W3C home > Mailing lists > Public > public-gld-wg@w3.org > March 2013

Re: Best Practices: was Re: Minutes for W3C GLD WG telecon 24-Jan-2013

From: Dave Reynolds <dave.e.reynolds@gmail.com>
Date: Mon, 18 Mar 2013 08:58:21 +0000
Message-ID: <5146D72D.9080005@gmail.com>
To: Bernadette Hyland <bhyland@3roundstones.com>
CC: public-gld-wg@w3.org
On 13/03/13 00:57, Bernadette Hyland wrote:
> Hi,
> Thanks Dave for taking the time to outline your thoughts vis a vis the
> procurement section.  The section was added last year at the suggestion
> of one of the members who had recently gone through the process of
> educating their procurement officer and wanted to share some of this so
> others could learn from that experience.

Such shared learning is great and an excellent topic for community 
forums such as the eGov interest group.

> That said, I agree that in every government agency will be different.
>   Procurement is a specialty with specific legal and process
> requirements as you noted.   I agree, the W3C is not the appropriate
> organization to publish procurement advice.  If someone is passionate
> about the topic, a community group properly scoped might be a better forum.

Glad we agree on this :)

> So I'm in agreement that there are several sections of the BP that
> should be shortened to "checklists" for consideration.  A checklist (to
> me) means, the following are issues may be unique to publishing &
> consuming LOD (i.e., it is open data vs. closed system, hosted on cloud
> vs. internal data center) and it would be helpful to cover this in your
> RFI / RFP process.  Period.
>
> In the version of the BP doc that we'll review at for this week's
> telecon, hopefully the doc't will reflect these changes to your
> satisfaction.

Looking at [1] there still seems to be a checklist.

I see it has been tweaked slightly (e.g. to mention APIs as well as 
SPARQL) but retains many of the problematic features that led to my -1 
vote. In particular, it seems to fail your test above in that it is 
mostly not LOD-specific.

[Minor: There seems to be a dangling bit of the Overview section that 
doesn't make sense on it is own.]

