[Date Prev] [Date Next] [Thread Prev] [Thread Next] Indexes: Main | Date | Thread | Author

Re: [ba-ohs-talk] New backlink metadata; mhpurple v0.2 released


Eugene Eric Kim wrote:    (01)

> You may want to review Eric Armstrong's summary of a talk I gave on purple
> numbers a while back:
>
>   http://www.bootstrap.org/dkr/discussion/3462.html
>
> His e-mail includes a link to my slides.    (02)

Note:
I noticed this comment in that message:
       Murray's Plink program (Java) works on XHTML. It does a
      great job of adding p#'s, if you're working in xhtml.    (03)

I've had a chance to play with Amaya (the open source xhtml editor)
recently. And while it is not great, it is definitely servicable. It makes
aall-xhtml all the time into a serious possibility. And that's great because
we get serious HIERARCHY using xhtml.    (04)

If anyone is interested, I can send my feature wish list, which wil give
you some idea what it's missing. I also need to send that list to the
Amaya group, once I revisit their page and find the feedback address.    (05)

Or, since it's open source, I should find some time and do some hacking!
 :_)    (06)

> Rod Welch wrote:
> >
> > I gather mpurple links the end of a para to the beginning of a para. ..
>
> The link is there for the user's convenience.  It allows the user to
> easily copy and paste the paragraph's URL.  Frode Heglund originally came
> up with this idea a few years ago.    (07)

This is a factoid worth capturing in that message, and worth attributiing
to its brilliant originator. :_)  Sure wish I could revise that message!    (08)

> Rod Welch wrote:
> >
> > As a result, mpurple could make KM faster and easier, depending on the mechanics
> > of accomplishing the tasks.  Can you point to a body of work product showing
> > where this has been applied that demonstrates folks can knock out a lot of stuff
> > using mpurple?    (09)

The whole  point of adding granular addressability is to make it possible.
Since it hasn't existed heretofore, and since knowledge-management
strategies like XTM (topic maps), RDF, and CYC are only now coming
together (as he waves his hands in relative ignorance...), there is
naturally very little in the way of concrete examples to point to.    (010)


> 1. All of the experimental work I have done owes a tremendous debt to
> many, many people, especially Jack, Lee, and Eric (and of course Doug).
> That's called collaboration.  I'm sure many others in our loose-knit group
> would share similar sentiments.    (011)

Wow. Thanks, Eugene. I'm honored to be included in that group!
Wish I could make a living as a brilliant theoretician!
:_)    (012)

> 2. That we are all not working on a single piece of software does not mean
> that we are not collaborating.    (013)

Actually, I kind of have to disagree here. I think that real collaboration would
have involved either working on the same piece of software, or fully focusing
on the standards and techniques necessary to permit interoperability.    (014)

The problem, of the course, is the chicken and egg dilemma that has beset
us since day one. We did the best we could in the couple of hours a week
we were able to meet together, and in the little spare time we had to devote.
But, given the lack of any serious tool for remote collaboration and discussion,
we were unable to make much progress in designing such a tool remotely!    (015)

> This is what we have.  My favorite example of this is NODAL.  Many moons
> ago, Eric announced his KRNL project, an initial attempt at building a
> node library for a collaboration system:
>
>   http://www.treelight.com/software/collaboration/index.html
>
> The response on the list was underwhelming.  However, we know that at
> least one person looked at his stuff.  Several months later, Lee announced
> NODAL, and on numerous occasions, he has gone out of his way to credit
> Eric's KRNL system as an intellectual predecessor.    (016)

And on numerous occasions, I have expressed my delight that he was able
to follow the paper trail I left, which carefully documented my confusions
(as opposed to conclusions!), and was then able to perpetrate a brilliant
simplification that at once reduced the design and extended its applicability.    (017)

I have to agree, that is a delightful form of collaboration. Ok. My baby
came out ugly. But something very good came out of it!    (018)

> A lot of people have done a lot of cool work on this list.  TouchGraph is
> a great example, and it has found its way into Jack and Murray's work.
> Nexist, GSIX, Metaglom, and many other projects are other notable
> examples.  SDS too!    (019)

Wish I had more opportunity to investigate to see the commonalities
and the opportunities for more direct collaboration!    (020)

Thanks for your note, Eugene.
Stimulating reading!    (021)