Re: [Urgent] Proposed changes to Bioschemas

Hi Tony, Melanie, Alasdair,

If MarRef and others are already using Samples as a type, I think it is 
just a matter to formalize it, i.e., including Sample as a type as part 
of this proposal shared by Alasdair.

@Alasdair, you have gone through the profiles already. Do you think it 
is possible to add Sample on this first round? I am aware that your 
deadline was today but Sample profile is mature enough and in use so 
nothing should changed when promoted to type, although new things could 
be added if needed.

Gene and Protein had been discussed by the community but none of them 
was being used already, there is some room for more changes on those and 
that has been seen on the discussions. But with Sample, everything that 
is already in place should be kept.

Any thoughts?

Regards,





On 2018-10-31 17:54, Tony Burdett wrote:
> Hi Alasdair,
> 
> On 31/10/2018 14:12, Gray, Alasdair J G wrote:
>> 
>>> Speaking for the samples group, I don't think everyone involved has 
>>> had time to evaluate the impact of this proposal on the samples 
>>> driving usecases, including the MarRef deployment and Biobanks 
>>> exchange work. At first reading I think this will make it harder to 
>>> demonstrate utility for these deployments. If the samples usecases 
>>> were deprioritized in favour of other more compelling examples, 
>>> that's reasonable, but the proposal document doesn't indicate how 
>>> this was determined and I'd like to have a chance to review the 
>>> results of that decision-making process.
>> 
>> Sample was not prioritised at this point as I felt that we had a 
>> strong working example of this being used in live deployment based on 
>> the BioChemEntity, i.e. Sample has not been deprioritised. I do not 
>> believe that the proposal would affect Sample profile or the 
>> deployments, but please correct me if I am wrong. My intention had 
>> been to follow up with the Sample group, but time has not permitted 
>> this to happen. We are more than happy to have discussions on whether 
>> there is a need for a Sample type.
> 
> Melanie is doing a more detailed analysis now and commenting on the
> proposal document. The existing Sample deployments require some
> mechanism to identify those BioChemEntities that are samples (whether
> this is via a property describing the type as an ontology URI or
> having a dedicated Sample type doesn't really matter, but the MarRef
> implementation assumes Sample as a new type) and the ability to link
> from PropertyValue to CategoryCode. We've built implementations on the
> assumption these will form part of the proposal and I think it would
> be good to keep these in order to send the right message to those
> early adopters who are using bioschemas markup for something other
> than search indexing.
> 
> Thanks,
> 
> Tony

Received on Wednesday, 31 October 2018 18:31:20 UTC