnik-0.6.0-win-py25.zip
23:07:33 <orls> ah, you're a saint :)
23:07:42 <springmeyer> not one bit
23:07:44 <ser> springmeyer: OK, so the answer is jon did not analyze other possibilities then apache module. imho fcgi for tile generation and fcgi redirect to static content directly served by an asynchronous server should be faster.
23:08:15 <ser> springmeyer: but of course it depends on a load of the server.
23:08:31 <springmeyer> and # of processors
23:08:38 <ser> yes, indeed
23:08:58 <w0lfie_> dane i'm thinking of trying to implement that thing we discussed
23:08:59 <w0lfie_> VPF
23:09:16 <springmeyer> interesting one to check out threads vs. processes:
23:09:17 <springmeyer> http://us.pycon.org/2009/conference/schedule/event/31/
23:09:24 <ser> but as jon said, nobody prohibits us from improving his software :-)
23:09:32 <springmeyer> w0lfie_: fun stuff
23:09:43 <w0lfie_> if i can get it working it would be :)
23:09:59 <springmeyer> ser: of course
23:10:22 <w0lfie_> im mostly just wondering what performance would be like
23:10:25 <springmeyer> which is makes sense to me :) just not saying that 'tilecache is pretty'
23:11:04 <ser> springmeyer: ok, it was not a perfectly designed question, you are right
23:11:19 <springmeyer> :) no worries, just calling you on it
23:11:47 <springmeyer> ser: in general too I'm no expert in threaded web services but I'm of the opinion...
23:12:15 <springmeyer> that there is a lot of talk and good ideas about how to optimize serving tiles
23:12:23 <springmeyer> (which is cool)
23:12:50 <springmeyer> but in comparision little effort at doing speed profiling in mapnik
23:13:15 <ser> fistly i need to have one before i will start to optimize anything :) and i am waiting for my 64G RAM as well
23:13:16 <w0lfie_> so you're saying it makes more sense to look at speeding up dynamic grabs
23:13:18 <w0lfie_> rather than tiling?
23:13:40 <springmeyer> and sense a lot of cpu is spent in rendering, why focus so much on the tileserving?
23:14:04 <springmeyer> ** not directed at anything/anyone **
23:14:18 <w0lfie_> well if you tile you only render once
23:14:24 <w0lfie_> so its not important how long it takes
23:14:28 <w0lfie_> im not following you
23:14:31 <ser> imho each static content should be servered directly by www server, without engaging any fcg/modules
23:14:39 <springmeyer> just hopeing that folks with more C++ experience than I use their skills at performance inside mapnik too
23:15:08 <springmeyer> w0lfie_: sure not important at all if your data never changes
23:15:20 <w0lfie_> most people's base data doesn't change much
23:15:25 <w0lfie_> OSM is pretty atypical
23:15:45 <springmeyer> most all people that ask me about performance are using osm
23:16:17 <w0lfie_> even then, OSM doesnt change fast enough to retile the whole world regularly
23:16:22 <w0lfie_> so what they're doing makes no sense imo
23:16:37 <springmeyer> heh
23:17:13 <w0lfie_> anyway its easy to solve
23:17:15 <ser> w0lfie_: idealy we need to regenerate only changed tiles
23:17:18 <w0lfie_> just track your additions to OSM
23:17:21 <w0lfie_> and redo changed tiles
23:17:22 <w0lfie_> yeah
23:17:27 <w0lfie_> i dont see what the big deal is
23:17:38 <springmeyer> whatever, I'm just being honest that I have an opinion that I'd like to see the ratio of hand waving about things being 'easy'
23:17:54 <springmeyer> change in relation to folks contributing to mapnik core
23:18:04 <ser> haha
23:18:28 <ser> you are an optimist ;
23:18:40 <w0lfie_> im not saying you shouldnt want to improve rendering time
23:18:45 <w0lfie_> im saying i dont think its a big problem
23:18:55 <springmeyer> and also that mod_tile hands a ton of stuff and it seems very few people understand it
23:19:19 * springmeyer ends rant :)
23:19:34 <ser> springmeyer: mostly because i have not seen apache for last 4 years
23:20:28 <ser> and i have to start learning it again :-)
23:20:30 <springmeyer> no worries ser - you are doing research - I don't mean to say anything personally to you :)
23:20:32 <w0lfie_> but yeah, i dont understand why anyone would like..
