Hi,

When ever anyone mentions ISO metadata I think of:

ISO 19103
ISO 19115

Are we talking about these?

If we are then I would just like to mention that both of these standards
lose information and context when you attempt to populate the fields from
actual data products.

There is also the issue of interpretation as to what data fields to copy to
which metadata fields.

Because of this I would suggest that each data format should store it's
metadata in a lossless form.

An xlst should be provided to convert each data-format metadata to the
appropriate ISO metadata format.

Regards
Damian

PS. Metadata for data does not excite me, but it is a necessary evil that
is becoming very important.


On 15 October 2015 at 22:07, Kurt Schwehr <[email protected]> wrote:

> If anyone wants to follow up on metadata, Sean is right... here is a
> different subject thread for metadata.
>
> On Thu, Oct 15, 2015 at 11:44 AM, Sean Gillies <[email protected]> wrote:
>
>> Kurt,
>>
>> On Wed, Oct 14, 2015 at 4:20 PM, Kurt Schwehr <[email protected]> wrote:
>>
>>> Sean got what I was meaning to say.  Do we want the output to have some
>>> record of which mode was used.  I think the answer is likely no.
>>>
>>
>> I agree, no. Without a larger provenance framework of some kind there's
>> not much to be done with that information. And such a provenance framework (
>> http://www.w3.org/TR/prov-o/ for example) covers this kind of thing
>> already.
>>
>>
>>> And it reminds me that I need to think about a driver for writing ISO
>>> metadata files.  (bleck)  The processing steps would go in an xml iso
>>> metadata sidecar.
>>>
>>
>> Let's discuss ISO metadata in a different thread :)
>>
>> --
>> Sean Gillies
>>
>
>
>
> --
> --
> http://schwehr.org
>
> _______________________________________________
> gdal-dev mailing list
> [email protected]
> http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
_______________________________________________
gdal-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Reply via email to