Cheers,
Dave

[1] https://dvcs.w3.org/hg/gld/raw-file/default/bp/index.html#procurement

> Cheers,
> Bernadette
>
>
> On Mar 7, 2013, at 4:37 AM, Dave Reynolds <Dave.e.Reynolds@gmail.com
> <mailto:Dave.e.Reynolds@gmail.com>> wrote:
>
>> -1
>>
>> I would like to register a formal objection to publishing BP with the
>> procurement section included.
>>
>> I've raised this issue several times over the last year, including the
>> last time BP was mentioned on a telecon [1].
>>
>> Since the section is still in there let me put my position on record
>> more strongly.
>>
>> I do not feel that W3C is an appropriate organization to publish
>> procurement advice and certainly not as a REC track document.
>> Publishing advice on technical capabilities that a procurement could
>> take into account might be reasonable but not advice on the
>> procurement process.
>>
>> The procurement checklist part of [2] is largely reasonable. Good work
>> was done there. However, the non-technical parts of it are superfluous
>> and incomplete,  the technical parts have omissions and questionable
>> inclusions. That is not a criticism of the work, it is in the nature
>> of the task that this will always be true. It is not an appropriate
>> goal to create a one-size-fits-all check list for the entire space of
>> public sector linked data procurements.
>>
>> On the non-technical parts then for every public sector linked data
>> procurement I have been involved in the procurement body has had
>> detailed guidelines and checks for things like selecting a suitable
>> procurement vehicle, vendor checks and service checks. The BP check
>> list is incomplete compared to any that I have been through (e.g.
>> omits financial stability checks, insurance levels, environmental
>> policy etc). I don't see how this list would add value to professional
>> public sector procurement departments who have a broad range of legal
>> constraints and policy guidelines already.
>>
>> On the technical parts that again it is largely reasonable, but omits
>> features that we have found critical to most public sector deployments
>> and includes items that are optional.
>>
>> In terms of omissions we have found it very important to also make
>> data available through a developer-friendly API, to document that API
>> and the data model in suitable terms, and to provide at least one
>> user-accessible application will pulls on that API and provides an
>> illustration of how that data can be used. A number of the tenders in
>> the UK these days in fact specify that such APIs and API documentation
>> are a mandatory part of the project. The BP checklist has no mention
>> of APIs, its mention of documentation seems to be aimed at the
>> software tooling whereas when purchasing a service it is the
>> documentation of the data-specific service and the community support
>> around that which is important.
>>
>> In terms of pieces that are not always appropriate the section assumes
>> a triple store ("graph database") publication of data and a SPARQL
>> endpoint. While this is true of every deployment my company has done,
>> it is perfectly possible to publish data as a set of static RDF & HTML
>> files which have been dynamically generated from a backend process. We
>> know public sector bodies who use this approach. They have very good
>> reasons for doing so and publish data which meets all the criteria for
>> TBL's 5* rating.
>>
>> I could pick on other details but the details aren't the point.
>>
>> To be crystal clear I am *NOT* suggesting that this section should be
>> rewritten to address these criticisms. I am suggesting it should be
>> withdrawn as not a suitable topic for W3C to address in a REC track
>> publication.
>>
>> Dave
>>
>> [1] http://www.w3.org/2011/gld/meeting/2013-02-14
>>
>> [2] https://dvcs.w3.org/hg/gld/raw-file/default/bp/index.html#procurement
>>
>> On 07/03/13 01:27, Boris Villazón-Terrazas wrote:
>>> Hi all
>>>
>>> I updated and made some cleaning to the BP document.
>>> @Bernadette, I modified the list of editors and included the authors
>>>
>>> The current ED version is here [1].
>>>
>>> I know there are still some open points, but I think we can publish it
>>> with some minor improvements as FPWD, what do you think?
>>>
>>> We have to:
>>> - Check the list of authors and their affiliations
>>> - Work on the current TO DOs
>>> - Include the references
>>> - Include the Acks section
>>>
>>> Sadly, I won't be able to join tomorrow's telecon.
>>> Please let me know what is the decision about this.
>>>
>>> TIA
>>>
>>> Boris
>>>
>>>
>>> [1] https://dvcs.w3.org/hg/gld/raw-file/default/bp/index.html
>>>
>>> On Thu, Feb 21, 2013 at 10:31 AM, Biplav Srivastava
