Re: Princples...

ACK

Dr. Lewis John McGibbney Ph.D., B.Sc., MAGU
Engineering Applications Software Engineer Level 2
Computer Science for Data Intensive Systems Group 398M
Jet Propulsion Laboratory
California Institute of Technology
4800 Oak Grove Drive
Pasadena, California 91109-8099
Mail Stop : 158-256C
Tel:  (+1) (818)-393-7402
Cell: (+1) (626)-487-3476
Fax:  (+1) (818)-393-1190
Email: lewis.j.mcgibbney@jpl.nasa.gov

           [cid:14E628FB-FFC7-4831-8D88-71D6BBF55746]

 Dare Mighty Things

From: Joshua Lieberman <jlieberman@tumblingwalls.com<mailto:jlieberman@tumblingwalls.com>>
Date: Thursday, April 23, 2015 at 9:03 AM
To: Clemens Portele <portele@interactive-instruments.de<mailto:portele@interactive-instruments.de>>
Cc: Simon Cox <Simon.Cox@csiro.au<mailto:Simon.Cox@csiro.au>>, Chris Little <chris.little@metoffice.gov.uk<mailto:chris.little@metoffice.gov.uk>>, "thiago.avila@ic.ufal.br<mailto:thiago.avila@ic.ufal.br>" <thiago.avila@ic.ufal.br<mailto:thiago.avila@ic.ufal.br>>, "Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au>" <Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au>>, "public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>" <public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>>
Subject: Re: Princples...
Resent-From: <public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>>
Resent-Date: Thursday, April 23, 2015 at 9:03 AM

This seems to be part of a larger discussion about principles of scalability, which we haven’t really paid any attention to (yet).

-Josh

On Apr 23, 2015, at 11:41 AM, Clemens Portele <portele@interactive-instruments.de<mailto:portele@interactive-instruments.de>> wrote:

Simon, that implementation pattern may be useful for those that want to publish their (spatial) data as RDF/SPARQL in addition to the other representations/APIs they support, but I would be concerned, if we try to promote this as a best practice in general.

Clemens


On 23 Apr 2015, at 15:42, <Simon.Cox@csiro.au<mailto:Simon.Cox@csiro.au>> <Simon.Cox@csiro.au<mailto:Simon.Cox@csiro.au>> wrote:

The opportunity here is to present it as ‘virtual’ linked data.
I.e. only generate a graph in response to a request. The canonical expression of the query is in SPARQL, using the W3C datacube vocabulary. An interface layer translates the request into the native API. The LDA enables you to mask the query as a linked data URI.

A provider should limit the size of responses, or use paging.

A couple of layers of indirection/virtualization.

??

From: Little, Chris [mailto:chris.little@metoffice.gov.uk]
Sent: Thursday, 23 April 2015 9:31 AM
To: thiago.avila@ic.ufal.br<mailto:thiago.avila@ic.ufal.br>; Taylor, Kerry (Digital, Acton)
Cc: public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>
Subject: RE: Princples...

Thiago, Kerry

Meteorology has use cases where linked data/RDF/triple stores/etc are not appropriate. Take a large (~1TByte) 5 dimensional data cube identified by only one URI/link, of one parameter, say the vector wind speed and direction,  and it is replaced every 6 hours. The data is all geo-referenced, and users wish to extract much more manageable subsets.

At this stage, I am not sure whether  all practices of W3C Best Practices for Publishing Linked Data<http://www.w3.org/TR/ld-bp/>
are helpful.

Chris

From: Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au> [mailto:Kerry.Taylor@csiro.au]
Sent: Thursday, April 23, 2015 8:01 AM
To: thiago.avila@ic.ufal.br<mailto:thiago.avila@ic.ufal.br>; public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>
Subject: RE: Princples...

Thiago,
I  do not think we should be reiterating those practices – that is not our job. It would be unfortunate if we were contradictory, though.

In our case, it is not obvious to what extent this group is focusing on linked data, or not, and I think our views in the group  may be divergent.

The UCR document should be “A document setting out the range of problems that the working groups are trying to solve.”  So in that context  I thought to bring up the question ( deliberately phrased in the form that reflects my point of view!).

Kerry

From: Thiago José Tavares Ávila [mailto:thiago.avila@ic.ufal.br]
Sent: Thursday, 23 April 2015 10:32 AM
To: public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>
Subject: Re: Princples...

Hello Kerry,

For 5th principle, Does geospatial data must follow all practices of W3C Best Practices for Publishing Linked Data<http://www.w3.org/TR/ld-bp/> ?

Regards. Thiago

2015-04-19 23:26 GMT-03:00 <Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au>>:
Here's 2 suggestions ( that may need improvement).

No 5. That community good practice for 5 star linked data be followed, including the use of so-called cool uris and  ontology annotations ( references need to be attached)

No 6. That ontologies  conform (" are valid" ? check) to the language of  owl2 dl.
(The latter is important for spatial and temporal reasoning.)

No 7. ( perhaps number  0) That these principles are aimed specifically at data published in RDF but where appropriate may also apply to other spatial data published on the web.


Kerry

On 20 Apr 2015, at 8:17 am, "John Machin" <john.machin@abs.gov.au<mailto:john.machin@abs.gov.au>> wrote:

Hi Ed, Andreas,

I like the proposed principles so far.

Based on some of the comments in the last call, I wonder if we have to have a principle related to keeping the practices up to date?

I realise that this might over-commitment from a WG with a specified lifespan but if maintaining currency is a principle then the two sponsor organisations may be encouraged to reconvene WGs to review and update the Best Practices periodically.

Cheers,
--
John Machin

<graycol.gif>Andreas Harth ---18/04/2015 05:57:59 AM---Hi Ed, On 2015-04-16 14:10, Ed Parsons wrote:

From: Andreas Harth <harth@kit.edu<mailto:harth@kit.edu>>
To: <public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>>,
Date: 18/04/2015 05:57 AM
Subject: Re: Princples...

________________________________



Hi Ed,

On 2015-04-16 14:10, Ed Parsons wrote:
> So to start the ball rolling....
>
> Princple No.1 : The linkabilty of Geospatial Information published on
> the web should be improved.
>
[...]
>
> Princple No.2 : We will not reinvent
>
[...]
>
> Feel free to add to these, develop more ... when we reach a level of
> agreement I will transfer them over to the wiki

How about the following?

Principle No.3 : Best Practices have to be visible.

We will link to at least one (or two, three?) publicly available
example(s) of a non-toy dataset that follows the best practice.

Cheers,
Andreas.