The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
Locked
12
12
The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
 
|
Congratulations Liam for your Master degree thesis
http://qspace.library.queensu.ca/handle/1974/8159 «... a study of Canadian municipal open data initiatives in order to assess the current state of the programs and to gauge the role(s) that these initiatives may play in regards to improving public engagement in local government issues.» via Open Data Research Network --- Liaison par | Curation by --- Dre Diane Mercier Ambassadrice de l'Open Knowledge Foundation - Groupe local au Canada ca.okfn.org | @okfnca | LinkedIn : dianemercier | Jabber/XMPP: [hidden email] Blogue : dianemercier.com Portail : donnees.ville.montreal.qc.ca _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
I would include the use of Free Libre and Open Source Software because open data without the right tools is not completely open data. And in open data, I would study open procurement processes. How many publish all public tenders online cost-free? My 2 cents. And Congrats. On 2013-08-30 1:53 PM, "Peder Jakobsen" <[hidden email]> wrote:
_______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
You don't need open source to have open data. One the most successful open data platforms on the planet is not built with open source software (Socrata). And most of the worlds best open data is stored in a decidedly non-open data data store which cannot be easily replaced (Oracle), nor should it.-1 This is so wrong, and on so many levels!! You cannot have open data without open source. Socrata is not the best platform, it is being replaced by CKAN. Oracle is being abandonned to the favor of MariaDB and Postgres and other open database solutions. As a subject matter, Open Source software is a very different realm than open data, something that people surprisingly get mixed up together and lump into one because they share is the word "Open". It's also something that would be difficult to write about with any depth unless you have actually worked in software development; it's perhaps the only way to separate the hype from reality….and there is a lot of hype to be sure. -1 again You don't need an degree in software development to analyse the adoption levels of FLOSS across different countries, levels of government and industries. No open data without open source! Just my 2 cents Immanuel On Mon, Sep 2, 2013 at 4:58 PM, Peder Jakobsen <[hidden email]> wrote:
_______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
Immanuel; What Stephane is saying, is that in the short term, as cities/Provs/Terr/Feds are transitioning to open data strategies, they need to cobble up what they have on hand to get data out. Be that open source or not. Once they get established a little, and do some organizational learning they can move to other platforms which in the best possible world, OS.I would argue that many of the current portals, even currently used open data OS ones, do not scale well when tens of thousands of datasets are in the systems. The TBS open data portal is an great example. Searching data with tags is by no means the best way, and common keyword vocabularies need to be used in order to actually find stuff. This is where the geomatics folks, and the librarians come into play, they have been doing catalogs and portals for a good long time. The Ottawa Public Library is an example of an excellent interoperable multi institutional catalog system. Just need to think of data as books! t On Wed, Sep 4, 2013 at 12:54 PM, Immanuel Giulea <[hidden email]> wrote:
-- _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
(Cutting out all the other lists so that we don't cross-post unnecessarily.)
Anyway, you can clearly have open data without open source. As mentioned, many governments use proprietary platforms like Socrata to manage and distribute open data, and even more governments use in-house proprietary code to do the same. A lot of open data currently published is created by proprietary software (ESRI ArcGIS) and stored in a proprietary format (ESRI Shapefile). The definition of open data does not set any requirements with respect to the licensing of the *software* used to create, manage and distribute the open data. The definition only sets requirements with respect to the licensing of the *data*. And you can clearly analyze open data and open source separately. Open source (and free software) have been around a lot longer as established concepts than open data has, and people have had no trouble analyzing open source in isolation for all that time. Anyway, no one is arguing against open source. People have simply pointed out that there is a distinction between the two concepts. James On 2013-09-04, at 8:30 AM, Tracey P. Lauriault wrote:
_______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
Hi Peder
>> I would include the use of Free Libre and Open Source Software because open data without the right tools is not completely open data. > > You don't need open source to have open data. One the most successful open data platforms on the planet is not built with open source software (Socrata). And most of the worlds best open data is stored in a decidedly non-open data data store which cannot be easily replaced (Oracle), nor should it. > > As a subject matter, Open Source software is a very different realm than open data, something that people surprisingly get mixed up together and lump into one because they share is the word "Open". It's also something that would be difficult to write about with any depth unless you have actually worked in software development; it's perhaps the only way to separate the hype from reality….and there is a lot of hype to be sure. You are right you don't need open source software to have open data technically speaking and Socrata is indeed a very successful company. That said, the common thread, I feel, is more of a belief in creating a 'commons' of data, software or whatever else, free of restrictive licenses that allow anyone to build upon, examine and reuse. If you are a software vendor like Oracle, that may not matter to you, as what's important is building a great product to serve a need, but for someone else, those principles of openness count whether it be software, data, hardware, or anything else. see also the wikipedia entry on open data here - http://en.wikipedia.org/wiki/Open_data Cheers, Michael On 2013-09-04, at 10:18 AM, James McKinney <[hidden email]> wrote: > (Cutting out all the other lists so that we don't cross-post unnecessarily.) > > Actually, Immanuel was quoting Peder :) But Stephane did agree with Peder with: > >> +1 Although I am a proponent of open source software, OS/FLOSS software cannot be tied to strongly with open data (or open gov). FLOSS software can enabled some benefits linked to open data, for example better collaboration around data management and software development. But it is not and should not be presented as a prerequisite. Many gov agencies are used to work with some software provider and those agencies and their team should not feel under attack when open data arrives. > > Anyway, you can clearly have open data without open source. As mentioned, many governments use proprietary platforms like Socrata to manage and distribute open data, and even more governments use