Re: Importing XMP data with SemWeb



On Thu, 2006-04-20 at 09:49 -0400, Warren Baird wrote:
> Larry Ewing wrote:
> 
> > Sadly, as far as I know, what I said and what you pasted here mean the
> > same thing.  It would definitely be preferable to nest the blank node in
> > the subject and avoid the nodeID altogether but I think this is a valid
> > serialization (I'd be happy to be proven wrong).  Improving the SemWeb
> > output would be welcome of course.  
> 
> Gah...   It definitely looks like you were right.   I read part of the 
> RDF primer last night and I've determined 3 things:
> 
> 1. I need to do a *lot* more reading on RDF to be able to write an 
> import mechanism that deals with general XMP data
> 2. I don't really care about importing general XMP data - my current 
> approach of just treating any rdf:Bag as a container of tags is good 
> enough for me (and probably 75% of the other people who want to import 
> metadata into f-spot)
> 3. XML schemas designed by committee are *scary* things...
> 
> I'm afraid I don't have *that* much time to work on f-spot, so I need to 
> focus my effort on the gaps that are affecting me the most.
> 

I understand.  When implementing the current f-spot metadata code I
found I had basically underestimated the complexity at every scale and
ended up spending much more time than I intended on getting it working.
So once things were minimally working I focused on other more visible
and important f-spot issues.   I haven't yet had the stomach to wade
back in to metadata and get it to the point where it is providing much
value to the user.  Hopefully that will change in the future, getting
your patch working is a good first step, so is the threaded metadata
writing patch.

> I'm going to finish off my current import work --- it's done in a fairly 
> general way - I've added a GetTagNames method to MetadataStore - it 
> currently just returns an array containing every string that shows up 
> inside a bag, but if someone else cares enough and knows enough RDF/XMP, 
> they could update it to do 'the right thing' and the rest of the import 
> mechanism would just work...
> 

I'll happily make the tag retrieval work properly.

> I'll send some patches in for review as soon as I can...
> 

Sounds great.

> 
> > p.s.> I just wanted to take this opportunity to say the whole thing
> > makes me sad again.
> 
> Yeah, it makes me sad too --- RDF and XMP seem to be a whole lot more 
> complicated than they really need to be...  *sigh*
> 

On a different day I'd go into a long rant about how software standards
grow geometrically in complexity by basing themselves on other standards
that in turn base themselves on other standards.  Then I'd continue with
a teatise on how using broad standards as a basis ends up meaning you
inherit all the complexities of the basis which are often counter
productive and/or inappropriate for your problem space.  But instead of
going deeper into that and the painfully obvious counter arguments
(there are many) I'll just say I think it is high time someone comes up
with a corollary to the Law of Software Envelopment [1] as it applies to
standards.

--Larry

[1] http://www.jwz.org/hacks/





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]