W3C home > Mailing lists > Public > public-dwbp-wg@w3.org > January 2015

Re: Best Practice 1: Provide metadata

From: Phil Archer <phila@w3.org>
Date: Wed, 14 Jan 2015 17:02:05 +0000
Message-ID: <54B6A10D.2050304@w3.org>
To: Laufer <laufer@globo.com>
CC: Public DWBP WG <public-dwbp-wg@w3.org>
I'd be grateful if you did it for 2 reasons:

1. I'm not in the ideal place to do it this week;
2. I like collaborative text rather than one person's text.

So please go ahead :-)

Thanks

Phil.

On 14/01/2015 15:45, Laufer wrote:
>> I will change the text - unless you can get to it before me
> Tell me what you prefer...
>
> Laufer
>
> 2015-01-14 13:18 GMT-02:00 Phil Archer <phila@w3.org>:
>
>> Oh crap ... that's what comes of writing in a hurry. You are quite correct
>> Laufer, sorry.
>>
>> In my very unconvincing defence, Void does include info about vocabs used
>> and exampke SPARQL queries but that's no excuse. I will change the text -
>> unless you can get to it before me. I've just landed in Greece for a 2 day
>> workshop.
>>
>> Phil
>>
>>
>>> Hi, Phil,
>>>
>>> In the text of BP1, VoID is listed as suited for the description of the
>>> structure of metadata but I think VoID is more similar to DCAT, but for
>>> Linked Data. It describes the dataset as a whole.
>>>
>>> Am I wrong?
>>>
>>> Best Regards,
>>> Laufer
>>>
>>> 2015-01-13 11:00 GMT-02:00 Phil Archer <phila@w3.org>:
>>>
>>>> Bernadette, Newton,
>>>>
>>>> I've added some more BPs to the metadata section today. There's one more
>>>> I
>>>> really want to write ASAP which will be about providing structural
>>>> metadata
>>>> (what Mark Harrison called intrinsic metadata). That will include links
>>>> to
>>>> things like VoID for RDF datasets, the CSV work for those files etc.
>>>>
>>>> IMO the doc is looking better but still needs work before Friday.
>>>>
>>>> Personally, I don't think we should include incomplete BP templates in
>>>> the
>>>> FPWD. I'd rather see a list of the BPs to be inserted at that pint in
>>>> the
>>>> doc, so, for example in data formats just a simple list like:
>>>>
>>>> Provide machine-readable data
>>>> Provide data in standardized formats
>>>> Provide data in open formats
>>>> Provide data in multiple formats
>>>> Provide locale parameters
>>>>
>>>> (that list prompts all sorts of questions by the way - why aren't we
>>>> just
>>>> using the 5 star model? Shouldn't locale parameters be in the metadata
>>>> section? etc.)
>>>>
>>>> I think that's a more honest reflection of the current situation and
>>>> allows people to see which BPs have been drafted and therefore in need
>>>> of
>>>> review (or ridicule!).
>>>>
>>>> @Christophe, might you have time to add normative statements to the BPs
>>>> you wrote? i.e. add in the RFC 2119 keywords in the intended outcome
>>>> sections?
>>>>
>>>> And, wrt. Issue-115, IMHO the 'What' section can be merged with the Why
>>>> -
>>>> I think in all cases, unless you/others think differently??
>>>>
>>>> Phil.
>>>>
>>>>
>>>> --
>>>>
>>>>
>>>> Phil Archer
>>>> W3C Data Activity Lead
>>>> http://www.w3.org/2013/data/
>>>>
>>>> http://philarcher.org
>>>> +44 (0)7887 767755
>>>> @philarcher1
>>>>
>>>>
>>>
>>>
>>> --
>>> .  .  .  .. .  .
>>> .        .   . ..
>>> .     ..       .
>>>
>>
>>
>> --
>>
>> Sent from my phone. Please excuse typos.
>>
>
>
>

-- 


Phil Archer
W3C Data Activity Lead
http://www.w3.org/2013/data/

http://philarcher.org
+44 (0)7887 767755
@philarcher1
Received on Wednesday, 14 January 2015 17:02:21 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 14 January 2015 17:02:21 UTC