#OSGEO IRC Log - 2008-11-15

For logs after Feb 3, 2007, all times are GMT-8. Prior logs are GMT-9.
Back to Logs
02:55:31 steko: http://blog.cleverelephant.ca/2008/11/what-is-osgeo-becoming.html
02:55:32 sigq: Title: Paul Ramsey: What is OSGeo becoming? ( at blog.cleverelephant.ca )
03:02:58 CIA-54: osgeo: yjacolin * r1284 /journal/volume_4/fr/annual_report/executive/ ( editorial.tex news.tex ): French translation for news, ... article Vol 4
03:02:58 sigq: osgeofeed: Changeset [1284]: French translation for news, ... article Vol 4 <http://trac.osgeo.org/osgeo/changeset/1284>
03:07:08 sigq: osgeofeed: Changeset [1286]: French translation for incubation article Vol 4 <http://trac.osgeo.org/osgeo/changeset/1286>
03:07:21 CIA-54: osgeo: yjacolin * r1285 /journal/volume_4/fr/annual_report/committees/incubation.tex: French translation for incubation article Vol 4
03:07:22 CIA-54: osgeo: yjacolin * r1286 /journal/volume_4/fr/annual_report/committees/incubation.tex: French translation for incubation article Vol 4
03:08:22 CIA-54: osgeo: yjacolin * r1287 /journal/volume_4/fr/annual_report/committees/journal.tex: French translation for journal article Vol 4
03:08:22 CIA-54: osgeo: yjacolin * r1288 /journal/volume_4/fr/annual_report/committees/journal.tex: French translation for journal article Vol 4
03:08:22 sigq: osgeofeed: Changeset [1288]: French translation for journal article Vol 4 <http://trac.osgeo.org/osgeo/changeset/1288>
03:11:09 sigq: osgeofeed: Changeset [1289]: final release for some article ( accent to latex tag ) <http://trac.osgeo.org/osgeo/changeset/1289>
03:11:39 CIA-54: osgeo: yjacolin * r1289 /journal/volume_4/fr/annual_report/ ( 4 files in 3 dirs ): final release for some article ( accent to latex tag )
04:34:06 seven: Hey. Someone around with English onomatopoetic proficiency? I am trying to understand Paul's blog ( as advised by steko ) and stumble on Glurp! What is he trying to tell us?
04:34:25 seven: As per http://www.urbandictionary.com/define.php?term=glurp means "Someone who farts in the bathtub and eats the bubbles"
04:34:26 sigq: Title: Urban Dictionary: glurp ( at www.urbandictionary.com )
04:34:52 seven: or "A burp that results in partially digested food coming back up."
04:34:52 seven: :- )
04:35:05 steko: seven: I think the "Glurp!" means "I hope what he's saying is not true, otherwise we are in serious trouble"
04:35:15 steko: does it make sense?
04:35:31 seven: Yes, that sounds reasonable.
04:35:39 seven: Btw. who is james Fee?
04:36:04 seven: Never saw him around here but he seems to have a strong opinion if I get it right.
04:37:01 steko: seven: I think he's the guy who owns http://www.spatiallyadjusted.com/
04:37:02 sigq: Title: James Fee GIS Blog Geospatial Technology, Web Mapping and Spatial Services ( at www.spatiallyadjusted.com )
04:38:47 seven: Cool. Big guy, we should listen to him. Good thing that he notices us. Must have been the logos. :- )
04:39:48 steko: yes, and the point he's raising is not trivial
04:39:57 steko: he's not a misinformed person for sure
04:44:08 steko: I think one possible solution to that problem could be putting all foss4g materials "upstream" on osgeo wiki, or something like that.
04:51:14 seven: Yes this should be high on our agenda, but besides we have to run an organization. Unfortunately there is just so many people actively helping.
04:52:25 seven: For what its worth, the history of the WIki shows where people invest time. This one for example unfortunatley is not that high on the agenda: http://wiki.osgeo.org/wiki/Advocacy
04:52:26 sigq: Title: Advocacy - OSGeo Wiki ( at wiki.osgeo.org )
04:59:15 steko: yes, but I can see lots of advocacy activities at local scale that perhaps aren't well integrated into "upstream" osgeo wiki
06:14:20 darkblue_B: ajturner: hello
06:15:09 ajturner: hey darkblue_B
06:15:44 darkblue_B: did anyone talk to you about the details of the Obama Houdini system?
