On emergent policy and self vs governance in common (Was: Re: [uf-discuss] Rough consensus and working code)

[Chris Messina]

I had "gone away" from this list a month or so back owing to Andy's
sometimes abrasive tone and pedantic reasoning. I simply didn't have
time to parse through all the hub-bub, as interesting as it might have
been to certain folks in the middle of it.

I'm glad that Tantek has taken action, as I previously encouraged him
to do, because, though I value Andy's positive contributions to the
list, the wiki and the community, many of his contributions worked to
unravel or undue the positive karma they earned him.

As Tantek said, it's a balancing act -- and Andy was very good
providing net neutral contributions.

But, I do not wish to dwell on that topic, for, at the very least,
groundbreaking action has been taken finally, and action that we can
learn from, in light of what's also come before us.

What I did want to talk about, however, are two things -- namely the
meta-centralization that the microformats-dot-org community represents
and the emergent policy that microformats, as an effort to codify a
series of best practices that become standard in web-transmittable
computer code, stands for. My goal is to illustrate the broader
purpose and perspective of the work we're doing, to propose a proper
ego-placement with regards to this work, and suggest potential
parallels which make the cabal-like governance work in certain
circumstances, and unravel in others, even within this community.

1. Where microformats fits in the broader picture.

I'll get this out of the way right now. The terms and names of
microformat classes, rel-values and so on don't matter. They don't. In
many sense, they're arbitrary, just as AJAX and HTML caught on.
They're simply placeholders for meaning, like the dollar bill is used
to transmit the meaning of value in society.

What is valuable, however, is agreement on terms. Agreement and
implementation between organizations and institutions -- for
implementation is non-binding, but by supporting a common cause, both
parties stand to benefit in ways neither is quite sure of yet, but
sees no reason to act to the contrary.

In this case, microformats such as hcard and hcalendar have found wide
support, because, unlike other external efforts that tried to reinvent
schema, we (Tantek in particular) dispensed with coming up with yet
more schema and went with existing convention (note that when we have
undertaken the "naming" process with new microformats, that process is
often where most of this community's contention and dissension lies).

But naming is an ego-driven event that is similar to an artist signing
his or her work; and when has a community produced a singular piece of
artwork? Rarely, if ever!

2. Why the microformats community operates as a cabal, and why it
should continue to do so.

Anyone who has participated in this community for some time will know
how hard it is to get a new microformat "blessed" -- that is,
accepted, documented, promoted and 'officialized' by the community.
There are many microformats efforts that have been relegated to the
scrapheap of semantic history or to the personal industry of smaller
parties, but very few efforts actually result in what we all would
call a microformat when we see it.

Truth be told, coming up with standards of any kind is a difficult and
harried process. There are those among us who have direct experience
with closed bodies who have and have not been successful with their
charge to develop interoperable standards and who could teach us all
about the quagmire that is standards development. But there is
strength is focus and in defending an ideal by intuitive fiat, even if
it seems unfair to those who have a great deal to offer but do not
have the same deftness that the incumbents possess.

As such, those who have been around from the beginning and have
weathered the hills and dales of this community have, in my opinion,
earned their seat at the table of the cabal. Fortunately, this cabal
is dependent upon the support of the community and upon obeisance of
its dicta or else it would simply cease to exist. In that way, the
controlling cabal is still very much subservient to the
implementations and good works of the community to give it its power;
if people stopped implementing or caring about microformats tomorrow,
regardless of their perceived arrogance or very real self-assurance,
their importance would only be to themselves.

And in that way, there is an important balance achieved, between
despotism and collaboration fueled by meritorious leadership. But,
this only scales to such a degree -- and feudalism can only hold so
long as the needs of the tenants are being met often enough. In the
case where centralization and cabalism leads to paralysis of natural
growth and species development, certain changes are in order.

3. On the continued rhizomatic development of microformats

A rhizome is a type of root-based plant that sends out lateral roots
to create offshoot new instantiations of itself. Strawberries are
rhizomatic as is ginger [1]. What's important about a rhizome  is that
it's growth path is predicated on similar and equal offshoots being
cultivated in environments in which the original may not have been
borne. As such, the offshoot is better healed to deal with the foreign
environment than if the original had simply been cloned or if it had
tried to impose itself on a foreign or hostile soil.

