[DAS] Adjacent feature extension

[Andy Jenkinson]

On 7 Mar 2011, at 15:49, Jonathan Warren wrote:

> 
> On 7 Mar 2011, at 15:04, Andy Jenkinson wrote:
> 
>> On 7 Mar 2011, at 12:16, Jonathan Warren wrote:
>> 
>>> I'd say if we don't have any more objections in the next couple of days then go with your proposal as is? I'll then put support into the registry this week if that is the case. If you could also then copy the proposal from here https://github.com/dasmoth/dalliance/wiki/AdjacentFeatures to the extensions page here:
>>> http://www.biodas.org/wiki/DAS1.6E#Adjacent_Feature_filter noting in large letters that it was agreed by the community on such a such a date?
>> 
>> I think there is a lot left to be clarified so adopting it "as is" is a no go for me. In particular, take a look at this diagram and see if you can work out what will be returned with "adjacent" queries for either side of the viewing area, and do they make sense for what the client is trying to achieve?
>> <DAS-Adjacent.png>
>> 
>> The client has "seen" gene 2 and all its parts.
>> 
>> If the client asks for features adjacent to the left/right sides of the viewing area, what should the server return?
> I don't think it makes sense to ask for a next right in this case as there are features here already. This is for sparse data sources so it's ok just to return whats there if someone specifically wants to hit the next feature button or a client can blank the next right button out. It's up to the client.

Agree. But what I don't want to see is clients implementing some weird hybrid where they offer a "next right" button that bypasses SNP 2. If we expect clients to behave in a certain way, we should say so.

> Next left should return SNP1 if asked for an adjacent request.... or genes and constituents if filtered on gene.

Why SNP 1, and not any of the others at the same position? How is the server supposed to decide? Does it matter? How would this be worded in the spec?

> If you take the intention of this as finding features where data is sparse then I don't think there are big issues.

These aren't big issues (taken in that context), but I absolutely want to make sure we don't make the mistakes of the past by leaving ambiguity in the spec - whether it's an extension or otherwise. It's all very well us knowing what we designed it for, but if it isn't written down then it's going to cause problems. For the avoidance of doubt, I am very keen to get this done, but I see no sense in doing it in a way that we're not going to regret later (of which there are already countless examples).

> Part of the point of the extensions phase is to try these things out with examples and refine the specs. To leave acceptance of this will be a big mistake in my view.

I'm not sure what you mean by "leave acceptance". I'm trying to work through these things, not put blocks in the way. I am trying right now to implement it and these are the things I have immediately come up against so I need to get input right now on how to do it. Or to put another way, I can't create my example without refining the specs.