Re: New LOD dataset for media types

[Graham Klyne]

Hi Silvio,

Nice work: this could be handy.  I have a couple of questions/comments:

1. I assume the machine formats are accessible by content negotiation?  It might 
be handy to also include direct links on the web page.  For example, I found I 
could change the .html suffix to .ttl or .json to get alternative 
representations, but that required guesswork on my part.


2. Are there any plans (and resources) in place to ensure longevity of the 
sparontologies.net domain?


3. I note that the HTML page indicates cc-by licensing of the content, but there 
is no such information in the machine readable formats.


#g
--


On 04/11/2015 23:21, Silvio Peroni wrote:

Dear friends,

We are pleased to announce a new LOD dataset

http://www.sparontologies.net/mediatype

that makes available media types defined as proper resources in RDF, according 
to the SPAR Ontologies [1] and DCTerms [2].

A media type is an identifier (for example "text/turtle") for file formats on the Internet composed 
by two parts: a registry ("text" in the example) and a record ("turtle" in the example). 
They are handled by the Internet Assigned Numbers Authority (IANA), which is the official authority for the 
standardisation and publication of these classifications.

The aforementioned web space, part of the SPAR Ontologies website, has been reserved for providing the RDF 
representation of each media type defined by IANA [3]. In particular, a media type is accessible by 
concatenating the URL "http://www.sparontologies.net/mediatype/; with its related identifier. For 
instance, "http://www.sparontologies.net/mediatype/text/turtle; allows one to access the specific 
resource for the "text/turtle" media type.

Each media type can be accompanied by agents who acted as contributors, the 
related RFC documents documenting it, its current status (either official, 
deprecated or obsoleted), and direct links to Wikipedia pages and DBpedia 
resources related with the media type.

All these resources defining media types, thus, can be used for specifying 
particular formats (e.g., by means of the DCTerms property dcterms:format) that 
a certain entity, such as a book or a dataset, can have.

Please do not hesitate to contact us (sparontolog...@gmail.com) for questions 
and additional information about this LOD dataset.

Have a nice day :-)

S.



1. http://www.sparontologies.net
2. http://dublincore.org/documents/dcmi-terms/
3. http://www.iana.org/assignments/media-types/media-types.xml



Silvio Peroni, Ph.D.
Department of Computer Science and Engineering
University of Bologna, Bologna (Italy)
Tel: +39 051 2094871
E-mail: silvio.per...@unibo.it
Web: http://www.essepuntato.it
Twitter: essepuntato

Re: New LOD dataset for media types

[Haag, Jason]

Silvio,

Thanks for this work on ontologies for media types. Is there any overlap
here with https://schema.org/MediaObject ?

BTW, the W3C has a really nice process in place for IRI persistence.
Perhaps establish something like https://w3id.org/sparontologies/mediatype/
and redirect for your server. I believe you could have stronger adoption
and use of your ontology if you adopt a good IRI design and persistence
strategy up front.

There are several nice things about the W3C permanent identifier approach.
For example:

1) longevity of the IRI (if sparontologies.net ceased to exist for whatever
reason, the ontologies could be maintained elsewhere by updating the w3id
target URL/redirect.

2) You don't have to worry about configuring a server for content
negotiation. The w3id.org server is already set up to handle content neg
using apache.

3) It uses github to sync with the w3id.org server and has a strong
commitment from several folks from the W3C and other organizations heavily
invested in the web. You simply add your .htaccess files and submit a pull
request. PURL.org is current disabled and not allowing even existing users
to login and manage their PURLs. The interface and approach for PURL.org is
cumbersome.


