W3C home > Mailing lists > Public > public-multilingualweb-lt@w3.org > April 2012

Re: [all] suggestions for consolidating requirements

From: David Lewis <dave.lewis@cs.tcd.ie>
Date: Fri, 27 Apr 2012 10:54:42 +0100
Message-ID: <4F9A6CE2.2040909@cs.tcd.ie>
To: public-multilingualweb-lt@w3.org
Hi Felix, guys
Yes, we discussed the need to start populating the data category 
requirements at the end of the WG call yesterday.

The current set of data categories is just our initial bucket of ideas, 
so I think it needs a little bit more work before investing the effort 
of writing use cases, to avoid too much duplication at that stage.

We've identified around 5 obvious consolidations, so I think we should 
first work through those on the list with the aim of revising them in 
the requirements document within the week, and then start actioning the 
authoring of use cases at the call next week. I'll send out actions 
about these consolidations later today, so by the call next friday we'll 
have a slightly more coherent set of data categories ready to be fleshed 
out with use cases.

To lay the ground for writing the use cases, we already have a suggested 
list of product classes in the requirements document at:
http://www.w3.org/International/multilingualweb/lt/wiki/Requirements#Product_Classes_Implementing_Requirements

I'd invite everyone to review these to make sure their individual 
products are properly classified.

In addition I reintroduced a list of potential use case roles at:
http://www.w3.org/International/multilingualweb/lt/wiki/Requirements#Use_Case_Roles

Again, if people can familiarise themselves with these and suggest 
revisions or different roles if they feel something is missing.

We could then actually put fictional people names against these roles, 
out 'cast of characters' to make the use case benefits more engaging 
(felix you pointed to 
http://www.w3.org/2008/WebVideo/Fragments/WD-media-fragments-reqs/ as a 
good examples of use cases).

Finally, I'm completing an action to produce a table that gives the list 
of data categories against: i) the current level of interest in the 
working group, ii) the level of interest shown in the questionnaire and 
iii) our perceived level of difficulty (technical and concensual) as a 
tool for helping us to prioritise work. We can use this then also to 
track implementation commitments as you suggest.

cheers,
Dave




On 27/04/2012 09:04, Felix Sasaki wrote:
> (not fully back yet, but throwing one aspect in here)
>
> In general, it might make sense to trigger these discussions and other 
> discussions of data categories by applications that will be developed 
> this or next year. In the next weeks we need to start consolidating 
> data categories asking ourselves
> 1) Who will implement this?
> 2) What tooling, usage scenario, users are behind this?
> 3) What test cases will be behind this
> I assume that answers to these questions will help a lot to resolve 
> issues coming up in the "top down" data category discussion in this 
> and other threads.
>
> Best,
>
> Felix
>
> Am 26. April 2012 16:07 schrieb Arle Lommel <arle.lommel@gmail.com 
> <mailto:arle.lommel@gmail.com>>:
>
>     Hi all,
>
>     Desí understanding reflects my own when I put these in. I see
>     these are pretty distinct, because confidentiality is not about
>     licensing, but about whether you can expose the content publicly
>     in a process. You might have a case where you have content that
>     will be GPLed (licensingTerms) but not want it exposed publicly
>     yet (confidentiality). The first would tend to be in published
>     materials (i.e., you've put something on the web and you are
>     telling a tool, potentially one you don't even know is looking at
>     your content, whether it can scrape it and use it; the second
>     would be process metadata telling you how to act in your process.
>     I think they are distinction enough (and operate in different
>     areas) that it makes sense to separate them.
>
>     -Arle
>
>
>
>     Sic scripsit Des Oates in Apr 26, 2012 ad 05:59 :
>
>>     Dave, re this case below, Iím not sure if this is a full
>>     overlap.  The use case mentioned in the contentLicensingTerms
>>     section covers  MT Training corpora.   Confidentiality is more
>>     applicable to a Translation Workflow use case, where the decision
>>     logic is more likely to be implemented in a TMS.   If both cases
>>     are covered by this data category, then it would be OK, but I
>>     want to confirm this is the case first.
>>
>>     Thanks
>>
>>     Des
>>
>>     Moritz, Des: I think there is potential to combine
>>     'confidentiality' with 'contentLicensingTerms', with
>>     confidentiality being a specific value of contentLicensingTerms,
>>     whcih could perhaps be relaised using Creative Commons license
>>     classes
>>
>>
>>
>
>
>
>
> -- 
> Felix Sasaki
> DFKI / W3C Fellow
>
Received on Friday, 27 April 2012 09:55:13 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 9 June 2013 00:24:55 UTC