Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
"Martijn van Oosterhout" wrote: > So to me it seems the dataset is a catalyst for a huge about of > programming work. Basically, open-source GIS is waking up due to the > availability of the dataset You're intentionally provoking dissense here, don't you !? ;-) Martin. -- Unix _IS_ user friendly - it's just selective about who its friends are ! -- ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
On Sat, Aug 2, 2008 at 8:45 PM, Frederik Ramm <[EMAIL PROTECTED]> wrote: > Hi, > >> For the last few months I've been wondering if OSM isn't more of a >> software project than a database. > > An interesting aspect. What I actually find interesting is that OSM as a project has subsequently lead to develeopment spurts in related projects, like Mapnik, OpenLayers, Marble (to greater or lesser extent) and all sorts of routing and mapping programs written from scratch. Mobile applications are also starting to appear. So to me it seems the dataset is a catalyst for a huge about of programming work. Basically, open-source GIS is waking up due to the availability of the dataset Have a nice day, -- Martijn van Oosterhout <[EMAIL PROTECTED]> http://svana.org/kleptog/ ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
Hi,
> But that whole argument is a bit theoretical, I wouldn't talk like that
> in public... the press will only quote the wrong parts ("Ramm: OSM
> basically worthless!").
@talk *is* in public :) .
Best regards,
ce
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
Hi,
> For the last few months I've been wondering if OSM isn't more of a
> software project than a database.
An interesting aspect.
It is very likely that none of the data we collect now will still be
used 20 years from now, because by then everything is so networked and
fully automatic and we have high resolution satellite images of
everywhere etc. etc. - will I then sit there and think it was all for
naught? Surely not, because the availability of free data *now* makes
sure that the market value of geodata goes down (makes ist more likely
that government agencies will provide them free), and also encourages
people to develop interesting techniques and software to work with that
data.
So in a way, it is possible that the database we are creating is just a
placeholder for future free geodatabases, and will be scrapped once a
better data source becomes available. But what we (collectively) learn
and develop working with OSM is very likely to still be of value even if
our data isn't any more.
But that whole argument is a bit theoretical, I wouldn't talk like that
in public... the press will only quote the wrong parts ("Ramm: OSM
basically worthless!").
Bye
Frederik
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
For the last few months I've been wondering if OSM isn't more of a software project than a database. I know everyone is spending more time mapping than writing their software, but coding, testing and documenting high performance and / or cutting edge software commercially will cost a lot more per man hour than inputting street names. Drawing the vectors are becoming easier and easier, as we switched from gpx to Y! applet to potlatch and now to mobile mapping. Developing the software has not really become easier. One could argue that many of the software development iterations could be skipped, but those experiments help us to find out what's possible and what's needed. On Tue, Jul 29, 2008 at 12:21 AM, Frederik Ramm <[EMAIL PROTECTED]> wrote: > Hi, > >the similarities between OSM and Wikipedia are many, and easily > spotted. In fact, we owe a lot of our success to Wikipedia as a "trail > blazer" - if I tell someone "we're like a Wikipedia for maps", that > saves me about 5 minutes explaining. > > However, there are also many conceptual differences between our > respective projects, and I would like to list a few of these that I've > been thinking about lately. > > I believe that some people are very quick to simply transfer "lessons > learned" from Wikipedia onto OSM, sometimes without properly taking into > account that while there are similarities, there are also lots of > differences. > > 1. One World > > In OSM, everything we have is in one database. It would be technically > possible to set up osm.de, osm.org, osm.fr etc. with national data sets > and just let everybody go along. It would even be possible to allow each > of these databases to contain a map of Karlsruhe, each styled > differently, with the French map of Karlsruhe highlighting those bits of > the city that seem important to the French and the American map focusing > on other stuff. Occasionally, users of OSM America would copy some bits > about Karlsruhe from OSM France and vice versa. All tagging would > conveniently be done in the native language of the community. If OSM > Estonia doesn't feature Reigate, then obviously Reigate is not > culturally important to Estonians, and who cares. > > This is how Wikipedia would do it. To a newcomer this looks very > puzzling at first - why should there be 50 independently authored > articles explaining how a laser works when there is one simple truth > that just has to be translated? But Wikipedia has considerable