Re: Start of profiles analysis page - 2nd reply

On 12/6/17 9:58 AM, Ruben Verborgh wrote:
> Hi Karen,
> 
>> Not all access to APs will be through content negotiation, AFAIK, so we
>> have to consider other access avenues, such as a document at is located
>> on a web site, profiles in wikis, etc.
> 
> Most Web servers that can host an RDF document
> will also be able to support conneg,
> but yes, linking could be a simple alternative.
> 
>> If there is a "concept" AP it needs to be something that can be
>> represented, thus is not entirely abstract.
> 
> I don't think there it _needs_ to be like that.
> 
> But the AP can be represented in RDF etc,
> so that's not a problem.
> 
>> That seems to be the thing
>> that we are tasked with describing - the "concept" xyz that can be
>> rendered in various formats.
> 
> The most important thing is that it gets an identifier
> that is independent of the concrete formats.
> 
> Bonus points if that identifier supports conneg,
> but it can also just link to specifications.
> 
>> profiles need to be easy to create
>> by people who don't necessarily write code.
> 
> I'm all for that idea,
> but I'm not sure if it is realistic (unfortunately).

Unfortunately, the majority case appears to be that profiles are not
created by coders. A look at the current generation of profiles shows
that they are Word or PDF documents, most likely written by folks who
are knowledgeable of the semantics of their community's metadata but who
do not themselves write code. So, at least in the area in which these
operate, requiring code as a profile would be a certain non-starter.

In addition, many of the users of the profiles are data creators who are
human beings. These folks are a key audience for the information in
profiles, unless one separates the machine-actionable from the
human-actionable, which itself has a downside.

Essentially, we need to satisfy the full range from human to machine in
as smooth a transition as possible. The full range is needed for a
successful outcome.

kc

> 
> Best,
> 
> Ruben
> 

-- 
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
m: 1-510-435-8234 (Signal)
skype: kcoylenet/+1-510-984-3600

Received on Wednesday, 6 December 2017 22:02:38 UTC