Regards,

J Haag




---
Advanced Distributed Learning Initiative
+1.850.266.7100(office)
+1.850.471.1300 (mobile)
jhaag75 (skype)
http://linkedin.com/in/jasonhaag

On Thu, Nov 5, 2015 at 7:37 AM, Silvio Peroni 
wrote:

> Hi Graham,
>
> Nice work: this could be handy.
>
>
> Thanks :-)
>
> 1. I assume the machine formats are accessible by content negotiation?
>
>
> They are accessible by content negotiation indeed. In particular, one can
> use:
> - "application/rdf+xml" for having the data in RDF/XML;
> - "text/turtle” or "text/n3" for having the data in Turtle;
> - "application/ld+json" or "application/json" for having the data in
> JSON-LD;
> - any other kind of request will result in having the data in HTML.
>
> E.g.:
>
> > curl -L -H "Accept: application/rdf+xml"
> http://www.sparontologies.net/mediatype/text/plain
>
> will return the RDF/XML representation of such media type.
>
> It might be handy to also include direct links on the web page.  For
> example, I found I could change the .html suffix to .ttl or .json to get
> alternative representations, but that required guesswork on my part.
>
>
> That’s right, you can access directly to the related formats by adding
> .rdf, .ttl, .json and .html respectively. I’ll add a note in the homepage
> of the dataset and in each HTML page of each mime type soon.
>
> 2. Are there any plans (and resources) in place to ensure longevity of the
> sparontologies.net domain?
>
>
> Well, actually now the sparontologies.net website is handled by me, with
> personal resources. I would like to keep it available without external
> funds, but in case it will be more expensive to maintain I think I can ask
> for support to the University of Bologna, that’s the plan at least. Do you
> have something different in mind?
>
> 3. I note that the HTML page indicates cc-by licensing of the content, but
> there is no such information in the machine readable formats.
>
>
> That’s right, I have to add these data into the RDF representation as well
> – in my todo list.
>
> Have a nice day :-)
>
> S.
>
>
>
>
>
> 
> Silvio Peroni, Ph.D.
> Department of Computer Science and Engineering
> University of Bologna, Bologna (Italy)
> Tel: +39 051 2094871
> E-mail: silvio.per...@unibo.it
> Web: http://www.essepuntato.it
> Twitter: essepuntato
>
>

Re: New LOD dataset for media types

[Graham Klyne]

On 05/11/2015 16:35, Silvio Peroni wrote:

>Perhaps establish something likehttps://w3id.org/sparontologies/mediatype/  
  and redirect for your server. I 
believe you could have stronger adoption and use of your ontology if you adopt a good 
IRI design and persistence strategy up front.

That’s a good point, and I’ve just finished to update everything, since it was 
possible in a relatively short time:-)

In particular, I’ve just made a request to w3id.org  for a 
space, and I’ve updated le URLs of all the entities. Now they are accessible using the



Hi Silvio,

Responding to your earlier question, this kind of community-underwritten hosting 
is the sort of thing I was contemplating.


Thanks!

#g
--

Re: New LOD dataset for media types

[Haag, Jason]

FYI: The process for using the W3C's permanent identifier/PURL service is
here in the readme: https://github.com/perma-id/w3id.org

As Jeff mentioned PURL.org still works in read-only mode, you just can't
login and edit or add new PURLS right now.

---
Advanced Distributed Learning Initiative
+1.850.266.7100(office)
+1.850.471.1300 (mobile)
jhaag75 (skype)
http://linkedin.com/in/jasonhaag

On Thu, Nov 5, 2015 at 9:42 AM, Daniel Garijo  wrote:

>
>
> 2015-11-05 16:22 GMT+01:00 Michael Brunnbauer :
>
>>
>> Hello Jason,
>>
>> On Thu, Nov 05, 2015 at 08:32:22AM -0600, Haag, Jason wrote:
>> > PURL.org is current disabled and not allowing even existing users
>> > to login and manage their PURLs.
>>
>> Does anybody have an idea how long the login on purl.org has been
>> disabled?
>>
> Around a week (since 27 October).
> This is the response provided in the public mailing list for persistent
> urls :
> "Sorry for the inconvenience. The SOLR index on the purl.org site has
> stopped updating, which prevents effective maintenance. The login mechanism
> is turned off until a solution is discovered. The timeframe for that isn't
> clear yet."
> Best,
> Daniel
>
>
>>
>> Should users of purl.org URIs start to think about alternatives?
>>
>> Regards,
>>
>> Michael Brunnbauer
>>
>> --
>> ++  Michael Brunnbauer
>> ++  netEstate GmbH
>> ++  Geisenhausener Straße 11a
>> ++  81379 München
>> ++  Tel +49 89 32 19 77 80
>> ++  Fax +49 89 32 19 77 89
>> ++  E-Mail bru...@netestate.de
>> ++  http://www.netestate.de/
>> ++
>> ++  Sitz: München, HRB Nr.142452 (Handelsregister B München)
>> ++  USt-IdNr. DE221033342
>> ++  Geschäftsführer: Michael Brunnbauer, Franz Brunnbauer
>> ++  Prokurist: Dipl. Kfm. (Univ.) Markus Hendel
>>
>
>