in-house proprietary code to do the same. A lot of open data currently published is created by proprietary software (ESRI ArcGIS) and stored in a proprietary format (ESRI Shapefile). > > The definition of open data does not set any requirements with respect to the licensing of the *software* used to create, manage and distribute the open data. The definition only sets requirements with respect to the licensing of the *data*. > > And you can clearly analyze open data and open source separately. Open source (and free software) have been around a lot longer as established concepts than open data has, and people have had no trouble analyzing open source in isolation for all that time. > > Anyway, no one is arguing against open source. People have simply pointed out that there is a distinction between the two concepts. > > James > > On 2013-09-04, at 8:30 AM, Tracey P. Lauriault wrote: > >> Immanuel; >> >> What Stephane is saying, is that in the short term, as cities/Provs/Terr/Feds are transitioning to open data strategies, they need to cobble up what they have on hand to get data out. Be that open source or not. Once they get established a little, and do some organizational learning they can move to other platforms which in the best possible world, OS. >> >> I would argue that many of the current portals, even currently used open data OS ones, do not scale well when tens of thousands of datasets are in the systems. The TBS open data portal is an great example. Searching data with tags is by no means the best way, and common keyword vocabularies need to be used in order to actually find stuff. This is where the geomatics folks, and the librarians come into play, they have been doing catalogs and portals for a good long time. The Ottawa Public Library is an example of an excellent interoperable multi institutional catalog system. Just need to think of data as books! >> >> Finally, this list has always been one where we communicate diplomatically. Your last post was, somewhat not of that nature. Breathe first. Absolutism should be tempered with the grey zones of institutional change that we are witnessing in our institutions. We will have to be patient or we will lose them. >> >> Cheers >> t >> >> >> On Wed, Sep 4, 2013 at 12:54 PM, Immanuel Giulea <[hidden email]> wrote: >> You don't need open source to have open data. One the most successful open data platforms on the planet is not built with open source software (Socrata). And most of the worlds best open data is stored in a decidedly non-open data data store which cannot be easily replaced (Oracle), nor should it. >> -1 >> >> This is so wrong, and on so many levels!! >> >> You cannot have open data without open source. >> Socrata is not the best platform, it is being replaced by CKAN. >> Oracle is being abandonned to the favor of MariaDB and Postgres and other open database solutions. >> >> >> As a subject matter, Open Source software is a very different realm than open data, something that people surprisingly get mixed up together and lump into one because they share is the word "Open". It's also something that would be difficult to write about with any depth unless you have actually worked in software development; it's perhaps the only way to separate the hype from reality….and there is a lot of hype to be sure. >> >> >> -1 again >> And wrong on so many levels again. Open data and open source are intimately linked and should be analysed together. >> You don't need an degree in software development to analyse the adoption levels of FLOSS across different countries, levels of government and industries. >> >> >> No open data without open source! >> >> Just my 2 cents >> >> >> Immanuel >> >> >> >> On Mon, Sep 2, 2013 at 4:58 PM, Peder Jakobsen <[hidden email]> wrote: >> >> On 2013-09-01, at 7:55 AM, Immanuel Giulea <[hidden email]> wrote: >> >>> I would include the use of Free Libre and Open Source Software because open data without the right tools is not completely open data. >> >> You don't need open source to have open data. One the most successful open data platforms on the planet is not built with open source software (Socrata). And most of the worlds best open data is stored in a decidedly non-open data data store which cannot be easily replaced (Oracle), nor should it. >> >> As a subject matter, Open Source software is a very different realm than open data, something that people surprisingly get mixed up together and lump into one because they share is the word "Open". It's also something that would be difficult to write about with any depth unless you have actually worked in software development; it's perhaps the only way to separate the hype from reality….and there is a lot of hype to be sure. >> >> I love open data for a thesis, I would just extend it beyond the Canadian border, otherwise the subject matter seems too small. >> >> Peder >> >> >> >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >> >> >> >> -- >> Tracey P. Lauriault >> http://traceyplauriault.wordpress.com/2013/07/23/moving-to-ireland/ >> https://gcrc.carleton.ca/confluence/display/GCRCWEB/Lauriault >> http://datalibre.ca/ >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss > > _______________________________________________ > CivicAccess-discuss mailing list > [hidden email] > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss -------------------------------------------------------------------------- Michael Roberts -- Acclar Open Aid Data Skype: mroberts_112 _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
In reply to this post by James McKinney-2
Dear friends, chers amis,
Please see the Open Definition (both for Open Data and Open Content) adopted by Open Knowledge foundation Network In English : http://opendefinition.org/okd/ En français : http://opendefinition.org/okd/francais/ « Quand dire c'est faire » (Austin, 1970) « How to do things with Words » (Austin, 1955) Cordialement, Regards, P.-S. Les envois multiples (crosspostings) sont souvent nécessaires en curation, car une personne est libre de faire partie de plusieurs réseaux ou média sociaux. À chacun de gérer ces propres abonnements et non aux expéditeurs de limiter leur force de liaison/curation. Suggestions for automatic translation tool : <a onclick="openLink(event);" href="http://www.translate.ua">http://www.translate.ua <a onclick="openLink(event);" href="http://igorgladkov.com/extensions/translator.html">http://igorgladkov.com/extensions/translator.html <a onclick="openLink(event);" href="https://translate.google.com">https://translate.google.com --- Liaison par | Curation by --- Dre Diane Mercier Ambassadrice de l'Open Knowledge Foundation - Groupe local au Canada Web : http://ca.okfn.org Discussion : http://lists.okfn.org/mailman/listinfo/okfn-ca Medias : @okfnca | LinkedIn : dianemercier | JiTsi - Jabber/XMPP: [hidden email] Blogue : http://dianemercier.com Portail : http://donnees.ville.montreal.qc.ca Le 2013-09-04 10:18, James McKinney a écrit :
_______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
In reply to this post by Michael Roberts
I don't see any need to link open source software with open data - indeed, mixing the two simply creates confusion, something that does not help advance the cause of either open data or open source. One is about a license applied to data, the other about a license applied to code.
