Attention.xml and OPML, why not?


Dave Wilkinson:



"More thoughts on OPML - I have been thinking about one of the uses of OPML. OPML can be used to store and share structured data. With self-referenced links to additional information, and then further reference to that referenced. The Free Open Protocol World Wide Linkup Tool."


(via Cristian Vidmar)


As I've speculated before, OPML is going to get more and more attention over the next few months. There's something that feels very 'right' about it, and Dave has touched on one of its attributes that makes it so tantalising.


Like RSS, once I 'got it' I saw lots of problems around me that could be potentially solved with RSS. I'm getting this now again with OPML.


Over the weekend (at Mind Camp), I attended a great session on attention.xml run by Andy Edmonds. We reviewed the file generated by the AttentionTrust firefox extention (bizzarely named attention.txt, not attention.xml - reason? the xml doesn't parse...oh well...).


During the session, it occurred to me that OPML might be what we need to center the attention.xml solution around and move things forward. This is not an original thought. Nick Bradbury has been pairing attention and OPML in his latest release of FeedDemon, I'm sure others have been thinking the same way.


What I didn't realise was that Steve Gilmore Steve Gillmor has pushed for this previously but seemed to be discouraged by others. I only found this out while researching this post. I think that's a shame. I think OPML looks a lot like the way to go.


Here are the non-technical reasons I think it would be beneficial to have attention.xml done via OPML:




  1. OPML is already integrated into many RSS-aware 'products' (client-side applications and services) today.



  2. OPML will be used significantly more as time passes (see my 7 reasons 2006 will be a big year for OPML)



  3. OPML is simple -this allows many more than just hardcore developers to understand and play



  4. OPML is flexible enough to take on this role



  5. Attention.xml seems to be going nowhere, fast. Let the fans of OPML and attention.xml solve the problem - let them run with it, or OPML will run all over attention.xml


There are probably several other reasons too, I just can't think of them.


I doubt that there would be any insurmountable technical problems we couldn't solve if the will was there to do this.  So, what would need to happen - politically - to make this happen? I don't know.


I'm not sure what the Dave Winer take on this is, but would love to hear his views or have them pointed out to me if they already exist (what do you think Dave?).


Update: Nov 8. Steve and Nick pick up the baton.


-


Tags: Attention, RSS, OPML, Attention.xml

Comments (6)
  1. why does nobody ever spell gillmor right? i even wrote a blog on the subject

    http://www.redmonk.com/jgovernor/archives/000915.html

    if you want to support his ideas, it might help to get his name right. i dont mean that rudely, its a question of feedback and trackability and so on.

  2. MSDNArchive says:

    Thanks James, corrected – v.valid point re tracking….

  3. Danny says:

    IMHO, having a usable data model for the attention domain is the starting point, a common format on top of that is desirable. I personally believe the actual format used is secondary, as long as it’s good enough for the job and people can agree to use it. I’ve been critical of the specifics of Attention.xml myself in the past, but it’s good enough and it is already out there. I think it would be more productive to support the current XHTML-based representation than inventing another based on OPML.

    Ok, on your non-tech points:

    1. (X)HTML is already integrated into many RSS-aware ‘products’ (client-side applications and services) today.

    2. XHTML will be used significantly more as time passes

    3. XHTML is simple -this allows many more than just hardcore developers to understand and play

    4. XHTML is flexible enough to take on this role

    5. Attention.xml seems to be going nowhere, fast

    – true, but it isn’t clear how simply using a different format will improve matters technically or socially (especially given the points above). How will forking into a different format push things forward?

    Some technical points:

    1. OPML doesn’t play nicely with other XML formats – the lack of a namespace limits reuse for a start.

    2. Whatever format is used, a clear, unambiguous specification is a prerequisite for interoperability. There isn’t one for Attention in OPML, and that format’s history to date doesn’t offer much prospect of there being one in the forseeable future.

  4. Danny says:

    PS. "The Free Open Protocol World Wide Linkup Tool" – er, yep, that’s HTML+HTTP.

  5. charlie bess says:

    I wrote a blog entry the other day about how I’d use this kind of capability, without knowing about your entry, but someone pointed yours out.

    http://www.eds.com/sites/cs/blogs/eds_next_big_thing_blog/archive/2005/11/09/8210.aspx

Comments are closed.

Skip to main content