23:20:36 <w0lfie_> download osm weekly and retile it
23:20:40 <w0lfie_> thats just asinine imo
23:21:09 * springmeyer heads back to django code.
23:21:12 <ser> w0lfie_: i would like to download it each menute
23:21:25 <ser> and retile only changed areas
23:21:32 <ser> but how to do it?
23:21:32 <w0lfie_> why
23:21:36 <w0lfie_> its not changing that fast
23:21:44 <ser> yes, it is changing each minute
23:22:02 <w0lfie_> not substantively
23:22:06 <ser> http://planet.openstreetmap.org/minute/
23:22:08 <ser> :-)
23:22:44 <w0lfie_> i wouldnt call a 10kb change substantive
23:22:58 <w0lfie_> roads just dont change that much
23:23:03 <ser> it is compressed 10kb
23:23:07 <w0lfie_> yes i know
23:23:09 <w0lfie_> so what?
23:23:41 <ser> 144KB each minute
23:23:59 <ser> 60MB each hour
23:24:17 <w0lfie_> that doesnt tell me whether the changes are substantive or not
23:24:40 <ser> if we would like to compete with google we need to use our advantages
23:25:44 <w0lfie_> heh i  could never use OSM anyway
23:25:56 <ser> you are wrong ;-)
23:26:07 <w0lfie_> um
23:26:14 <w0lfie_> and you are arrogant to think you know anything about my organization
23:26:18 <w0lfie_> that i could ever use OSM
23:26:44 <ser> hah i do now nothing about your organization :)
23:27:08 <ser> but i hope you are not working day and night ;
23:27:20 <w0lfie_> its 1.30 and im sitting at home, so nah
23:27:43 <ser> ;)
23:28:15 <ser> so probably you are living in continental europe ;
23:28:15 <w0lfie_> street data wouldnt be the sort of thing to keep me up at night anyway
23:28:23 <w0lfie_> 1330, not 0130
23:28:45 <w0lfie_> you could just /whois me
23:29:30 <ser> :)
23:29:36 <w0lfie_> anyway the problem with OSM is authoritativeness
23:30:21 <ser> yes and not, our research shows sometimes osm is more authorative
23:30:47 <w0lfie_> what's relevant is you have no idea
23:30:48 <ser> it is standard wikipedia vs. britannica problem
23:30:54 <w0lfie_> yep
23:31:21 <w0lfie_> and you're also assuming beneficence
23:31:33 <w0lfie_> if someone malicious wants to change OSM data, there's nothing to stop them
23:32:10 <ser> sure, but generally there is no need for army-trustful-level maps
23:32:29 <w0lfie_> nah, i work for the marines
23:32:32 <w0lfie_> not the army ;)
23:32:44 <ser> haha, OK - now i can see your point
23:32:51 <w0lfie_> but yeah i don't hate OSM
23:32:55 <ser> and you are perfectly right
23:33:02 <ser> from your point of view
23:33:09 <w0lfie_> but i can't rely on it
23:33:10 <w0lfie_> and..