While I believe both are "good" I'm still looking for the technical or even policy reason to conflate them in this thread. Just like open source code run on proprietary hardware is still open, open data (that is data licensed in a way the meets the open definition) can be used for commercial means, for non-profit means and/or run, created and shared on proprietary software and it will be no less open. If advocates of open source software want to advance their cause in government (which I'm broadly in favour of) that's great, but as an open data advocate, I get uncomfortable when open data is held hostage to that cause. Perhaps more simply put does open source software that runs proprietary data suddenly not become open? Of course not. I understand that some people are ideologically in favour of all things being open. I understand if people want to argue that a CKAN implementation may be technically better (I'm not claiming this), or a better policy choice for the government, but this thread started with a discussion about how you can't have open data without open source, this is patently not the case. And again, I get concerned as an advocate since claiming that data shared via a Socrata portal is somehow less open than that shared via a CKAN run portal in my mind does a disservice to the cause of open data and can only serve to confuse public servants (and the public) trying to wrap their head around these concepts for the first time. David Eaves [hidden email] www.eaves.ca On 2013-09-04, at 7:27 AM, Michael Roberts <[hidden email]> wrote: > Hi Peder > >>> I would include the use of Free Libre and Open Source Software because open data without the right tools is not completely open data. >> >> You don't need open source to have open data. One the most successful open data platforms on the planet is not built with open source software (Socrata). And most of the worlds best open data is stored in a decidedly non-open data data store which cannot be easily replaced (Oracle), nor should it. >> >> As a subject matter, Open Source software is a very different realm than open data, something that people surprisingly get mixed up together and lump into one because they share is the word "Open". It's also something that would be difficult to write about with any depth unless you have actually worked in software development; it's perhaps the only way to separate the hype from reality….and there is a lot of hype to be sure. > > You are right you don't need open source software to have open data technically speaking and Socrata is indeed a very successful company. That said, the common thread, I feel, is more of a belief in creating a 'commons' of data, software or whatever else, free of restrictive licenses that allow anyone to build upon, examine and reuse. If you are a software vendor like Oracle, that may not matter to you, as what's important is building a great product to serve a need, but for someone else, those principles of openness count whether it be software, data, hardware, or anything else. > > see also the wikipedia entry on open data here - http://en.wikipedia.org/wiki/Open_data > > Cheers, > Michael > > On 2013-09-04, at 10:18 AM, James McKinney <[hidden email]> wrote: > >> (Cutting out all the other lists so that we don't cross-post unnecessarily.) >> >> Actually, Immanuel was quoting Peder :) But Stephane did agree with Peder with: >> >>> +1 Although I am a proponent of open source software, OS/FLOSS software cannot be tied to strongly with open data (or open gov). FLOSS software can enabled some benefits linked to open data, for example better collaboration around data management and software development. But it is not and should not be presented as a prerequisite. Many gov agencies are used to work with some software provider and those agencies and their team should not feel under attack when open data arrives. >> >> Anyway, you can clearly have open data without open source. As mentioned, many governments use proprietary platforms like Socrata to manage and distribute open data, and even more governments use in-house proprietary code to do the same. A lot of open data currently published is created by proprietary software (ESRI ArcGIS) and stored in a proprietary format (ESRI Shapefile). >> >> The definition of open data does not set any requirements with respect to the licensing of the *software* used to create, manage and distribute the open data. The definition only sets requirements with respect to the licensing of the *data*. >> >> And you can clearly analyze open data and open source separately. Open source (and free software) have been around a lot longer as established concepts than open data has, and people have had no trouble analyzing open source in isolation for all that time. >> >> Anyway, no one is arguing against open source. People have simply pointed out that there is a distinction between the two concepts. >> >> James >> >> On 2013-09-04, at 8:30 AM, Tracey P. Lauriault wrote: >> >>> Immanuel; >>> >>> What Stephane is saying, is that in the short term, as cities/Provs/Terr/Feds are transitioning to open data strategies, they need to cobble up what they have on hand to get data out. Be that open source or not. Once they get established a little, and do some organizational learning they can move to other platforms which in the best possible world, OS. >>> >>> I would argue that many of the current portals, even currently used open data OS ones, do not scale well when tens of thousands of datasets are in the systems. The TBS open data portal is an great example. Searching data with tags is by no means the best way, and common keyword vocabularies need to be used in order to actually find stuff. This is where the geomatics folks, and the librarians come into play, they have been doing catalogs and portals for a good long time. The Ottawa Public Library is an example of an excellent interoperable multi institutional catalog system. Just need to think of data as books! >>> >>> Finally, this list has always been one where we communicate diplomatically. Your last post was, somewhat not of that nature. Breathe first. Absolutism should be tempered with the grey zones of institutional change that we are witnessing in our institutions. We will have to be patient or we will lose them. >>> >>> Cheers >>> t >>> >>> >>> On Wed, Sep 4, 2013 at 12:54 PM, Immanuel Giulea <[hidden email]> wrote: >>> You don't need open source to have open data. One the most successful open data platforms on the planet is not built with open source software (Socrata). And most of the worlds best open data is stored in a decidedly non-open data data store which cannot be easily replaced (Oracle), nor should it. >>> -1 >>> >>> This is so wrong, and on so many levels!! >>> >>> You cannot have open data without open source. >>> Socrata is not the best platform, it is being replaced by CKAN. >>> Oracle is being abandonned to the favor of MariaDB and Postgres and other open database solutions. >>> >>> >>> As a subject matter, Open Source software is a very different realm than open data, something that people surprisingly get mixed up together and lump into one because they share is the word "Open". It's also something that would be difficult to write about with any depth unless you have actually worked in software development; it's perhaps the only way to separate the hype from reality….and there is a lot of hype to be sure. >>> >>> >>> -1 again >>> And wrong on so many levels again. Open data and open source are intimately linked and should be analysed together. >>> You don't need an degree in software development to analyse the adoption levels of FLOSS across different countries, levels of government and industries. >>> >>> >>> No open data without open source! >>> >>> Just my 2 cents >>> >>> >>> Immanuel >>> >>> >>> >>> On Mon, Sep 2, 2013 at 4:58 PM, Peder Jakobsen <[hidden email]> wrote: >>> >>> On 2013-09-01, at 7:55 AM, Immanuel Giulea <[hidden email]> wrote: >>> >>>> I would include the use of Free Libre and Open Source Software because open data without the right tools is not completely open data. >>> >>> You don't need open source to have open data. One the most successful open data platforms on the planet is not built with open source software (Socrata). And most of the worlds best open data is stored in a decidedly non-open data data store which cannot be easily replaced (Oracle), nor should it. >>> >>> As a subject matter, Open Source software is a very different realm than open data, something that people surprisingly get mixed up together and lump into one because they share is the word "Open". It's also something that would be difficult to write about with any depth unless you have actually worked in software development; it's perhaps the only way to separate the hype from reality….and there is a lot of hype to be sure. >>> >>> I love open data for a thesis, I would just extend it beyond the Canadian border, otherwise the subject matter seems too small. >>> >>> Peder >>> >>> >>> >>> _______________________________________________ >>> CivicAccess-discuss mailing list >>> [hidden email] >>> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >>> >>> >>> >>> -- >>> Tracey P. Lauriault >>> http://traceyplauriault.wordpress.com/2013/07/23/moving-to-ireland/ >>> https://gcrc.carleton.ca/confluence/display/GCRCWEB/Lauriault >>> http://datalibre.ca/ >>> _______________________________________________ >>> CivicAccess-discuss mailing list >>> [hidden email] >>> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >> >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss > > -------------------------------------------------------------------------- > Michael Roberts -- Acclar Open Aid Data > Skype: mroberts_112 > > > > > > > _______________________________________________ > CivicAccess-discuss mailing list > [hidden email] > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
In reply to this post by dianemercier
Sorry I have been late to this conversation. I have been in the midst of preparing, packing, and driving across the country for the past week and a half. I have only just gotten my laptop and a secure net connection. I want to respond to a couple of things.