success > with this scheme, and probably avoids a million pitfalls. > > OSM has only one database that is supposed to contain the truth(tm). If > the Estonians and the Londoners cannot agree on how Reigate should be > mapped, we have a problem; Wikipedia wouldn't. > > 2. Commercially Valuable Product > > OSM is creating something of considerable commercial value. The > estimated market volume of geodata in Europe is way over one billion > Euros per year (I found varying figures, some even say it's 1.5 billion > for Germany alone, others are more conservative). - I'm sure there was a > market for encyclopedias before Wikipedia arrived but it cannot have > been this big, ever. Or can it? Let me hear figures if you have some. > > This might make a difference in attracting funding. I could imagine, for > example, that OSM could be much more successful in talking to individual > sponsors, whereas Wikipedia usually turns to the community to raise money. > > 3. Not an End Product > > Working with Wikipedia, what you see is what is there: You always have > the current version of some article in front of your eyes, and you will > usually access this product with your web browser and, ultimately, your > eyes. Wikipedia does not collect raw data, it collects/creates an end > product. In contrast, OSM does collect data, and you only ever see a > highly processed version of it. I'm sure there are *some* people who use > Wikipedia articles as some sort of text body over which to run > statistical analyses and so on, but certainly not to the degree this is > done over here at OSM. > > This means, among other things, that OSM will always be one more step > away from the unsuspecting user - OSM is about what is "behind" the map > you see. Makes some things more complicated. Also, this means that > software is likely to play a greater role in OSM than it does in Wikipedia. > > > Just a few ideas. - Not meant to be negative about Wikipedia in any way, > it's a great project that I use a lot. Just pointing out where we are > different. I'm sure you will have additional ideas about differences? > > Bye > Frederik > > -- > Frederik Ramm ## eMail [EMAIL PROTECTED] ## N49°00'09" E008°23'33" > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
> So it may be that it sounds like a good idea to be a "data provider" and > that other people will provide the primary user-facing interface to your > data, but that in fact if you want it done well what you have to do is > go out there and do it yourself. :-) > > We're currently caught between the two positions. > > If we are only a data provider, why is the Cycle Map not hosted > elsewhere and linked to from www.openstreetmap.org, along with any other > interesting maps and views that people provide? The cycle map _is_ hosted elsewhere. There is no OSM(F) resources being used to design, render, or host this map... other than the weekly planet dump we grab. The osm front page happens to link the tiles directly... as do quite a few other people it has to be said. Last month it served approximately 200GB of tiles... I think about half of them have referrers other than the osm.org website. So the osm.org website could disappear tomorrow, and we would still be providing this map. Dave ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
Frederik Ramm wrote: > 2. Commercially Valuable Product > > OSM is creating something of considerable commercial value. The > estimated market volume of geodata in Europe is way over one billion > Euros per year (I found varying figures, some even say it's 1.5 billion > for Germany alone, others are more conservative). - I'm sure there was a > market for encyclopedias before Wikipedia arrived but it cannot have > been this big, ever. Or can it? Let me hear figures if you have some. I suspect that if Wikipedia took Google ads, their revenue would be in the hundreds of millions of dollars per year. They are the top hit in Google for most factual queries, and people read it looking for facts and info (rather than entertainment) and it's a short step from their to purchasing. So their data also has considerable commercial value, although the value is associated with the eyeballs viewing the most-commonly-used expression of the data (which they control) rather than the data itself. > 3. Not an End Product Not to contradict what you've said, but maybe there is an interesting parallel here between OSM and mozilla.org. Originally, mozilla.org was a "technology provider", the idea being that lots of different companies and organizations would build Foo Browser and Bar Browser and be the distributors. Netscape was the biggest, but they did a fairly poor job of it and still there weren't really many others. After mozilla.org split from AOL/TW/Netscape, we went into the browser business ourselves. The result is Firefox. So it may be that it sounds like a good idea to be a "data provider" and that other people will provide the primary user-facing interface to your data, but that in fact if you want it done well what you have to do is go out there and do it yourself. :-) We're currently caught between the two positions. If we are only a data provider, why is the Cycle Map not hosted elsewhere and linked to from www.openstreetmap.org, along with any other interesting maps and views that people provide? Why doesn't the default map show everything including errors and maplint, so we can more easily see what's there and what's not? But if we are, in fact, the primary front end, then we should decide to go for it, get some super-fast hardware, host as many layers of interest as we can find, and tell everyone to come to www.openstreetmap.org to get their maps rather than maps.google.com. Gerv ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