23:33:12 <w0lfie_> for certain people
23:33:17 <w0lfie_> it's better to have nothing than something unreliable
23:33:33 <ser> but i am a young GIS researcher, phd student
23:33:38 <w0lfie_> which i know seems like a weird way of thinking to some people
23:33:46 <w0lfie_> b ut it depends on your requirements
23:33:58 <ser> and the essential part of my research is to look how community is able to generate a GIS content
23:34:24 <ser> and i need to provide them tools for an experiment
23:34:31 <w0lfie_> right now i'm looking at rangoon in OSM
23:34:34 <w0lfie_> it looks pretty good
23:34:40 <w0lfie_> problem is i have no idea where this data came from
23:34:44 <w0lfie_> when it was accurate
23:34:47 <w0lfie_> etc etc
23:35:01 <w0lfie_> if i get it from an official source
23:35:04 <w0lfie_> chances are its more outdated
23:35:14 <w0lfie_> but at least i can say that someone reasonably competent looked at it, etc etc
23:35:25 <w0lfie_> which is better than 'heres some data, i dunno anything about it'
23:35:25 <ser> w0lfie_: sure, but as i said, generally there is no need for "official" data
23:35:38 <ser> and official data is often not actual
23:35:50 <w0lfie_> its just a question of how risk-averse you are
23:35:58 <w0lfie_> and most people don't even think about that quantitatively
23:36:05 <w0lfie_> so frankly i wouldn't agree that there's generally no need
23:36:12 <w0lfie_> i'd say rather that most people don't really have a clue
23:36:17 <w0lfie_> and just sorta assume good faith
23:36:36 <ser> imho 99% true is almost a true
23:36:57 <ser> and for 99% of people 99% true is enough
23:37:14 <w0lfie_> i don't think most people understand their requirements well enough to put an actual number on it
23:37:17 <ser> especially if they do not have to pay for a map
23:37:24 <w0lfie_> so i don't agree with you
23:37:41 <w0lfie_> i think most people would probably take a lower number
23:37:50 <w0lfie_> when they go buy a street map
23:37:52 <ser> please notice people do not have such a budget like youes
23:37:54 <w0lfie_> they don't look at who made it
23:38:03 <w0lfie_> they look at how flashy the cover is
23:38:07 <w0lfie_> maybe the year it was published
23:38:08 <w0lfie_> and the price
23:38:34 <w0lfie_> oh on budget you can't compete with the commercial world
23:38:44 <ser> if it would be true, nobody want to use linux
23:38:44 <w0lfie_> they're always ahead
23:39:15 <w0lfie_> for a long time many people don't
23:39:18 <w0lfie_> didn't*
23:39:20 <w0lfie_> and many people still don't
23:39:23 <w0lfie_> especially on the desktop
23:39:29 <w0lfie_> where despite many great efforts, its notably absent
23:39:49 <w0lfie_> again, because people don't make decisions based solely or even often primarily on merit
23:39:55 <ser> today was announced in poland Firefox got 50.3% of web-browser market
23:40:00 <w0lfie_> they choose things they're familiar with, and what other people use
23:40:46 <orls> sorry to butt in; been idling while tinkering with my fresh mapnik install. w0lfie's dead right; I work for a large map store, people go for National Geographic or Insight Maps (OK and awful respectively) despite there being far, far better stuff around.
23:41:24 <w0lfie_> it's because people are lazy decision makers
23:41:27 <orls> Insight have a big fat Discovery Channel logo on them - people jump straight for them ;)
23:41:32 <w0lfie_> and most of the time its not worth putting effort into it
23:41:41 <ser> of course chaps, you are right if capitalism persists
23:41:42 <w0lfie_> frankly its often rational to make poor decisions for that reason
23:42:02 <w0lfie_> because its cheaper to do so
23:42:14 <w0lfie_> you can spend hours researching the best map book to have in your car
23:42:26 <w0lfie_> and you probably wont really reap benefits of that later on if you only casually look at it
23:42:40 <w0lfie_> now if you're say.. a taxi driver, or work for a delivery company
23:42:49 <w0lfie_> then yeah, you might be more inclined to consider the problem more carefully
23:43:09 <ser> everything you say is true if you have still your credit card handy
23:43:20 <ser> but imagine it disapears
23:43:33 <w0lfie_> i don't use credit cards
23:43:55 <ser> you are in 0.0001% of citizens
23:44:18 <w0lfie_> i know lots of people that don't use credit cards
23:44:27 <w0lfie_> doesn't mean we don't have them
23:44:50 <ser> hah, so you have one but you do not use it? why do you have it?
23:45:10 <w0lfie_> same reason i own a printer that i use maybe 3x a year
23:45:18 <w0lfie_> occasional need
23:45:56 <w0lfie_> anyway we're pretty far offtopic
23:46:24 <ser> no, because imho people are taking un-rational decisions because they have a pleny of money
23:46:38 <ser> and it may change
23:46:41 <w0lfie_> i don't see how that factors into OSM
23:46:53 <ser> OSM is free
23:46:54 <w0lfie_> since you dont have to pay to use google or MSVE's map tiles
23:47:13 <w0lfie_> it's easy to get free map data in the US
23:47:22 <w0lfie_> outside of the US is a different story
23:47:36 <ser> and if people will have no money to buy advertised products, google will not survive and will not offer a free maping service :)
23:47:52 <w0lfie_> and if you're a large company; it's probably a better business decision to pay NAVTEQ for data anyway
23:47:55 <w0lfie_> if you're talking road data
23:48:00 <w0lfie_> again, risk
23:48:21 <w0lfie_> because when it matters; smart decision makers will take a quantifiable risk over an unknown one
23:48:25 <ser> what kind of risk, UK is almost fully ready
23:48:37 <w0lfie_> we've been over this
23:48:41 <w0lfie_> it's not authoritative
23:49:04 <ser> authoritive enough to visit an aunt
23:49:44 <w0lfie_> that's not much of a selling point
23:49:53 <w0lfie_> OSM - Good enough to find directions to your relatives!