>>> <sbiplav@in.ibm.com <mailto:sbiplav@in.ibm.com>
>>> <mailto:sbiplav@in.ibm.com>> wrote:
>>>
>>>
>>>    Yes from my side.
>>>
>>>    Regards,
>>>    --Biplav
>>>
>>>    **
>>>
>>>
>>>
>>>    From: Boris Villazón-Terrazas <boris.villazon@terrazas.name
>>> <mailto:boris.villazon@terrazas.name>
>>>    <mailto:boris.villazon@terrazas.name>>
>>>    To: Biplav Srivastava/India/IBM@IBMIN
>>>    Cc: Bernadette Hyland <bhyland@3roundstones.com
>>> <mailto:bhyland@3roundstones.com>
>>>    <mailto:bhyland@3roundstones.com>>, Joăo Paulo Almeida
>>>    <jpalmeida@inf.ufes.br <mailto:jpalmeida@inf.ufes.br>
>>> <mailto:jpalmeida@inf.ufes.br>>, Benedikt
>>>    Kaempgen <kaempgen@fzi.de <mailto:kaempgen@fzi.de>
>>> <mailto:kaempgen@fzi.de>>,
>>>    "public-gld-wg@w3.org <mailto:public-gld-wg@w3.org>
>>> <mailto:public-gld-wg@w3.org>"
>>>    <public-gld-wg@w3.org <mailto:public-gld-wg@w3.org>
>>> <mailto:public-gld-wg@w3.org>>
>>>    Date: 02/21/2013 02:57 PM
>>>    Subject:
>>>    Re: Minutes for W3C GLD WG telecon 24-Jan-2013
>>>    Sent by: boris.villazon.terrazas@gmail.com
>>> <mailto:boris.villazon.terrazas@gmail.com>
>>>    <mailto:boris.villazon.terrazas@gmail.com>
>>>
>>>
>>>    ------------------------------------------------------------------------
>>>
>>>
>>>
>>>    Hi Biplav
>>>
>>>    Thanks!
>>>    So, I can update section Source data [1] from here [2], right?
>>>
>>>    Best
>>>
>>>    Boris
>>>
>>>    [1]
>>>    _https://dvcs.w3.org/hg/gld/raw-file/default/bp/index.html#source-data_
>>>    [2]
>>>    _http://www.w3.org/2011/gld/wiki/Best_Practices_Discussion_Summary#Source_Data_
>>>
>>>    On Thu, Feb 21, 2013 at 9:22 AM, Biplav Srivastava
>>>    <_sbiplav@in.ibm.com_ <mailto:sbiplav@in.ibm.com>> wrote:
>>>
>>>    Hi
>>>
>>>    I have made changes to the "Source Data" section. The aim of the
>>>    section is to list issues someone publishing data may face and
>>>    provide practical guidelines. It turns out that listing issues is
>>>    simple (and quite general) but resolving them is very context and
>>>    government dependent. So a point-by-point response to issues is not
>>>    practical.
>>>
>>>    With the current content, I have improved the text listing issues. I
>>>    have also modified/ added to the general guidelines Ghislain had
>>>    kindly added (thanks!). Especially on being privacy aware (#8).
>>>
>>>    Please review.
>>>
>>>    Regards,
>>>    --Biplav
>>>
>>>    **
>>>
>>>
>>>    From:Bernadette Hyland <_bhyland@3roundstones.com_
>>>    <mailto:bhyland@3roundstones.com>>
>>>    To:Boris Villazón-Terrazas <_boris.villazon@terrazas.name_
>>>    <mailto:boris.villazon@terrazas.name>>
>>>    Cc:Joăo Paulo Almeida <_jpalmeida@inf.ufes.br_
>>>    <mailto:jpalmeida@inf.ufes.br>>, Benedikt Kaempgen
>>>    <_kaempgen@fzi.de_ <mailto:kaempgen@fzi.de>>,
>>>    "_public-gld-wg@w3.org_ <mailto:public-gld-wg@w3.org>"
>>>    <_public-gld-wg@w3.org_ <mailto:public-gld-wg@w3.org>>
>>>    Date:02/21/2013 01:33 AM
>>>    Subject:Re: Minutes for W3C GLD WG telecon 24-Jan-2013
>>>
>>>
>>>
>>>    ------------------------------------------------------------------------
>>>
>>>
>>>
>>>
>>>    Hi Boris,
>>>    I don't think you should remove the sections.  I think we should
>>>    still put a small section in to specify why these sections are
>>>    relevant and need to be considered [by someone publishing and/or
>>>    consuming LOD].  We just don't have a lot of detailed guidance to
>>>    offer at this time.  This is just my opinion.
>>>
>>>    Others?
>>>
>>>    Cheers,
>>>
>>>    Bernadette Hyland, co-chair
>>>    W3C Government Linked Data Working Group
>>>    Charter: _http://www.w3.org/2011/gld/_
>>>
>>>    On Feb 20, 2013, at 6:36 AM, Boris Villazón-Terrazas
>>>    <_boris.villazon@terrazas.name_
>>>    <mailto:boris.villazon@terrazas.name>> wrote:
>>>
>>>    Hi Bernadette
>>>
>>>    So, for FPWD purposes I'm going to remove those sections, ok?
>>>    I'll do it tonight.
>>>
>>>    Thanks
>>>
>>>    Boris
>>>
>>>
>>>    On Tue, Feb 19, 2013 at 6:37 PM, Bernadette Hyland
>>>    <_bhyland@3roundstones.com_ <mailto:bhyland@3roundstones.com>> wrote:
>>>    +1
>>>
>>>    We've briefly discussed in the WG the that there are a couple parts
>>>    of the BP document should be omitted, Stability  & Versioning being
>>>    two.  We can & should include language describing the issue for
>>>    consideration however, not propose practices to address it IMO.
>>>
>>>    Thanks,
>>>    Bernadette
>>>
>>>
>>>    On Feb 14, 2013, at 11:15 AM, Joăo Paulo Almeida
>>>    <_jpalmeida@inf.ufes.br_ <mailto:jpalmeida@inf.ufes.br>> wrote:
>>>
>>>    Dear Bernadette and others,
>>>
>>>    I have reviewed the section on Stability of the BP document, and I