06:16:10 darkblue_B: apparently it crashed by 11:30 Central time
06:16:44 darkblue_B: I dont know any of the inside story.. apparently it worked very well though
06:17:29 darkblue_B: not sure how much visualization it had.. but very thorough coverage of their voter leads for Election day
06:18:10 darkblue_B: .. at any rate, I have inported the data feeds from California and the US votereport, a couple of different ways
06:18:40 darkblue_B: but, stitching together the Location and Reporter records from inside the Record, to be a real relational DB has been tricky
06:19:03 darkblue_B: I am a good programmer, but still klunking along with Pg SQL
06:19:35 darkblue_B: I have tried to do a few things on the SQL/plpgsql side and its not very co-operative
06:19:57 darkblue_B: ( makes me appreciate the Refractions crew all the more )
06:20:31 darkblue_B: how do you handle the data stream.. Record inside Atom, I suppose
06:21:04 darkblue_B: .. do you bother with it once it has been generated? or do you handle it all before it is emitted as XML/JSON/etc ?
06:21:54 ajturner: darkblue_B: what is the Obama Houdini system?
06:22:39 ajturner: I'm not quite following you or what you're trying to do?
06:23:17 darkblue_B: 2 answers
06:24:08 darkblue_B: Houdini - I was told by an Indiani poll worker that the Democrats had a system in place.. IN requires ID to vote. the Demo poll worker had a print out of tear-off tabs
06:24:28 darkblue_B: .. with all of the names of anyone for that polling place, that had expressed interest in Obama
06:24:57 darkblue_B: when someone came and idenitified themselves, the Demo volunteer tore off the tab for that name
06:25:32 darkblue_B: .. and put it in a bag.. later, those remaining were phoned in , so that other call volunteers could contact those that had not voted by phone
06:25:54 darkblue_B: it apparently worked very well.. until ti crashed ;- )
06:26:26 darkblue_B: 2 ) your votereport data is great.. so I want to do various visualizations with it
06:26:31 ajturner: how did a bag of names crash?
06:26:45 darkblue_B: ;- ) the national call in system crashed
06:26:54 Schuyler: that's an IRC channel topic if I ever heard one
06:27:02 darkblue_B: hey Schuyler
06:27:12 Schuyler: I didn't do it!
06:27:18 darkblue_B: :p
06:27:28 Schuyler: I mean
06:27:31 Schuyler: Hi!
06:28:02 darkblue_B: 2 ) votereport data is great.. I wrote a small python program that sucked in the cached data feeds all day
06:28:23 darkblue_B: but since it didnt know when the data was new, it got a lot of duplicated
06:28:24 ajturner: the db is available for download I believe
06:28:30 darkblue_B: ah
06:28:35 ajturner: thre are unique "guids" for each report too
06:28:41 ajturner: so you can strip out dupes easily
06:28:48 darkblue_B: yes, each report, I have done that
06:28:56 ajturner: and can also walk the entire db in the feeds - there is pagination
06:29:06 darkblue_B: create table t2 as select distinct on ( id ) * from tblOrig;
06:29:30 darkblue_B: sure, but the report has locations, and reporter
06:29:38 darkblue_B: those are sub-records, as in a relation
06:29:48 darkblue_B: its hard to stitich that together right
06:29:57 darkblue_B: I can do it but some hand programming
06:30:17 darkblue_B: yea, I would like to copy the db and look.. I'll be interested if they do the relations
06:30:29 darkblue_B: 600 locations and 800 reports, for example
06:30:33 darkblue_B: I saw
06:32:46 ajturner: there is some estimation of a Report has-a Location
06:33:06 ajturner: but not great, since most were just submitted w/ Lat/Lon - and not 'snapped' to nearest polling location
06:42:38 darkblue_B: ajturner: I see the schema in ActiveRecord at github.com/davetroy/votereport/tree/master/db
06:42:49 darkblue_B: .. but I dont see the data itself yet.. url?