23:50:20 <ser> w0lfie_: currently everything is business oriented, but it looks business may not survive a financial collapse
23:50:23 <w0lfie_> personally, i think OSM probably is 95%+ accurate
23:50:35 <w0lfie_> but it's not about what i believe via intuition
23:50:44 <w0lfie_> its about what i can get backed
23:51:06 <ser> w0lfie_: sure, but wikipedia is still improving their quality
23:51:15 <ser> and the same would be with OSM
23:51:22 <w0lfie_> you're not listening
23:51:32 <w0lfie_> it's not quantified
23:51:35 <w0lfie_> it's not authoritative
23:52:08 <w0lfie_> it could be made authoritative
23:52:10 <ser> what does it mean, authoriitative? who is your authority?
23:52:11 <w0lfie_> but that hasn't been done
23:52:42 <w0lfie_> do you believe wikipedia is authoritative?
23:52:58 <ser> w0lfie_: no. as well as britannica
23:53:00 <w0lfie_> if you read an article, can you tell me why i should believe it's accurate?
23:53:04 <w0lfie_> exactly
23:53:08 <w0lfie_> then it's not authoritative
23:53:18 <ser> but who is?
23:53:39 <w0lfie_> if i get data from NGA
23:53:42 <w0lfie_> and it's wrong
23:53:45 <w0lfie_> then they're responsible
23:53:48 <w0lfie_> if i get data from OSM
23:53:50 <w0lfie_> and it's wrong
23:53:52 <w0lfie_> nobody is responsible
23:54:20 <ser> yes, if you buy a windows and it is wrong, you can call microsoft and receive confirmation it is wrong
23:54:29 <ser> and nobody is responsible
23:55:02 <w0lfie_> if i'm running a business
23:55:05 <ser> until you pay double fare to repair it
23:55:06 <w0lfie_> or let's say a county government
23:55:12 <w0lfie_> and i buy servers
23:55:16 <w0lfie_> let's say for databases
23:55:27 <w0lfie_> and i have the choice of buying from some random guy on craigslist
23:55:33 <w0lfie_> or buying from a reputable company
23:55:42 <w0lfie_> what's the correct decision?
23:55:51 <ser> w0lfie_: no, microsoft has excluded any guarantee of quality of software and your data
23:56:15 <w0lfie_> that's because legally you havent bought software
23:56:17 <w0lfie_> you've rented it
23:56:27 <w0lfie_> you're a licensee
23:56:30 <w0lfie_> according to their EULA
23:56:36 <ser> you cannot buy any software
23:56:44 <ser> or maps
23:56:51 <ser> all is licensed only
23:56:51 <w0lfie_> i can't maps?
23:56:57 <w0lfie_> er buy*
23:56:59 <ser> yes, you cannot
23:57:00 <w0lfie_> i can here
23:57:10 <ser> no, you can buy a license
23:57:10 <w0lfie_> i have a map book right here
23:57:13 <w0lfie_> i can sell it to someone
23:57:14 <w0lfie_> want to buy it?
23:57:25 <ser> yes, you can sell a paper
23:57:30 <ser> a copy
23:57:35 <ser> but not a map itself
23:57:55 <ser> you cannot reproduce it as well
23:58:05 <w0lfie_> not outside of fair use, no
23:58:16 <ser> so it is not your map
23:58:21 <w0lfie_> depends where i got it
23:58:23 <w0lfie_> i have some USGS data
23:58:31 <w0lfie_> i can do whatever i want
23:58:38 <w0lfie_> want some?
23:59:07 <ser> not whatever, the author rights are not transferable
23:59:14 <w0lfie_> yeah they are
23:59:15 <w0lfie_> its public domain
23:59:32 <ser> no, author still have rights to them
23:59:49 <ser> not transferable, it is a basic stuff from the IP law