>>>    believe it is not ready for prime time.
>>>
>>>    Honestly, I would recommend dropping this section or replacing it by
>>>    a fairly generic piece of text that only raises stability as a
>>>    concern (one or two paragraphs), and points to useful references
>>>    (further reading). I'm not an expert on the topic, but I am willing
>>>    to volunteer to produce this, if the group agrees this is the way
>>> to go.
>>>
>>>    There is a lot of work on long-term data preservation and the
>>>    "properties" described in the text (section 6.3 are quite confusing
>>>    and do not seem focused). This is not a simple issue, see, e.g. (a
>>>    lot of initiatives are listed including standards on long-term data
>>>    preservation):
>>>    _
>>>    __http://en.wikipedia.org/wiki/Digital_preservation__
>>>    __http://en.wikipedia.org/wiki/OAIS_
>>>
>>>    Best regards,
>>>    Joăo Paulo
>>>
>>>
>>>
>>>
>>>    *
>>>    From: *Boris Villazón-Terrazas <_boris.villazon@terrazas.name_
>>>    <mailto:boris.villazon@terrazas.name>>*
>>>    Date: *Sunday, January 27, 2013 5:50 PM*
>>>    To: *Bernadette Hyland <_bhyland@3roundstones.com_
>>>    <mailto:bhyland@3roundstones.com>>*
>>>    Cc: *Benedikt Kaempgen <_kaempgen@fzi.de_ <mailto:kaempgen@fzi.de>>,
>>>    "_public-gld-wg@w3.org_ <mailto:public-gld-wg@w3.org>"
>>>    <_public-gld-wg@w3.org_ <mailto:public-gld-wg@w3.org>>*
>>>    Subject: *Re: Minutes for W3C GLD WG telecon 24-Jan-2013*
>>>    Resent-From: *<_public-gld-wg@w3.org_ <mailto:public-gld-wg@w3.org>>*
>>>    Resent-Date: *Sun, 27 Jan 2013 19:50:54 +0000
>>>
>>>    Thanks Michael, Bernadette
>>>
>>>    Since Anne and Ron are not available, any volunteer to take care of
>>>    section 13. Stability? Biplav? Joao Paulo?
>>>
>>>    Best,
>>>
>>>    Boris
>>>
>>>
>>>
>>>    On Fri, Jan 25, 2013 at 5:52 PM, Bernadette Hyland
>>>    <_bhyland@3roundstones.com_ <mailto:bhyland@3roundstones.com>> wrote:
>>>    Hi Boris,
>>>    Thanks for circulating.  A couple WG members in yesterday's telecon
>>>    agreed to do a thorough review and comment.  This included Biplav
>>>    and Joao Paulo.
>>>
>>>    Note: We should update assignments to Anne and Ron who are no longer
>>>    involved in the WG AFAIK.
>>>
>>>    Cheers,
>>>    Bernadette
>>>
>>>
>>>    On Jan 25, 2013, at 4:08 AM, Boris Villazón-Terrazas
>>>    <_boris.villazon@terrazas.name_
>>>    <mailto:boris.villazon@terrazas.name>> wrote:
>>>
>>>    Dear all
>>>
>>>    Apologies for missing yesterday telecon's ... and for not sending my
>>>    regrets in advance.
>>>
>>>    Regarding the BP document, I was working with the BP document
>>>    including some comments from outside, e.g, Olaf, but I think it's
>>>    time to try to finalize this first version. We can make the rest of
>>>    improvements later on
>>>
>>>    The current version is here [1].
>>>
>>>    Currently, there are some sections to review, so I would suggest the
>>>    following section and reviewers:
>>>    - Background - Bernadette
>>>    - Linked Open Data Lifecycle -  Boris & Solve the current issue
>>>    (ISSUE-15)
>>>    - Vocabulary Selection - Boris/Ghis
>>>    - URI Construction - John ERickson/Boris
>>>    - 6. URI Policy for Persistence - Bernadette/John Erickson
>>>    - 8. Specifying an appropiate License - Bernadette
>>>    - 9. Security and hosting -  Michael Pendleton
>>>    - 10. Publishers "Social Contract" -  Bernadette
>>>    - 11. Pragmatic Provenance -  John Erickson
>>>    - 12. Versioning -  John Erickson
>>>    - 13. Stability -  Anne Washington (GMU), Ron Reck
>>>    - 14. Source Data -  Biplav
>>>
>>>    Reviewers would have one week to review their section, then we'll
>>>    have another week to fix the document. I'll be the one to be doing
>>>    the whole edition, and Bernadette can help me identifying possible
>>>    minor issues. More help is appreciated.
>>>    In summary after two weeks we should have a preliminary stable
>>>    version of the document, so we can publish as a FPWD ....
>>>    Then, we can continue with the weekly updates and improvements.
>>>
>>>    What do you think?
>>>
>>>    Best
>>>
>>>    Boris
>>>
>>>    [1] _https://dvcs.w3.org/hg/gld/raw-file/default/bp/index.html_
>>>
>>>    On Thu, Jan 24, 2013 at 5:04 PM, Benedikt Kaempgen
>>>    <_kaempgen@fzi.de_ <mailto:kaempgen@fzi.de>> wrote:
>>>    Hello,
>>>
>>>    See [1] for today's minutes.
>>>
>>>    Best,
>>>
>>>    Benedikt
>>>
>>>    [1] <_http://www.w3.org/2011/gld/meeting/2013-01-24_>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
>
Received on Monday, 18 March 2013 08:58:57 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:32:38 UTC