RE: Cities publish no good data. Peder made some valid criticisms of doing a narrow examination of Canadian municipal open data programs (thinking that I was just starting a PhD, and not finishing up an MA). He pointed out that, “The municipalities in Canada that I'm aware of publish such a small sliver of data that it's really moot to study now. In 5 years the timing for a study of just Open data might be more interesting. In other words, you could write the answer to your current thesis in one sentence: The current state is dismal, case closed”. While in general this is fairly accurate (if perhaps a bit cynical), I do think that there are some interesting things going on that I discovered when looking at the programs in more detail and speaking to those involved. First of all, there a great deal of variation in how different cities are approaching their programs and what kind/quality of data they are making available. For example the District of North Vancouver (http://www.geoweb.dnv.org/) built their open data program early (2009) and have focused exclusively on geographic databases (since it is run out of the GIS department). The data they publish is quite detailed, available in multiple formats (including open versions like KML and CSV for every dataset), and has a good metadata description that includes an accuracy and completeness score and a data history. Sure, there is nothing in there that is particularly groundbreaking in terms of government transparency (e.g. no public spending or procurement records), but they have laid a solid basis for adding that type of data in the future (the program is set to expand out of the GIS department in the future). As well, if you are interested in geographic infrastructure data, then this is a very good resource. Most cities are publishing this type of data. Transportation, infrastructure, cultural amenities, and planning/development data was by far the most common (~70% of all published data as of Nov 2012). However, this is to be expected because that type of data is often already available, it’s in easy to publish formats, and it’s politically safe. It makes sense that this is what gets published first. Some cities are moving beyond that now. Toronto, Ottawa, Montreal, Edmonton, and Vancouver all publish some interesting government data such as councilor voting records/attendance, budget data, lobbyist and public tenders data, and other small but valuable additions. Tracey already expounded a bit on this earlier, but these little steps are how things get started. Nobody wants to put their neck out on the line and publish something that could come back to bite them, so they’re cautious. I think it’s worthwhile to encourage this movement and not complain (too loudly) that they’re not acting fast enough. RE: Open data vs open source The OKF definition of open data is the one that I prefer, and it does include open source aspect in that the data be freely available for access and modification without undue costs. Publishing data in proprietary formats such as ESRI’s .shp or Autodesk’s .dwg format is very common and I think it severely limits the ability of people to access and modify the data. I understand why this is done though. Often the dataset exists and is available to be published, except that it is sitting in the format of the software used by the city. It can be time consuming and/or technically difficult to convert all of the data to open formats, so the realistic choice is to either publish in the format available or not publish at all. If the programs has budgets and staff, then this could be alleviated, but most programs are not well funded (if at all), and decide that other things have higher priority. It would be great if everything was available in open formats and published through OS platforms like CKAN, but like Laura nicely explained, this might not be reasonable to expect yet and hopefully it will become more common in the future.
Thanks for taking an interest! The full thesis is available at https://dl.dropboxusercontent.com/u/38342349/CA_Muni_OpenData_Initiatives_LiamCurrie_2013.pdf if anyone wants to look through it. It’s far too long and I need to chop out the useful info and put it into a separate document. I also have to clean up my data and provide it online as well. SOON! Liam Currie D309 Macintosh-Corry Hall, _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
Thanks for this great study, Liam. I found the quotes from interviewees particularly illuminating. The local data team at the Sunlight Foundation has also been taking a close look at it, kudos!
Somewhat off-topic, but for what it's worth, when cities convert Shapefile to KML or other open formats, information (in particular, attributes for each geographic feature in the shapefile) is often lost or truncated due to poor conversion software. Information is also often more difficult to access after a shapefile has been converted to KML, due to the fact that KML has four different ways of storing attributes and not all software has equal support for all of them. Although Shapefile is proprietary, it is supported by GDAL/OGR, a very popular open source GIS library used by many other tools like QuantumGIS (an open source alternative to ArcGIS) and PostGIS (an extension to the PostgreSQL relational database, itself an open source alternative to Oracle or Microsoft products). James On 2013-09-04, at 6:14 PM, Liam Currie wrote:
_______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
In reply to this post by David Eaves
On 13-09-04 04:58 PM, David Eaves wrote:
> While I believe both are "good" I'm still looking for the technical > or even policy reason to conflate them in this thread. While my own policy priority is FLOSS over data, I suspect the issue is that open interfaces (APIs, file formats, etc) is being conflated with the software itself being FLOSS. With that in mind, wouldn't people agree that the data is not open if the method to interface with the data is closed? -- Russell McOrmond, Internet Consultant: <http://www.flora.ca/> Please help us tell the Canadian Parliament to protect our property rights as owners of Information Technology. Sign the petition! http://l.c11.ca/ict "The government, lobbied by legacy copyright holders and hardware manufacturers, can pry my camcorder, computer, home theatre, or portable media player from my cold dead hands!" _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
Related: data that is _only_ behind an API (open or closed) and is not
available in bulk download is not Open: 1 - API often have daily hit limits making it de facto impossible to download the entire dataset 2 - And when the provider of the API goes away, the access to the data goes away For #2, more organizations should consider https://aws.amazon.com/publicdatasets/ https://aws.amazon.com/publicdatasets/#3 The data user must pay for access. Some may baulk at the cost associated with access, perhaps conflating free-as-in-beer with free-as-freedom. But it is a good backup for public data that do not have stable hosts (which in my opinion is most sites, despite claims). -Glen On Thu, Sep 5, 2013 at 12:45 PM, Russell McOrmond <[hidden email]> wrote: > On 13-09-04 04:58 PM, David Eaves wrote: >> While I believe both are "good" I'm still looking for the technical >> or even policy reason to conflate them in this thread. > > While my own policy priority is FLOSS over data, I suspect the issue > is that open interfaces (APIs, file formats, etc) is being conflated > with the software itself being FLOSS. > > With that in mind, wouldn't people agree that the data is not open if > the method to interface with the data is closed? > > -- > Russell McOrmond, Internet Consultant: <http://www.flora.ca/> > Please help us tell the Canadian Parliament to protect our property > rights as owners of Information Technology. Sign the petition! > http://l.c11.ca/ict > > "The government, lobbied by legacy copyright holders and hardware > manufacturers, can pry my camcorder, computer, home theatre, or > portable media player from my cold dead hands!" > _______________________________________________ > CivicAccess-discuss mailing list > [hidden email] > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss -- - http://zzzoot.blogspot.com/ - _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