Re: New LOD dataset for media types

[Silvio Peroni]

Hi Jason,

> Thanks for this work on ontologies for media types. Is there any overlap here 
> with https://schema.org/MediaObject  ? 

Well, it’s not an overlap, is more a complement. The media types described in 
such dataset (class dcterms:MediaType) are possible formats that media objects 
may have. This can be specified by means of dcterms:format property, for 
instance:

:my-video a schema:MediaObject ;
dcterms:format  .

> BTW, the W3C has a really nice process in place for IRI persistence.

I know it and I’ve already used it indeed for another project.

> Perhaps establish something like https://w3id.org/sparontologies/mediatype/ 
>  and redirect for your server. I 
> believe you could have stronger adoption and use of your ontology if you 
> adopt a good IRI design and persistence strategy up front.

That’s a good point, and I’ve just finished to update everything, since it was 
possible in a relatively short time :-) 

In particular, I’ve just made a request to w3id.org  for a 
space, and I’ve updated le URLs of all the entities. Now they are accessible 
using the

https://w3id.org/spar/mediatype/[mediatype] 
 

e.g.

https://w3id.org/spar/mediatype/text/turtle 


Does it sound good?

Jason, Graham, thanks again for all your suggestions and support!
Have a nice day :-)

S.




Silvio Peroni, Ph.D.
Department of Computer Science and Engineering
University of Bologna, Bologna (Italy)
Tel: +39 051 2094871
E-mail: silvio.per...@unibo.it
Web: http://www.essepuntato.it
Twitter: essepuntato

RE: New LOD dataset for media types

[Young,Jeff (OR)]

The SOLR indexing on purl.org has been misbehaving for a couple of months. 
Login has been disabled for a few weeks to minimize the damage. The service is 
still running, but in read-only mode. There is no estimate for when this will 
be resolved.

One alternative is to use the http://w3id.org/ service, which is run by the W3C 
Permanent Identifier Community Group. Their mechanism is simpler and has a 
broader base of support.

Jeff

> -Original Message-
> From: Michael Brunnbauer [mailto:bru...@netestate.de]
> Sent: Thursday, November 05, 2015 10:22 AM
> To: Haag, Jason
> Cc: Linking Open Data Mailing List Data; Semantic Web Mailing List
> Subject: Re: New LOD dataset for media types
> 
> 
> Hello Jason,
> 
> On Thu, Nov 05, 2015 at 08:32:22AM -0600, Haag, Jason wrote:
> > PURL.org is current disabled and not allowing even existing users to
> > login and manage their PURLs.
> 
> Does anybody have an idea how long the login on purl.org has been disabled?
> 
> Should users of purl.org URIs start to think about alternatives?
> 
> Regards,
> 
> Michael Brunnbauer
> 
> --
> ++  Michael Brunnbauer
> ++  netEstate GmbH
> ++  Geisenhausener Straße 11a
> ++  81379 München
> ++  Tel +49 89 32 19 77 80
> ++  Fax +49 89 32 19 77 89
> ++  E-Mail bru...@netestate.de
> ++  http://www.netestate.de/
> ++
> ++  Sitz: München, HRB Nr.142452 (Handelsregister B München)  USt-IdNr.
> ++ DE221033342
> ++  Geschäftsführer: Michael Brunnbauer, Franz Brunnbauer
> ++  Prokurist: Dipl. Kfm. (Univ.) Markus Hendel

Re: New LOD dataset for media types

[Daniel Garijo]

2015-11-05 16:22 GMT+01:00 Michael Brunnbauer :