06:43:27 : * ajturner is looking in the mail list
06:45:10 ajturner: hrm, darkblue_B - email Dave Troy
06:45:14 crschmidt: seven: "Glurp" is probably the onomantopaeic version of "gulp"
06:45:21 ajturner: for a dump- he had one that was a little more cleaned up
06:45:57 crschmidt: seven: James Fee is a long time well-known "ESRI" blogger who has ( within the past year ) started moving many of his clients to Open Source solutions.
06:48:49 crschmidt: seven: He is a smart, opinionated guy. He has funded some development work in OpenLayers ( about a year ago ) for a project he was working on, and has long been a proponent of seeking out Open Source solutiosn that work -- especially ones that allow you to replace more expensive ArcGIS licenses with things like MapServer + ArcSDE support.
06:56:51 ajturner: crschmidt: what is the comment seven made ( my log doesnt' go back that far )
06:57:54 darkblue_B: they were asking about what Glurp might mean in his blog
06:58:04 darkblue_B: - though I just looked for it and dont see it
06:58:25 steko: http://blog.cleverelephant.ca/2008/11/what-is-osgeo-becoming.html
06:58:26 sigq: Title: Paul Ramsey: What is OSGeo becoming? ( at blog.cleverelephant.ca )
06:58:32 steko: glurp is here ; )
07:09:18 ajturner: I agree with JFee
07:09:50 darkblue_B: hmm.. I think his rage against the logo is misplaced, but, I agree with JFee also
07:09:51 ajturner: OSGeo and OSM are having problems maturing beyond the hacker ethos - i.e. going from Startup to Enterprise
07:10:20 ajturner: for an example, reference the discussion at the AGM on needing Corporate Materials for sponsorship and the lashback
07:10:29 darkblue_B: I question whether enterprise is the only model for maturity! but, I think something new is desirable
07:10:49 steko: ajturner: but I think it would be wrong to put OSM and OSGeo together with respect to maturity
07:11:01 crschmidt: I agree: OSM is much more mature
07:11:11 crschmidt: and successful at market penetration
07:11:19 darkblue_B: much more under Venture Capital pressure, too ;- )
07:11:33 crschmidt: enh, I don't think CloudMade has anything to do with it
07:11:42 steko: crschmidt: let alone market penetration, why would you label OSM as "more mature" ?
07:12:47 crschmidt: steko: Wider general knowledge about the project. More success at integrating itself into all kinds of marketing venues. Wider net of contributors, better marketing materials, greater number of local groups pushing forward OSM development and data collection.
07:13:03 darkblue_B: OSM is a phenomenon, like Wikipedia.. dont need to try to exactly compare
07:13:05 crschmidt: In every metric that OSGeo could want to succeed in, OSM has been there, done that, got the T-Shirt.
07:13:43 steko: I mean, James Fee was not saying that OSGeo is not good at marketing, but rather that it lacks a mature set of "resellable" case studies. I'm not convinced of that, however
07:14:54 crschmidt: That was an example of OSGeo's failure to create materials which help to push forward its agenda.
07:15:56 ajturner: my point is that both are being looked at and potentially used in very large scale companies/projects - and perceptions are important and how they present themselves to continue this adoption beyond just hackers
07:16:08 ajturner: look at Linux ( Torvalds ) vs. GNU ( Stallman )
07:16:46 : * crschmidt isn't sure whether that's a 'vs.', since he puts them both in the same boat : )
07:17:12 ajturner: Linux is comfortable ( now ) for enterprises, GNU is scary
07:17:14 crschmidt: "Red Hat" or "ubuntu" vs. "linux", I could see as being a comparison, maybe? or maybe FSF is successful outsdie the hacker world and I just don't see it
07:17:17 darkblue_B: well I have done an MBA program, and we spent months reading and talking about market position and values
07:17:34 crschmidt: Well, I think Linux is comfortable because of resources other than Linus...