Lars Aronsson wrote: > Frederik Ramm wrote about OSM vs. Wikipedia: > >> Wikipedia does not collect >> raw data, it collects/creates an end product. > > This description of Wikipedia is wrong. It's not, because... > It would be better for Wikipedia if more readers went to other > mirror websites ...is the difference. The cyclemap isn't a mirror. OSM-on-Garmin isn't a mirror. OpenRouteService isn't a mirror. They are "creative and unexpected uses of the data" - heck, even [EMAIL PROTECTED] could be considered not a mirror. Whereas all those pagerank exercises that rehost Wikipedia to get some Google Adsense income are just that - mirrors. They add nothing to the original content. So when you say > Just like OSM, Wikipedia > is about compiling free contents. How this is presented can be > determined by the user, who downloads the database dump and > converts it to something useful: on the web, on CDROM or in print. it kind of ignores the fact that a good 20% of OSM's userbase is involved in alternative presentations of the data, whereas barely 2% of Wikipedia page views come through anything other than the default Mediawiki view at somethingorother.wikipedia.org. [1] cheers Richard [1] spurious statistics entirely made up for purpose of proving argument ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
Frederik Ramm wrote about OSM vs. Wikipedia: > 3. Not an End Product > > Working with Wikipedia, what you see is what is there: You > always have the current version of some article in front of your > eyes, and you will usually access this product with your web > browser and, ultimately, your eyes. Wikipedia does not collect > raw data, it collects/creates an end product. This description of Wikipedia is wrong. Just like OSM, Wikipedia is about compiling free contents. How this is presented can be determined by the user, who downloads the database dump and converts it to something useful: on the web, on CDROM or in print. What you happen to see on Wikipedia's website is just one example. In this respect, Wikipedia works exactly like OpenStreetMap does. In both cases, many users are (mis-)led to believe that what they see is the one and only end product. When Wikipedia's website is one of the world's ten most visited ones, this is just a big cost, and not the purpose of Wikipedia. It would be better for Wikipedia if more readers went to other mirror websites, so that Wikipedia's servers could just serve the active contributors. -- Lars Aronsson ([EMAIL PROTECTED]) Aronsson Datateknik - http://aronsson.se ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Frederik Ramm wrote: | Hi, | | the similarities between OSM and Wikipedia are many, and easily | spotted. In fact, we owe a lot of our success to Wikipedia as a "trail | blazer" - if I tell someone "we're like a Wikipedia for maps", that | saves me about 5 minutes explaining. | | However, there are also many conceptual differences between our | respective projects, and I would like to list a few of these that I've | been thinking about lately. | | I believe that some people are very quick to simply transfer "lessons | learned" from Wikipedia onto OSM, sometimes without properly taking into | account that while there are similarities, there are also lots of | differences. | | 1. One World | | 2. Commercially Valuable Product | | 3. Not an End Product I agree with this strongly, and believe it's worth mentioning that in all 3 aspects, MusicBrainz and possibly dmoz.org and similar projects are a lot closer to us than Wikipedia is. voxforge.org is also building something interesting which will have similar issues to us. Robert (Jamie) Munro -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.8 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkiPANAACgkQz+aYVHdncI1UvwCgmp79bz+Q3WJlYXfs2Vxi+KXz nmQAoKRFHkIL6I44DXdPu6TZrrZaETUC =nWmh -END PGP SIGNATURE- ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
On Jul 29, 2008, at 0:21 , Frederik Ramm wrote: > > I believe that some people are very quick to simply transfer "lessons > learned" from Wikipedia onto OSM, sometimes without properly taking > into > account that while there are similarities, there are also lots of > differences. There's another difference, which is quite important (to me at least). Wikipedia collects knowledge in general and a great deal of this knowledge (if not most of it) is partly subjective; in the end, the good faith of its contributors and the existence of a mechanism to verify it is important. Furthermore, there is stuff where the "objective truth" doesn't exist at all - all of this bring up the point of how much one trust in Wikipedia, if you prefer such an approach or the traditional one with an editor, a board of controllers, etc... On the contrary, OSM is documenting mostly factual data based on empirical observation (the GPS tracks). Yes, there are the boundary controversies etc, but fortunately they involve only a part of the world. Summing up, there are no strong problems of trust in OSM, while there are in Wikipedia, IMHO. -- Fabrizio Giudici, Ph.D. - Java Architect, Project Manager Tidalwave s.a.s. - "We make Java work. Everywhere." weblogs.java.net/blog/fabriziogiudici - www.tidalwave.it/blog [EMAIL PROTECTED] - mobile: +39 348.150.6941 ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] Why OpenStreetMap is not Wikipedia