>
> Hello Jason,
>
> On Thu, Nov 05, 2015 at 08:32:22AM -0600, Haag, Jason wrote:
> > PURL.org is current disabled and not allowing even existing users
> > to login and manage their PURLs.
>
> Does anybody have an idea how long the login on purl.org has been
> disabled?
>
Around a week (since 27 October).
This is the response provided in the public mailing list for persistent urls
:
"Sorry for the inconvenience. The SOLR index on the purl.org site has
stopped updating, which prevents effective maintenance. The login mechanism
is turned off until a solution is discovered. The timeframe for that isn't
clear yet."
Best,
Daniel


>
> Should users of purl.org URIs start to think about alternatives?
>
> Regards,
>
> Michael Brunnbauer
>
> --
> ++  Michael Brunnbauer
> ++  netEstate GmbH
> ++  Geisenhausener Straße 11a
> ++  81379 München
> ++  Tel +49 89 32 19 77 80
> ++  Fax +49 89 32 19 77 89
> ++  E-Mail bru...@netestate.de
> ++  http://www.netestate.de/
> ++
> ++  Sitz: München, HRB Nr.142452 (Handelsregister B München)
> ++  USt-IdNr. DE221033342
> ++  Geschäftsführer: Michael Brunnbauer, Franz Brunnbauer
> ++  Prokurist: Dipl. Kfm. (Univ.) Markus Hendel
>

Re: New LOD dataset for media types

[Michael Brunnbauer]

Hello Jason,

On Thu, Nov 05, 2015 at 08:32:22AM -0600, Haag, Jason wrote:
> PURL.org is current disabled and not allowing even existing users
> to login and manage their PURLs.

Does anybody have an idea how long the login on purl.org has been disabled?

Should users of purl.org URIs start to think about alternatives?

Regards,

Michael Brunnbauer

-- 
++  Michael Brunnbauer
++  netEstate GmbH
++  Geisenhausener Straße 11a
++  81379 München
++  Tel +49 89 32 19 77 80
++  Fax +49 89 32 19 77 89 
++  E-Mail bru...@netestate.de
++  http://www.netestate.de/
++
++  Sitz: München, HRB Nr.142452 (Handelsregister B München)
++  USt-IdNr. DE221033342
++  Geschäftsführer: Michael Brunnbauer, Franz Brunnbauer
++  Prokurist: Dipl. Kfm. (Univ.) Markus Hendel


signature.asc
Description: PGP signature

Re: New LOD dataset for media types

[Silvio Peroni]

Hi Graham,

> Nice work: this could be handy.  

Thanks :-)

> 1. I assume the machine formats are accessible by content negotiation?  

They are accessible by content negotiation indeed. In particular, one can use:
- "application/rdf+xml" for having the data in RDF/XML;
- "text/turtle” or "text/n3" for having the data in Turtle;
- "application/ld+json" or "application/json" for having the data in JSON-LD;
- any other kind of request will result in having the data in HTML.

E.g.:

> curl -L -H "Accept: application/rdf+xml" 
> http://www.sparontologies.net/mediatype/text/plain 
> 

will return the RDF/XML representation of such media type.

> It might be handy to also include direct links on the web page.  For example, 
> I found I could change the .html suffix to .ttl or .json to get alternative 
> representations, but that required guesswork on my part.

That’s right, you can access directly to the related formats by adding .rdf, 
.ttl, .json and .html respectively. I’ll add a note in the homepage of the 
dataset and in each HTML page of each mime type soon.

> 2. Are there any plans (and resources) in place to ensure longevity of the 
> sparontologies.net domain?

Well, actually now the sparontologies.net  website 
is handled by me, with personal resources. I would like to keep it available 
without external funds, but in case it will be more expensive to maintain I 
think I can ask for support to the University of Bologna, that’s the plan at 
least. Do you have something different in mind?

> 3. I note that the HTML page indicates cc-by licensing of the content, but 
> there is no such information in the machine readable formats.

That’s right, I have to add these data into the RDF representation as well – in 
my todo list.

Have a nice day :-)

S.





Silvio Peroni, Ph.D.
Department of Computer Science and Engineering
University of Bologna, Bologna (Italy)
Tel: +39 051 2094871
E-mail: silvio.per...@unibo.it
Web: http://www.essepuntato.it
Twitter: essepuntato

New LOD dataset for media types

[Silvio Peroni]

Dear friends,

We are pleased to announce a new LOD dataset

http://www.sparontologies.net/mediatype

that makes available media types defined as proper resources in RDF, according 
to the SPAR Ontologies [1] and DCTerms [2].