07:17:42 darkblue_B: .. of course ajturner is right about taking it seriously
07:18:14 ajturner: Linus has a better personality for a public face, than the "crazy" that is Stallman
07:18:15 ajturner: : )
07:18:45 : * crschmidt has spent enough time in the same ( stinky ) room as stallman to be aware of that ; )
07:19:22 darkblue_B: stallman is crazy, and pure values driven, and his band of followers has changed the game
07:20:00 steko: ... but going back to the source of discussion... what should OSGeo do to go beyond this blocking problem?
07:20:41 darkblue_B: a formal set of professional case studies sounds like a reasonable thing to do
07:20:55 darkblue_B: I notced inthe board report they have some money
07:21:03 darkblue_B: the certainly paid ofr the visual Identity part
07:21:23 darkblue_B: some professional design of full glossy studies may not be a bad investment
07:21:57 darkblue_B: I have noticed some aversion to monetization, and I think thats a mistake to let go too far
07:22:07 steko: true
07:22:08 crschmidt: darkblue_B: Whether we should spend money on case studies ( in my opinion ) deserves further consideration. I'm not convinced that "Because James Fee says it would help him" is enough evidence that Case Studies are important.
07:22:10 darkblue_B: pure values is powerful, I support it
07:22:26 darkblue_B: .. but pure values alone isnt enough
07:22:55 darkblue_B: crschmidt: healthy scepticism, and, I think this is a good thing
07:23:49 darkblue_B: ( meanwhile I think I can save my attempt at a poster for GIS day ;- )
07:24:04 steko: and, before making some shiny promotional materials, we should question which is the intended audience
07:24:17 darkblue_B: steko: yes, of course
07:24:37 steko: and most importantly *where* is the audience : )
07:25:03 darkblue_B: we had a panel at OSBootCamp on the effects of the economy on FOSS
07:25:22 darkblue_B: recently, and Josh Berkus said "I expect to be very busy in the coming years"
07:25:24 ajturner: steko: the discussion I brought up at the AGM was that OSGeo wants more sponsorship and it's tapping out the "I'm friend with so&so that owns the company"
07:25:29 steko: I mean, local action has its weight, despite the importance of global networks of information
07:25:55 darkblue_B: ".. when times are good, CEOs like to brag about their acquisitions and expansion. When times are hard, CEOs like to brag about their cost savings.."
07:26:18 darkblue_B: "FOSS is clearly a cost savings.. therefore, I believe I will be geting a lot of calls for FOSS consulting"
07:30:11 darkblue_B: steko: ohh, localizable glossy material.. thats good actually
07:31:47 steko: darkblue_B: strategies should be localizable and glossy, not just flyers and posters : )
07:36:04 darkblue_B: steko: probably not possible, really.. Brazil, and Japan, and DACH ??
07:36:17 darkblue_B: just too different
07:36:41 crschmidt: darkblue_B: What kind of strategy works in Brazil but doesn't work in Japan?
07:37:07 darkblue_B: crschmidt: What kind of strategy works in Brazil and works in Japan?
07:38:03 darkblue_B: ohh !! Fuimori strategy! of course! but he's in Jaill now
07:38:52 crschmidt: Perhaps I don't understand what you mean by strategy. For me, "OpenLayers is a project that allows you to use your own data in a Javascript mapping application quickly and easily" has worked well as a strategy for getting corporate usage in both Brazil and Japan.
07:39:25 darkblue_B: http://en.wikipedia.org/wiki/Alberto_Fujimori
07:39:26 sigq: Title: Alberto Fujimori - Wikipedia, the free encyclopedia ( at en.wikipedia.org )
07:40:18 darkblue_B: sales is high-touch.. to be 'professional', eg responsible, very expert, have some money
07:40:34 darkblue_B: .. is communicated differently in different cultures
07:40:48 steko: crschmidt: yes, because it's just a software strategy. but things are different when you try to promote a whole new philosophy ( with related software stack ) in government agencies e.g.
07:41:40 crschmidt: It seems to me thatthe way a strategy is implemented is different than the strategy itself.
