Re: Data on the Web BP for review please

Hi Frans, Phil, BP Editors, and DQV Editors

Thank you for sharing your question about a best practice for significant
figures.  Here is my rationale including significant digits somewhere in
our BP work:
1) Adding a best practice for datum significant figures to Data Quality or
Data Enrichment
                         - OR -
2) Adding a significant figure literal property to the DQV dqv:Precision.
                         - OR -
3) Both 1 & 2.

Rationale:


Counting deals with exact numbers(counting 5 fingers)

All measurement data is inexact.


Accuracy refers to how closely a measured value agrees with the correct
value.
Precision refers to how closely individual measurements agree with each
other.


Significant figures are only directly linked to the precision. Significant
figures include all known digits plus one estimated digit.


Example:

known digits: 20 degrees centigrade measured by looking at a mercury
thermometer with 1 degree marks

estimated digits: 0.5 if the thermometer mercury appeared half way between
the 20 and 21 degree marks.

For this example the measurement with significant digits is 20.5.


Kind regards,


Eric S


On Fri, May 27, 2016 at 8:22 AM, Phil Archer <phila@w3.org> wrote:

> Hi Frans,
>
> Thanks for this.
>
> We had a telco a little earlier today and discussed your comment - so far
> without any resolution. There is sympathy with the point you raise but it's
> debatable whether it applies specifically to data on the Web as opposed to
> numeric data in general. Plus we're concerned that there is pretty much no
> time to create and receive reviews of a new BP.
>
> No definite decision yet though!
>
> By copying this reply to the DWBPs comment list, it is already in the
> system. no need to send it again unless you want to add more detail.
>
> Thanks
>
> Phil.
>
> On 27/05/2016 11:05, Frans Knibbe wrote:
>
>> Hello Phil, all,
>>
>> Thanks, the document looks great and hope many will read it and take it to
>> heart.
>>
>> One thing I miss is the advice to use significant figures in numerical
>> data. It is an easy way to make the data match their uncertainty, and in
>> many cases it helps to compact data too. Numerical data with the wrong
>> number of significant digits is a very common problem in geographical data
>> (e.g. geographic coordinates with nanometre precision).
>>
>> Should I post a personal comment about this to
>> public-dwbp-comments@w3.org?
>> Or do want to comment on significant figures as a group?
>>
>> Regards,
>> Frans
>>
>>
>>
>> 2016-05-25 10:26 GMT+02:00 Phil Archer <phila@w3.org>:
>>
>> Dear all,
>>>
>>> As you may have seen, last Thursday the DWBP WG published a new version
>>> on
>>> its BP doc and supporting vocabularies. These are now stable with a
>>> couple
>>> of specific issues in the vocabs and no known issues for the BP doc which
>>> is expected to transition to Candidate Recommendation next month.
>>>
>>> So we're now in the final call for comments ahead of the call for
>>> implementations which comes next.
>>>
>>> The input provided by the individuals in the To line of this mail are
>>> already acknowledged but if there are further comments, please get them
>>> to
>>> the WG by 12 June. I'm thinking in particular of the alignment with SDW
>>> BP
>>> to make sure that the latter sensibly builds on DWBP.
>>>
>>> Please see
>>>
>>> Data on the Web Best Practices
>>>  https://www.w3.org/TR/2016/WD-dwbp-20160519/
>>>
>>> Data Quality Vocabulary
>>>  https://www.w3.org/TR/2016/WD-vocab-dqv-20160519/
>>>
>>> Dataset Usage vocabulary
>>>  https://www.w3.org/TR/2016/WD-vocab-duv-20160519/
>>>
>>> Thanks
>>>
>>> Phil.
>>>
>>>
>>> --
>>>
>>>
>>> Phil Archer
>>> W3C Data Activity Lead
>>> http://www.w3.org/2013/data/
>>>
>>> http://philarcher.org
>>> +44 (0)7887 767755
>>> @philarcher1
>>>
>>>
>>>
>>
> --
>
>
> Phil Archer
> W3C Data Activity Lead
> http://www.w3.org/2013/data/
>
> http://philarcher.org
> +44 (0)7887 767755
> @philarcher1
>
>