A media type is an identifier (for example "text/turtle") for file formats on 
the Internet composed by two parts: a registry ("text" in the example) and a 
record ("turtle" in the example). They are handled by the Internet Assigned 
Numbers Authority (IANA), which is the official authority for the 
standardisation and publication of these classifications.

The aforementioned web space, part of the SPAR Ontologies website, has been 
reserved for providing the RDF representation of each media type defined by 
IANA [3]. In particular, a media type is accessible by concatenating the URL 
"http://www.sparontologies.net/mediatype/; with its related identifier. For 
instance, "http://www.sparontologies.net/mediatype/text/turtle; allows one to 
access the specific resource for the "text/turtle" media type. 

Each media type can be accompanied by agents who acted as contributors, the 
related RFC documents documenting it, its current status (either official, 
deprecated or obsoleted), and direct links to Wikipedia pages and DBpedia 
resources related with the media type.

All these resources defining media types, thus, can be used for specifying 
particular formats (e.g., by means of the DCTerms property dcterms:format) that 
a certain entity, such as a book or a dataset, can have.

Please do not hesitate to contact us (sparontolog...@gmail.com) for questions 
and additional information about this LOD dataset.

Have a nice day :-)

S.



1. http://www.sparontologies.net
2. http://dublincore.org/documents/dcmi-terms/
3. http://www.iana.org/assignments/media-types/media-types.xml



Silvio Peroni, Ph.D.
Department of Computer Science and Engineering
University of Bologna, Bologna (Italy)
Tel: +39 051 2094871
E-mail: silvio.per...@unibo.it
Web: http://www.essepuntato.it
Twitter: essepuntato

Re: New LOD dataset for media types

[Melvin Carvalho]

On 5 November 2015 at 00:21, Silvio Peroni  wrote:

> Dear friends,
>
> We are pleased to announce a new LOD dataset
>
> http://www.sparontologies.net/mediatype
>
> that makes available media types defined as proper resources in RDF,
> according to the SPAR Ontologies [1] and DCTerms [2].
>
> A media type is an identifier (for example "text/turtle") for file formats
> on the Internet composed by two parts: a registry ("text" in the example)
> and a record ("turtle" in the example). They are handled by the Internet
> Assigned Numbers Authority (IANA), which is the official authority for the
> standardisation and publication of these classifications.
>
> The aforementioned web space, part of the SPAR Ontologies website, has
> been reserved for providing the RDF representation of each media type
> defined by IANA [3]. In particular, a media type is accessible by
> concatenating the URL "http://www.sparontologies.net/mediatype/; with its
> related identifier. For instance, "
> http://www.sparontologies.net/mediatype/text/turtle; allows one to access
> the specific resource for the "text/turtle" media type.
>
> Each media type can be accompanied by agents who acted as contributors,
> the related RFC documents documenting it, its current status (either
> official, deprecated or obsoleted), and direct links to Wikipedia pages
> and DBpedia resources related with the media type.
>
> All these resources defining media types, thus, can be used for specifying
> particular formats (e.g., by means of the DCTerms property dcterms:format)
> that a certain entity, such as a book or a dataset, can have.
>
> Please do not hesitate to contact us (sparontolog...@gmail.com) for
> questions and additional information about this LOD dataset.
>
> Have a nice day :-)
>

Thanks for some great work!

Question:  do you find this work related to the IANA discussion on media
types here:

https://github.com/mnot/I-D/issues/140


>
> S.
>
>
>
> 1. http://www.sparontologies.net
> 2. http://dublincore.org/documents/dcmi-terms/
> 3. http://www.iana.org/assignments/media-types/media-types.xml
>
>
>
> 
> Silvio Peroni, Ph.D.
> Department of Computer Science and Engineering
> University of Bologna, Bologna (Italy)
> Tel: +39 051 2094871
> E-mail: silvio.per...@unibo.it
> Web: http://www.essepuntato.it
> Twitter: essepuntato
>
>