Here’s Arnaud’s reply to my comments….

Great, Marc Canter found some time to answer my questions. SO time to comment on his answers.

He confirms my suspicion that OPML is a good way to store and exchange MicroContent (he wrote it as micro-content by the way). They did use Weboutliner for that. I have to play with outliners to see what can be done. See if I can get the Structured Blogging approach into OPML somehow.

He also explains why he likes Structured Blogging. The main reason is that it is an effort to get things going. I agree with him here. If I look at the number of clients which I can relate to MicroContent, it is staggering. It is time to get this local information published in a structured way. It shouldn’t be left on someone’s PC. And Structured Blogging helps here. Time to dive into it a bit more.

Marc Canter sees a lot happening in the field of MicroContent. Great! I’ve become a lot less pessimistic, so it is good to hear he sees some movement going on. We will get at MicroContent nirvana in the end!

By the way, I like the fact that Marc Canter answers me on his weblog and not in my comments. We have a public conversation, but we stay owner of the things we write, as we publish it on our own weblog. This kind of conversation corresponds to the ideas put forward on the Datalibre mailing-list. Any evidence of a conversation should be visible in the trackbacks.

[MicroContent Musings]

There’s another reason I’ve been using the public reply technique - comments are broken. Not only my TypeKey comments - but comments in general. They’re being turned off in droves and they hide the thread from view. I prefer to see things out in the open. Perhaps notifation or some sort of Attention.xml standards can change all that.

But that’s not my area or work or exertise right now.

Micro-content is. However we spell it.

Arnaud is one of those true believers too - so I’m going to use this thread with him - to highlight a bunch of different things I’ve wanted to be known in public - on the record - for sometime now.

I’m not really amazed at how smart Arnaud is - as it takes that to grok all this stuff. As is Thomas Van Der Wahl - who I’ve been getting to know receenly - but that’s another post. I met Thomas at SXSW and he calls DLAs ‘Info Clouds’ (with a space in the middle) - but that’s what’s MOST beautiful about these ideas.

I’m sure Michael Sippey and Ben Trott have their own name for all this at 6A - and I bet Barak has HIS own term for it - but we’re all (hopefully) working on the same principles.

So here are some fresh replies to Arnaud……

1. Yes OPML rocks, It has since the day it was explained to me by Dave Winer at his house the week he created it. Our outliner - the WebOutliner - uses OPML as it’s native form and we did LOTS of work on extending it for various purposes (see below.)

2. So let me tell you about some experiments we did on BOTH Structured Blogging (which is what it’s being called right now) and storing MicroContent (properly spelled with the in-cap) in OPML.

2a.

The Birthday party activity

So way back like almost three years ago I run into a writer and XSLT scriptor named Chuck White who had written several books on it - and who wanted to put it to use - with a real interface.

So Chuck and I worked on an example UI - which took some wizard collected info, converted it to OPML and sent it to our weboutliner. That structured data was designed to represent all the info needed to encompass a group based community birthday party.

It worked. It rocked. And one day I’ll put it up to show everyone. But for now let me explain how it worked and what it proved.

3. First one selected: what kind of Group, named the Group and decided who was an initial member of a Group. This was all created from a simple web page - but ws designed to work via mobile. To become what I promised Hward Rheingold “Tools for the Mob”.

4. The human was then passed into an activity grid - at which point they’d select one category of activities; such as after-school or sports or at-home fun - and then were sent through detailed tab dialogs - which enabled one to fill in all the meta-data regarding this selected activity (how many tables to set up?, what games?, by what rules?)

5. Now with our micro-content (sorry I mean MicroContent) activity data all loaded up - we XSLTed the data into OPML and sent it to our weboutliner - which (presumably)_ would be running an extension to support that particular activity and it’s associated meta-data. The underlying notion was that we could transform this data into any form, for any kind of output, format or device.

6. It worked like a charm, was easy enough to be categorized as “situated” by Clay Shirky and allowed me to extend the outliner tool metaphor into what I now call a “stucture editor”.

7. Humans certainly understand:

I - Plan party– define basic party meta-data
– select music and images to play at party
– set up party web site and blog
– establish plans/agenda for party

II - Send out invites– plug in list of names or retrieve previous names
– send out invites
– status of invites (data sent to dashboard)

III - Throw party– start executing agenda for the day - drive media devices
– execute milestone during party
– change party machines to upload stations

IV - Memories– collect memeoires after party
– send out thank you notes
– archival commenting

It’s just a matter of how the UIs implement this sort of structure.

7. All this work will appear - soon - in a ‘Family Oriented DLA’ we’re working on.

Still trying to figure out Arnaud’s last name…… And where he lives and what he does as a day job. I hope he’s really a php hacker looking for work - that would make it PERFECT!

This entry was posted on Friday, April 8th, 2005 at 8:02 am and is filed under . You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.