07:42:10 crschmidt: The strategy of "This software is better than anything else out there" works across all cultures -- how you implement that 'sale' is the part that gets 'lost in translation'
07:42:54 : * steko agrees
07:44:46 : * darkblue_B agrees
07:46:55 steko: so it's just a matter of making the better geospatial software in the solar system : )
07:56:34 steko: s/better/best
07:56:46 : * steko is too tired
08:06:20 dpatton: it can be argued that OSGeo's purpose isn't to "market", but to "support" - to provide the foundation that allows projects to have a solid base on which they can build and do their own marketing
08:09:48 darkblue_B: a trade association often does public outreach.. "marketing".. the projects in OSS are typically over worked and under resourced to do specialized work like that..
08:10:09 darkblue_B: you are right - it could be that way, dpatton, but, perhaps this is a better direction
08:10:39 darkblue_B: if it were me, I would say to do the outreach..
08:11:16 darkblue_B: btw- I am making a poster for our GIS day.. I want to add a map overlay.. any contrbutions?
08:11:32 darkblue_B: vectors... I dont think OSM would be appropriate
08:13:14 darkblue_B: http://24.7.79.159/osgeo/
08:13:15 sigq: Title: Index of /osgeo ( at 24.7.79.159 )
09:18:27 steko: hi TylerM
09:19:34 darkblue_B: TylerM: !!
09:19:45 darkblue_B: hey I have a poster for next week
09:20:06 darkblue_B: I just need a vector map to place on it.. in perspective.. you know, map over land
09:20:26 darkblue_B: the one I chose from GeoServer sample data had problems
09:20:42 darkblue_B: I ddnt figure out how to clip to a rectangle in Illutrator :-/
09:21:01 darkblue_B: dont know why it was so tricky.. the clip filter 'removes strokes'
09:21:09 darkblue_B: .. which of course was all the map detail
09:21:49 darkblue_B: any nice OSS generated cartography come to mind?
09:22:27 darkblue_B: I suppose I could use a raster in their, too.. dont know yet
09:23:33 crschmidt: darkblue_B: Why not oSM?
09:23:57 crschmidt: good open source cartography is somewhat rare -- OSM is among the absolute best, imho
09:24:16 TylerM: hi steko darkblue_B crschmidt : )
09:24:24 TylerM: darkblue_B: glad to hear it!
09:24:32 TylerM: yeah osm road rendering is really cool
09:25:21 TylerM: crschmidt: do you know the best way to pull/dump some hi res osm rendering into a single graphic file?
09:25:58 TylerM: i used to only online browsing or wms : )
09:26:18 darkblue_B: hmmm
09:26:28 darkblue_B: ok .. if you say so .. yes, OSM is among the best
09:26:45 TylerM: or were you lookng more just for some data to play with?
09:26:56 darkblue_B: no play! complete!
09:27:00 darkblue_B: :- )
09:27:09 darkblue_B: ship it!
09:27:30 crschmidt: TylerM: yes, click 'export' on the OSM hoempage
09:27:35 crschmidt: there is an option to export .ps, .pdf, etc.
09:27:42 tbonfort: here's a link to mapserver-rendered osm data: http://mapserverosm.s3.amazonaws.com/parisosm.html
09:27:56 crschmidt: simply navigate the map to what you want to see, and then export the data
09:28:20 crschmidt: tbonfort: nice
09:28:33 crschmidt: MapServer's vector export isn't so great though, no?
09:28:58 hobu: svg
09:29:11 crschmidt: Yes, that's what I'm talking about.
09:29:37 tbonfort: the pdf/svg/swf is kind of lagging behind feature-wise
09:29:52 hobu: patches, as you say : )
09:30:05 TylerM: crschmidt it can be quite nice too.. takes a lot of work regardless, imo
09:30:11 TylerM: tbonfort: beauty
09:30:44 hobu: tbonfort is working to unify the raster rendering side, someone would also need to unify the vector side ( WFS, SWF, SVG, etc ) IMO at some point too
09:30:59 hobu: it's stuff no one wants to pay for though
09:31:28 crschmidt: My point is that if darkblue_B wants something *now*, since GIS day is like, Tuesday, he's going to need something that works nice as a set of vectors before the weekend is out
09:31:38 crschmidt: and if he wants that, then OSM's vector export from their homepage works well
09:31:47 TylerM: +1 go for it darkblue_B : )
09:34:31 crschmidt: tbonfort: is the lack of zoom controls on that map intentional?