The OKD includes "Absence of Technological Restriction" http://opendefinition.org/okd/
> The work must be provided in such a form that there are no technological obstacles to the performance of the above activities. This can be achieved by the provision of the work in an open data format, i.e. one whose specification is publicly and freely available and which places no restrictions monetary or otherwise upon its use. According to Wikipedia, ESRI Shapefile is a "(mostly) open specification", so ESRI Shapefile may pass the test. Russell McOrmond wrote: > With that in mind, wouldn't people agree that the data is not open if the method to interface with the data is closed? According to the OKD, what matters is the specification of the data format, not the software that implements that specification. Glen Newton wrote: > Related: data that is _only_ behind an API (open or closed) and is not available in bulk download is not Open: The OKD says "The work shall be available as a whole". Since APIs typically do not give access to a dataset as a whole, you might say that the dataset as a whole is not open data. On the other hand, each response from the API may be considered open data. > 2 - And when the provider of the API goes away, the access to the data goes away The OKD does not require that data be available in perpetuity. That is not a condition for open data. Archival is certainly a concern, but it's separate from the definition of open data. James On 2013-09-05, at 1:02 PM, Glen Newton wrote: > Related: data that is _only_ behind an API (open or closed) and is not > available in bulk download is not Open: > 1 - API often have daily hit limits making it de facto impossible to > download the entire dataset > 2 - And when the provider of the API goes away, the access to the > data goes away > > For #2, more organizations should consider > https://aws.amazon.com/publicdatasets/ > https://aws.amazon.com/publicdatasets/#3 > The data user must pay for access. > Some may baulk at the cost associated with access, perhaps conflating > free-as-in-beer with free-as-freedom. > But it is a good backup for public data that do not have stable hosts > (which in my opinion is most sites, despite claims). > > -Glen > > On Thu, Sep 5, 2013 at 12:45 PM, Russell McOrmond <[hidden email]> wrote: >> On 13-09-04 04:58 PM, David Eaves wrote: >>> While I believe both are "good" I'm still looking for the technical >>> or even policy reason to conflate them in this thread. >> >> While my own policy priority is FLOSS over data, I suspect the issue >> is that open interfaces (APIs, file formats, etc) is being conflated >> with the software itself being FLOSS. >> >> With that in mind, wouldn't people agree that the data is not open if >> the method to interface with the data is closed? >> >> -- >> Russell McOrmond, Internet Consultant: <http://www.flora.ca/> >> Please help us tell the Canadian Parliament to protect our property >> rights as owners of Information Technology. Sign the petition! >> http://l.c11.ca/ict >> >> "The government, lobbied by legacy copyright holders and hardware >> manufacturers, can pry my camcorder, computer, home theatre, or >> portable media player from my cold dead hands!" >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss > > > > -- > - > http://zzzoot.blogspot.com/ > - > _______________________________________________ > CivicAccess-discuss mailing list > [hidden email] > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
That is of course if we all agree that the OKD is the definition to follow. It is one of many, and compliance is always a tricky thing. On Thu, Sep 5, 2013 at 6:21 PM, James McKinney <[hidden email]> wrote: The OKD includes "Absence of Technological Restriction" http://opendefinition.org/okd/ -- _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
Are any of these other definitions defined on some website? If we're going to disagree on definitions, then we need to at least know what those other definitions are - otherwise we run a high risk of assuming too much about what's included in or excluded from each others' definitions.
The OKD is useful in that you can point to it, and if you disagree with it you can point to specific provisions that you disagree with. For example, I've pointed to a few provisions, and if Russell or Glen disagree, then we can have that conversation. The OKD helps bring a lot more clarity to that discussion. In any case, I think there's a huge benefit to having broad adoption of the same definition for open data. I don't think a pluralistic definition of open data is very helpful. "What is open data?" shouldn't be as contentious as "What is art?". On 2013-09-05, at 1:23 PM, Tracey P. Lauriault wrote:
_______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
My issue with the oknf one is it does not work well for science or research data or geomatics data for that matter. It misses interoperability, standards, provenance, archiving and while description is there, metadata are not explicit. The discussion about OS and OD here for instance challenges the OKNF one or points to different understandings.t On Thu, Sep 5, 2013 at 6:44 PM, James McKinney <[hidden email]> wrote:
-- _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
+10 Tracey
On the science side of open data - which is where I work - keeping data for perpetuity is the goal (in a perfect world). Minimal data loss is the realistic goal, and in particular data that cannot be recreated (lost LANDSAT data from the 70s is gone-gone; whereas the data from a lab experiment with a well defined methods/protocol and easy to obtain materials can be recreated). And you do not know how other people may (subsequently) use collected data, beyond the original creator's intent and often in a very different domain. like the previously reported "18th century ships' logs used for climate change research", http://www.wired-gov.net/wg/wg-news-1.nsf/0/3E832C26266DCDA580257647005766FA?OpenDocument and "Monks' diaries aid understanding of 500 years of climate change", http://www.bbc.co.uk/news/uk-scotland-edinburgh-east-fife-12205971 and my just discovered - "Ships’ Log-Books, Sea Ice and the Cold Summer of 1816 in Hudson Bay and Its Approaches", ARCTIC VOL. 38, NO. 2 (JUNE 1985) P. 121.128, http://arctic.synergiesprairies.ca/arctic/index.php/arctic/article/download/2121/2098 which interestingly pre-dates the climate change debate - "Hudson's Bay Company ship's logbooks: a source of far North Atlantic weather data", https://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8503228 [PAYWALLED] The OKNF position suggests a shallow/naive view of data and limited range of data use / use cases. Gotta go: we keep filling up our 62TB storage device here at my office with all our genomics data... :-) -Glen On Thu, Sep 5, 2013 at 2:03 PM, Tracey P. Lauriault <[hidden email]> wrote: > see a list here - http://datalibre.ca/links-resources/#Principles > > My issue with the oknf one is it does not work well for science or research > data or geomatics data for that matter. It misses interoperability, > standards, provenance, archiving and while description is there, metadata > are not explicit. > > The discussion about OS and OD here for instance challenges the OKNF one or > points to different understandings. > > OD has been around for way longer than the OKNF has, and many of the > practioners on this list have been engaged with the topic of open access to > public data for nearly 20 years. > > The new comers have adopted by habit, defacto or lack of exposure to the > historical discourse on open access to data issues the oknf definition, > which has traction among - new media/api/mobile dev.-app/ crowd - but by no > means has traction among scientists, librarians, archivists and researchers, > and that is why there is often a disconnect among understandings and > communities. > > It was one of the reasons why a few years ago when the oknf contacted us to > become a part of it somehow, we said that we liked our canadian flavour of > things, and the fact that we were doing well without them. > > Cheers > t > > > On Thu, Sep 5, 2013 at 6:44 PM, James McKinney <[hidden email]> wrote: >> >> Are any of these other definitions defined on some website? If we're going >> to disagree on definitions, then we need to at least know what those other >> definitions are - otherwise we run a high risk of assuming too much about >> what's included in or excluded from each others' definitions. >> >> The OKD is useful in that you can point to it, and if you disagree with it >> you can point to specific provisions that you disagree with. For example, >> I've pointed to a few provisions, and if Russell or Glen disagree, then we >> can have that conversation. The OKD helps bring a lot more clarity to that >> discussion. >> >> In any case, I think there's a huge benefit to having broad adoption of >> the same definition for open data. I don't think a pluralistic definition of >> open data is very helpful. "What is open data?" shouldn't be as contentious >> as "What is art?". >> >> >> On 2013-09-05, at 1:23 PM, Tracey P. Lauriault wrote: >> >> That is of course if we all agree that the OKD is the definition to >> follow. It is one of many, and compliance is always a tricky thing. >> >> >> On Thu, Sep 5, 2013 at 6:21 PM, James McKinney <[hidden email]> wrote: >>> >>> The OKD includes "Absence of Technological Restriction" >>> http://opendefinition.org/okd/ >>> >>> > The work must be provided in such a form that there are no >>> > technological obstacles to the performance of the above activities. This can >>> > be achieved by the provision of the work in an open data format, i.e. one >>> > whose specification is publicly and freely available and which places no >>> > restrictions monetary or otherwise upon its use. >>> >>> According to Wikipedia, ESRI Shapefile is a "(mostly) open >>> specification", so ESRI Shapefile may pass the test. >>> >>> Russell McOrmond wrote: >>> > With that in mind, wouldn't people agree that the data is not open if >>> > the method to interface with the data is closed? >>> >>> According to the OKD, what matters is the specification of the data >>> format, not the software that implements that specification. >>> >>> Glen Newton wrote: >>> > Related: data that is _only_ behind an API (open or closed) and is not >>> > available in bulk download is not Open: >>> >>> The OKD says "The work shall be available as a whole". Since APIs >>> typically do not give access to a dataset as a whole, you might say that the >>> dataset as a whole is not open data. On the other hand, each response from >>> the API may be considered open data. >>> >>> > 2 - And when the provider of the API goes away, the access to the data >>> > goes away >>> >>> The OKD does not require that data be available in perpetuity. That is >>> not a condition for open data. Archival is certainly a concern, but it's >>> separate from the definition of open data. >>> >>> James >>> >>> On 2013-09-05, at 1:02 PM, Glen Newton wrote: >>> >>> > Related: data that is _only_ behind an API (open or closed) and is not >>> > available in bulk download is not Open: >>> > 1 - API often have daily hit limits making it de facto impossible to >>> > download the entire dataset >>> > 2 - And when the provider of the API goes away, the access to the >>> > data goes away >>> > >>> > For #2, more organizations should consider >>> > https://aws.amazon.com/publicdatasets/ >>> > https://aws.amazon.com/publicdatasets/#3 >>> > The data user must pay for access. >>> > Some may baulk at the cost associated with access, perhaps conflating >>> > free-as-in-beer with free-as-freedom. >>> > But it is a good backup for public data that do not have stable hosts >>> > (which in my opinion is most sites, despite claims). >>> > >>> > -Glen >>> > >>> > On Thu, Sep 5, 2013 at 12:45 PM, Russell McOrmond <[hidden email]> >>> > wrote: >>> >> On 13-09-04 04:58 PM, David Eaves wrote: >>> >>> While I believe both are "good" I'm still looking for the technical >>> >>> or even policy reason to conflate them in this thread. >>> >> >>> >> While my own policy priority is FLOSS over data, I suspect the issue >>> >> is that open interfaces (APIs, file formats, etc) is being conflated >>> >> with the software itself being FLOSS. >>> >> >>> >> With that in mind, wouldn't people agree that the data is not open if >>> >> the method to interface with the data is closed? >>> >> >>> >> -- >>> >> Russell McOrmond, Internet Consultant: <http://www.flora.ca/> >>> >> Please help us tell the Canadian Parliament to protect our property >>> >> rights as owners of Information Technology. Sign the petition! >>> >> http://l.c11.ca/ict >>> >> >>> >> "The government, lobbied by legacy copyright holders and hardware >>> >> manufacturers, can pry my camcorder, computer, home theatre, or >>> >> portable media player from my cold dead hands!" >>> >> _______________________________________________ >>> >> CivicAccess-discuss mailing list >>> >> [hidden email] >>> >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >>> > >>> > >>> > >>> > -- >>> > - >>> > http://zzzoot.blogspot.com/ >>> > - >>> > _______________________________________________ >>> > CivicAccess-discuss mailing list >>> > [hidden email] >>> > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >>> >>> _______________________________________________ >>> CivicAccess-discuss mailing list >>> [hidden email] >>> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >> >> >> >> >> -- >> Tracey P. Lauriault >> http://traceyplauriault.wordpress.com/2013/07/23/moving-to-ireland/ >> https://gcrc.carleton.ca/confluence/display/GCRCWEB/Lauriault >> http://datalibre.ca/ >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >> >> >> >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss > > > > > -- > Tracey P. Lauriault > http://traceyplauriault.wordpress.com/2013/07/23/moving-to-ireland/ > https://gcrc.carleton.ca/confluence/display/GCRCWEB/Lauriault > http://datalibre.ca/ > > _______________________________________________ > CivicAccess-discuss mailing list > [hidden email] > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss -- - http://zzzoot.blogspot.com/ - _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
