> On 19 Sep 2015, at 19:16, Daniel Kinzler <[email protected]> wrote:
> 
> Am 19.09.2015 um 10:27 schrieb Egon Willighagen:
>> 
>> 
>> On Tue, Sep 15, 2015 at 10:12 PM, Michael Peel <[email protected]
>> <mailto:[email protected]>> wrote:
>> 
>>    It seems to assume a default uncertainty on values, though: I just added 
>> the
>>    elevation above sea level to:
>>    https://www.wikidata.org/wiki/Q1513315
>>    just specifying the central value, and it assumes that this value is +- 
>> 0.1
>>    km - which isn't a good assumption to make...
>> 
>> 
>> Did you mean to write 2.80 km about sea level? Then the error would be 0.01 
>> km
>> .... I am guessing the uncertainty follows the scientific notation of the
>> number... 2.8 has the numeric uncertainty of (about) +/- 0.1...
>> 
>> That sounds like a reasonable approach to me...
> 
> Yes, the uncertainty follows the scientific convention about significant 
> digits.

Taking that approach is a *really* bad idea. You can't just assume/make up 
uncertainties!

As an example, say you have a length of 100m. Which significant digit do you 
assume is correct? Is this +- 100m, 10m or 1m? What if it's referring to the 
length of a 100m run, where the accuracy could be much higher than the 
significant digit given, e.g. 100m +- 1cm? Or what if it's the size of a crater 
on a distant planet, where it might be 100m+-50m? Or if the actual value is 
100m +- 3m, but we say that it's +- 1m (which I see is the default in this 
case), which might be believable to readers but very misleading in reality?

> We are currently thinking about adjusting this a bit, see
> https://phabricator.wikimedia.org/T105623

As thiemowmde says there:
"The fact that the parser "guesses" a precision based on basically zero 
information always was and still is wrong. It must default to ±0. Everything 
else is misleading and a source of significant confusion and actual errors."
... although perhaps a better approach if possible might be to default to -1, 
or something else indicating the absence of data.

Thanks,
Mike


_______________________________________________
Wikidata mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikidata

Reply via email to