09:35:19 tbonfort: crschmidt, you don't have a mouse wheel : ) ?
09:35:23 crschmidt: tbonfort: nope
09:35:32 TylerM: heh
09:35:38 crschmidt: I don't have a mouse.
09:36:22 tbonfort: I won't teach you how to firebug that control in : )
09:37:38 crschmidt: tbonfort: I was sort of asking cause I was considering bloggerizing the link
09:39:47 tbonfort: then I guess I'll have to reread my s3 docs to upload an updated html
09:40:00 crschmidt: heh, heh
09:41:59 crschmidt: tbonfort: no need if you're not interested
09:42:03 crschmidt: I just thought it looked like a super-cool map
09:42:28 tbonfort: crschmidt, I'm cool with it, thanks for proposing
09:42:58 tbonfort: will ping you once its updated
09:43:09 crschmidt: Do you pay for bandwidth/hits from S3?
09:43:18 tbonfort: yep
09:43:45 crschmidt: so maybe bloggerizing is not good for you : )
09:47:09 darkblue_B: tbonfort: tasteful coloring in the OSM style.. I pulled out WebKit nightly for the experience
09:47:53 tbonfort: the color scheme is kind of ripped off ; )
09:49:05 crschmidt: tbonfort: you don't say : )
09:51:56 tbonfort: crschmidt: should be updated now
09:54:54 crschmidt: thx!
10:04:57 CIA-54: osgeo: yjacolin * r1290 /journal/volume_4/fr/annual_report/projects/openlayers.tex: French translation for journal article Vol 4
10:04:59 sigq: osgeofeed: Changeset [1290]: French translation for journal article Vol 4 <http://trac.osgeo.org/osgeo/changeset/1290>
11:05:18 darkblue_B: ok, the poster is up for anyone to look at ...
11:05:35 darkblue_B: http://24.7.79.159/osgeo/
11:05:36 sigq: Title: Index of /osgeo ( at 24.7.79.159 )
11:09:45 FrankW: nice!
16:26:33 darkblue_B: so I have some space left on this OSGeo poster
16:26:47 darkblue_B: I was thinking of what might go there
16:26:48 crschmidt: got a URL?
16:27:05 darkblue_B: http://24.7.79.159/osgeo/
16:27:06 sigq: Title: Index of /osgeo ( at 24.7.79.159 )
16:27:11 darkblue_B: 2
16:27:24 darkblue_B: perhaps 'code + spatial'
16:27:30 crschmidt: enough room for project names?
16:27:40 crschmidt: i don't have a pdf viewer on this client, sorry
16:27:40 darkblue_B: perhaps the 9 areas of concern for GEOSS
16:27:45 darkblue_B: oh
16:27:55 adamhill: man you guys spend too much time on logos : )
16:27:59 darkblue_B: yes, the project names with their chairs are there
16:28:08 darkblue_B: and the committtees
16:28:20 darkblue_B: .. or lastly, "What you can do"
16:28:33 darkblue_B: but since it is UC Berkeley, and the specialization is huge
16:28:50 darkblue_B: I dont thnk I could really do justice to that last one..
16:28:55 darkblue_B: .. or that it would be sensible
16:29:07 darkblue_B: hi adamhill
16:31:35 crschmidt: darkblue_B: one comment...
16:31:41 crschmidt: I might try to make the OSM layer transparent if you can
16:32:13 crschmidt: not sure if it's possible with your tools...
16:32:44 darkblue_B: hmm
16:33:16 darkblue_B: any ideas on what to fill in on the left
16:33:41 crschmidt: i might actually just leave it
16:33:51 crschmidt: there's already a lot of content, ya know
16:33:57 darkblue_B: yeah
16:36:58 darkblue_B: what about a mascot ? :P
23:24:38 Kapivara: There is an open source library with lots of image processing services with good interfaces with commercial languages like Java?
23:26:52 Kapivara: I known TerraLib, but there's no way to use it with Java.. they're working on it but will take time...