> - "Hudson's Bay Company ship's logbooks: a source of far North
> Atlantic weather data", > https://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8503228 > [PAYWALLED] Polar Record / Volume 48 / Issue 02 / April 2012, pp 165-176 On Thu, Sep 5, 2013 at 2:30 PM, Glen Newton <[hidden email]> wrote: > +10 Tracey > > On the science side of open data - which is where I work - keeping > data for perpetuity is the goal (in a perfect world). Minimal data > loss is the realistic goal, and in particular data that cannot be > recreated (lost LANDSAT data from the 70s is gone-gone; whereas the > data from a lab experiment with a well defined methods/protocol and > easy to obtain materials can be recreated). > > And you do not know how other people may (subsequently) use collected > data, beyond the original creator's intent and often in a very > different domain. like the previously reported "18th century ships' > logs used for climate change research", > http://www.wired-gov.net/wg/wg-news-1.nsf/0/3E832C26266DCDA580257647005766FA?OpenDocument > and "Monks' diaries aid understanding of 500 years of climate change", > http://www.bbc.co.uk/news/uk-scotland-edinburgh-east-fife-12205971 > and my just discovered > - "Ships’ Log-Books, Sea Ice and the Cold Summer of 1816 in Hudson Bay > and Its Approaches", ARCTIC VOL. 38, NO. 2 (JUNE 1985) P. 121.128, > http://arctic.synergiesprairies.ca/arctic/index.php/arctic/article/download/2121/2098 > which interestingly pre-dates the climate change debate > - "Hudson's Bay Company ship's logbooks: a source of far North > Atlantic weather data", > https://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8503228 > [PAYWALLED] > > > The OKNF position suggests a shallow/naive view of data and limited > range of data use / use cases. > > Gotta go: we keep filling up our 62TB storage device here at my office > with all our genomics data... :-) > > -Glen > > On Thu, Sep 5, 2013 at 2:03 PM, Tracey P. Lauriault <[hidden email]> wrote: >> see a list here - http://datalibre.ca/links-resources/#Principles >> >> My issue with the oknf one is it does not work well for science or research >> data or geomatics data for that matter. It misses interoperability, >> standards, provenance, archiving and while description is there, metadata >> are not explicit. >> >> The discussion about OS and OD here for instance challenges the OKNF one or >> points to different understandings. >> >> OD has been around for way longer than the OKNF has, and many of the >> practioners on this list have been engaged with the topic of open access to >> public data for nearly 20 years. >> >> The new comers have adopted by habit, defacto or lack of exposure to the >> historical discourse on open access to data issues the oknf definition, >> which has traction among - new media/api/mobile dev.-app/ crowd - but by no >> means has traction among scientists, librarians, archivists and researchers, >> and that is why there is often a disconnect among understandings and >> communities. >> >> It was one of the reasons why a few years ago when the oknf contacted us to >> become a part of it somehow, we said that we liked our canadian flavour of >> things, and the fact that we were doing well without them. >> >> Cheers >> t >> >> >> On Thu, Sep 5, 2013 at 6:44 PM, James McKinney <[hidden email]> wrote: >>> >>> Are any of these other definitions defined on some website? If we're going >>> to disagree on definitions, then we need to at least know what those other >>> definitions are - otherwise we run a high risk of assuming too much about >>> what's included in or excluded from each others' definitions. >>> >>> The OKD is useful in that you can point to it, and if you disagree with it >>> you can point to specific provisions that you disagree with. For example, >>> I've pointed to a few provisions, and if Russell or Glen disagree, then we >>> can have that conversation. The OKD helps bring a lot more clarity to that >>> discussion. >>> >>> In any case, I think there's a huge benefit to having broad adoption of >>> the same definition for open data. I don't think a pluralistic definition of >>> open data is very helpful. "What is open data?" shouldn't be as contentious >>> as "What is art?". >>> >>> >>> On 2013-09-05, at 1:23 PM, Tracey P. Lauriault wrote: >>> >>> That is of course if we all agree that the OKD is the definition to >>> follow. It is one of many, and compliance is always a tricky thing. >>> >>> >>> On Thu, Sep 5, 2013 at 6:21 PM, James McKinney <[hidden email]> wrote: >>>> >>>> The OKD includes "Absence of Technological Restriction" >>>> http://opendefinition.org/okd/ >>>> >>>> > The work must be provided in such a form that there are no >>>> > technological obstacles to the performance of the above activities. This can >>>> > be achieved by the provision of the work in an open data format, i.e. one >>>> > whose specification is publicly and freely available and which places no >>>> > restrictions monetary or otherwise upon its use. >>>> >>>> According to Wikipedia, ESRI Shapefile is a "(mostly) open >>>> specification", so ESRI Shapefile may pass the test. >>>> >>>> Russell McOrmond wrote: >>>> > With that in mind, wouldn't people agree that the data is not open if >>>> > the method to interface with the data is closed? >>>> >>>> According to the OKD, what matters is the specification of the data >>>> format, not the software that implements that specification. >>>> >>>> Glen Newton wrote: >>>> > Related: data that is _only_ behind an API (open or closed) and is not >>>> > available in bulk download is not Open: >>>> >>>> The OKD says "The work shall be available as a whole". Since APIs >>>> typically do not give access to a dataset as a whole, you might say that the >>>> dataset as a whole is not open data. On the other hand, each response from >>>> the API may be considered open data. >>>> >>>> > 2 - And when the provider of the API goes away, the access to the data >>>> > goes away >>>> >>>> The OKD does not require that data be available in perpetuity. That is >>>> not a condition for open data. Archival is certainly a concern, but it's >>>> separate from the definition of open data. >>>> >>>> James >>>> >>>> On 2013-09-05, at 1:02 PM, Glen Newton wrote: >>>> >>>> > Related: data that is _only_ behind an API (open or closed) and is not >>>> > available in bulk download is not Open: >>>> > 1 - API often have daily hit limits making it de facto impossible to >>>> > download the entire dataset >>>> > 2 - And when the provider of the API goes away, the access to the >>>> > data goes away >>>> > >>>> > For #2, more organizations should consider >>>> > https://aws.amazon.com/publicdatasets/ >>>> > https://aws.amazon.com/publicdatasets/#3 >>>> > The data user must pay for access. >>>> > Some may baulk at the cost associated with access, perhaps conflating >>>> > free-as-in-beer with free-as-freedom. >>>> > But it is a good backup for public data that do not have stable hosts >>>> > (which in my opinion is most sites, despite claims). >>>> > >>>> > -Glen >>>> > >>>> > On Thu, Sep 5, 2013 at 12:45 PM, Russell McOrmond <[hidden email]> >>>> > wrote: >>>> >> On 13-09-04 04:58 PM, David Eaves