Frederik Ramm wrote: >Sent: 28 July 2008 11:22 PM >To: Talk Openstreetmap >Subject: [OSM-talk] Why OpenStreetMap is not Wikipedia > >Hi, > >the similarities between OSM and Wikipedia are many, and easily >spotted. In fact, we owe a lot of our success to Wikipedia as a "trail >blazer" - if I tell someone "we're like a Wikipedia for maps", that >saves me about 5 minutes explaining. > >However, there are also many conceptual differences between our >respective projects, and I would like to list a few of these that I've >been thinking about lately. > >I believe that some people are very quick to simply transfer "lessons >learned" from Wikipedia onto OSM, sometimes without properly taking into >account that while there are similarities, there are also lots of >differences. > >1. One World > >In OSM, everything we have is in one database. It would be technically >possible to set up osm.de, osm.org, osm.fr etc. with national data sets >and just let everybody go along. It would even be possible to allow each >of these databases to contain a map of Karlsruhe, each styled >differently, with the French map of Karlsruhe highlighting those bits of >the city that seem important to the French and the American map focusing >on other stuff. Occasionally, users of OSM America would copy some bits >about Karlsruhe from OSM France and vice versa. All tagging would >conveniently be done in the native language of the community. If OSM >Estonia doesn't feature Reigate, then obviously Reigate is not >culturally important to Estonians, and who cares. > >This is how Wikipedia would do it. To a newcomer this looks very >puzzling at first - why should there be 50 independently authored >articles explaining how a laser works when there is one simple truth >that just has to be translated? But Wikipedia has considerable success >with this scheme, and probably avoids a million pitfalls. > >OSM has only one database that is supposed to contain the truth(tm). If >the Estonians and the Londoners cannot agree on how Reigate should be >mapped, we have a problem; Wikipedia wouldn't. > >2. Commercially Valuable Product > >OSM is creating something of considerable commercial value. The >estimated market volume of geodata in Europe is way over one billion >Euros per year (I found varying figures, some even say it's 1.5 billion >for Germany alone, others are more conservative). - I'm sure there was a >market for encyclopedias before Wikipedia arrived but it cannot have >been this big, ever. Or can it? Let me hear figures if you have some. > >This might make a difference in attracting funding. I could imagine, for >example, that OSM could be much more successful in talking to individual >sponsors, whereas Wikipedia usually turns to the community to raise money. > >3. Not an End Product > >Working with Wikipedia, what you see is what is there: You always have >the current version of some article in front of your eyes, and you will >usually access this product with your web browser and, ultimately, your >eyes. Wikipedia does not collect raw data, it collects/creates an end >product. In contrast, OSM does collect data, and you only ever see a >highly processed version of it. I'm sure there are *some* people who use >Wikipedia articles as some sort of text body over which to run >statistical analyses and so on, but certainly not to the degree this is >done over here at OSM. > >This means, among other things, that OSM will always be one more step >away from the unsuspecting user - OSM is about what is "behind" the map >you see. Makes some things more complicated. Also, this means that >software is likely to play a greater role in OSM than it does in Wikipedia. > > >Just a few ideas. - Not meant to be negative about Wikipedia in any way, >it's a great project that I use a lot. Just pointing out where we are >different. I'm sure you will have additional ideas about differences? > You have summed up very well my own feelings. I've no idea if OSM really has learnt from any other project or if its steered its own course, Steve kicked off with the cookbook, but so many of the ingredients for the recipe have been change along the way that I'm not sure if most of us really do may a comparison with other projects. It is very right I feel that we reinforce not so much our differences to other projects but rather the specific strengths that OSM has. They are basically in what you have written and I'd list them (in terms of comparing with the alternatives out there) as: * Unique * Valuable * Reliable Cheers Andy ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