wrote: >>>> >>> While I believe both are "good" I'm still looking for the technical >>>> >>> or even policy reason to conflate them in this thread. >>>> >> >>>> >> While my own policy priority is FLOSS over data, I suspect the issue >>>> >> is that open interfaces (APIs, file formats, etc) is being conflated >>>> >> with the software itself being FLOSS. >>>> >> >>>> >> With that in mind, wouldn't people agree that the data is not open if >>>> >> the method to interface with the data is closed? >>>> >> >>>> >> -- >>>> >> Russell McOrmond, Internet Consultant: <http://www.flora.ca/> >>>> >> Please help us tell the Canadian Parliament to protect our property >>>> >> rights as owners of Information Technology. Sign the petition! >>>> >> http://l.c11.ca/ict >>>> >> >>>> >> "The government, lobbied by legacy copyright holders and hardware >>>> >> manufacturers, can pry my camcorder, computer, home theatre, or >>>> >> portable media player from my cold dead hands!" >>>> >> _______________________________________________ >>>> >> CivicAccess-discuss mailing list >>>> >> [hidden email] >>>> >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >>>> > >>>> > >>>> > >>>> > -- >>>> > - >>>> > http://zzzoot.blogspot.com/ >>>> > - >>>> > _______________________________________________ >>>> > CivicAccess-discuss mailing list >>>> > [hidden email] >>>> > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >>>> >>>> _______________________________________________ >>>> CivicAccess-discuss mailing list >>>> [hidden email] >>>> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >>> >>> >>> >>> >>> -- >>> Tracey P. Lauriault >>> http://traceyplauriault.wordpress.com/2013/07/23/moving-to-ireland/ >>> https://gcrc.carleton.ca/confluence/display/GCRCWEB/Lauriault >>> http://datalibre.ca/ >>> _______________________________________________ >>> CivicAccess-discuss mailing list >>> [hidden email] >>> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >>> >>> >>> >>> _______________________________________________ >>> CivicAccess-discuss mailing list >>> [hidden email] >>> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >> >> >> >> >> -- >> Tracey P. Lauriault >> http://traceyplauriault.wordpress.com/2013/07/23/moving-to-ireland/ >> https://gcrc.carleton.ca/confluence/display/GCRCWEB/Lauriault >> http://datalibre.ca/ >> >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss > > > > -- > - > http://zzzoot.blogspot.com/ > - -- - http://zzzoot.blogspot.com/ - _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
In reply to this post by Tracey P. Lauriault
I think what is actually being disagreed upon is the goal of the definitions, as opposed to the definitions themselves. The OKD is based on the Open Source Definition (http://opensource.org/osd-annotated). The OSD similarly has no criteria in terms of metadata for software, provenance of that software, interoperability between software, standards for software, etc. All those things are important, but they don't have anything to do with what the OSD is trying to achieve, which is to set a standard for what rights users ought to have to software. The goal of the OKD is similarly to set a standard for what rights users ought to have to data. I think the OKD's goal is well scoped. If you set too big a goal, it becomes incredibly hard to get adoption and compliance. That's why I like approaches like the W3C "5-star" http://5stardata.info/ Like people on this list, the W3C has things that it would like to see in open data (specifically, linked data), and it came up with an approach where open data gets an additional star for each target it achieves. I believe a similar approach with respect to achieving specific targets for interoperability, etc. would be a more fruitful strategy than attempting to increase adoption of a very demanding definition of open data, that contains the laundry list of things we want data to be. I might add that not all the principles in that list on DataLibre are intended to be interpreted as definitions. For example, the IPC's Access by Design is a document describing rules of beliefs for governing one's behavior in a way that achieves access goals. Indeed, several of the principles have more to do with the publisher than with the data being published. It is also written specifically for government. Same goes for OECD document. I think the definition of open data and the principles that government ought to follow with respect to data are two separate conversations. James On 2013-09-05, at 2:03 PM, Tracey P. Lauriault wrote:
_______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
Re: [Okfn-ca] The Role of Canadian Municipal Open Data: A Multi-city Evaluation | Currie, Liam (MA Thesis)
|
|
In reply to this post by Tracey P. Lauriault
Just to pitch in to this discussion - the OpenDefinition is supposed to do one thing: define the 'openness' in open data. That doesn't mean it is supposed to be a data standard for any data that might be open. In particular it provides principles for the baseline legal and technical requirements (machine readable, legally reusable without restriction, downloadable in bulk, etc). It isn't supposed to be a data standard to end all data standards. Nor is it claiming that people haven't been putting data on the web (and sharing data more generally) for decades - but people sharing data which is explicitly marked as open with legal tools like CC0 is a fairly recent development. [1]
As a legal definition it actually is widely considered to be one of the most influential definitions for the openness of scientific data and has been widely cited by many organisations, funding bodies, policy documents and so on, along with the Panton Principles (which won an award from SPARC). [2] We also host Panton Fellowships to encourage the dissemination of these principles.
In terms of geodata, we do have a license called the Open Database License (ODbL), part of a project called Open Data Commons, which is used by Open Street Map and for many other geospatial datasets. [3] The Open Definition was a crucial part of discussions that went into the creation of this license, so perhaps geospatial data experts who work with open data are gradually going to be more likely to encounter the OpenDefinition in some shape or form.
With our various local groups and working groups we're working hard to increase awareness of best practises for making data open (put it online in machine readable format, use a suitable legal tool to explicitly mark it as open) in a variety of different fields [4] - but it is critical that this is *not* the only consideration to take into account for data to be good data. There are plenty of other data standards and best practises out there that are crucial in different fields. We mainly (but not exclusively) focus on the open bit and that's the bit that the Open Definition is supposed to help with.
J. [1] with the possible exception of data which is exempt from copyright and other rights, like US federal data On 5 September 2013 20:03, Tracey P. Lauriault <[hidden email]> wrote:
Jonathan Gray Director of Policy and Ideas | @jwyg Empowering through Open Knowledge okfn.org | @okfn | OKF on Facebook | Blog | Newsletter _______________________________________________ CivicAccess-discuss mailing list [hidden email] http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss |
«
Return to CivicAccess
|
1 view|%1 views
| Free forum by Nabble | Edit this page |