What does this have to do with this community? Well, for one thing,
the cabal-like institution of the microformats community leadership is
powerful because we give it its power. And I trust it to look out for
our best interests; at the same time, I think that there are
opportunities to both relieve some pent up pressure as well as
consider alternative models that would continue to effectively spread
microformats and the practices that this community espouses beyond our
areas of natural influence.

I think a salient example of this came recently when my partner, Tara
Hunt, was consider for deletion on Wikipedia (as I have been consider
before). Now, Wikipedians obviously have the interest of Wikipedia in
mind when they consider removing things from the index and they also,
one might surmise, have the readers in mind as well. However, in both
discussions over whether to remove Tara and myself from the index (and
this has been repeated for other people in the index as well) it was
the *individual bias of Wikipedia editors* that ruled out over the
unspoken interest of the minority communities that stood to benefit
from our inclusion (one person even suggested that I be kept in the
index since I was a "Notable programmer that assisted in creating a
few notable groups and browsers" [2] -- those who know me know that I
can't code for shyte -- and thus the reasoning for keeping would have
been arbitrary at best).

So, coming back to microformats, I think that it's time, as a matter
of governance and Darwinian evolution, that we actual begin thinking
about allowing new species of microformats to exist in the wild --
they may not receive a "blessing" by us, but I hardly think that all
the creatures on earth today were predicted in any non-secular books.

To this end, I would recommend the specific explanation and
characterization, vis-a-vis the microformats process, of efforts that
fall into any of these categories:

1. best practice -- a technique has been discovered to make the
composition of XHTML documents more consistent or more semantically
accurate, for example, using the <cite> tag
2. design pattern -- this isn't necessary a "data format" in the sense
that microformats should be about data interchange, but a design
pattern is XHTML that can be used to facilitate the development of
human interfaces, and may, for example, leverage existing microformats
to achieve its affect (an example could be if flickr applied a
behavior to hcards that allowed you to add a person marked up with the
hcard microformat to your friends list)... the presence of
microformats for a design pattern, however, is purely optional
3. exploratory/brainstorming -- gee, wouldn't it be great to have a
format for Smooth Peanut Butter? -- primarily at the early stages, no
code is necessary to explore a concept, but an interested or committed
following is present and is willing to document the problem they'd
like to solve and existing behavior
4. working draft -- essentially a series of conventions or best
practices have been developed that may show up in the wild and that
are probably "good enough" to start putting into use, with the
understanding that changes are still likely
5. recommendation/specification -- this is where things solidify
enough so that making a change has some impact... in fact, you could
use this stage to definitively mark up your documents knowing that a
change is unlikely; what separates this stage from becoming a "real"
microformat is implementations in the wild; if no one adopts or puts
this work into practice, you have a dead standard that would serve
only to clutter the microformat ecosystem
6. microformat -- only when there is mass deployment in the wild, such
that, given any significant sampling of pages on the open web, you
*might* bump into this format, should it then be considered an actual
microformat -- for in practice, the community at large (the one that
subsumes the microformats community and its leading cabal) has shown
its support by adopting the conventions recommended in the spec and
have shown their approval of it by *actually deploying it*

The last and final stage is the hardest, as it requires influence,
political might and campaigning; but those are the microformats that
will likely last and be embraced -- and, futhermore, are the most
indisputable because there are real, rather than imagined or
potential, statistics behind them.

Note that that list is preliminary, but does pay homage to the W3C
process stages, but in a much more informal way [3]:

   1. Working Draft (WD)
   2. Last Call Working Draft
   3. Candidate Recommendation (CR)
   4. Proposed Recommendation (PR)
   5. W3C Recommendation (REC)

Finally, to conclude, I would like to suggest that expanding and
making more explicity the preliminary stages of "microformat
crystalization" allows external communities to take this effort and
expand it beyond our natural sphere of influence or first-hand
knowledge. The purpose, of course, is to avoid the kind of
Wikipedian-myoptic purview that would lead the effort down the path of
exclusivity and stagnation. If anything stands out about the current
governance structure, it's that we have a strong political will in
Tantek who does a damn fine job keeping us on target but who, to the
detriment of the whole, hasn't allowed for market forces to take care
of the nascent efforts that might emerge external to this list.

If anything else, I want to avoid at all costs, now that we're seeing
popular support from Firefox et al, the conversion of our rich and
diverse community into a Tech Crunch-like kingmaker -- that people
somehow think they have to win favor with in order to be successful. I
think the point is that anyone should be able to build out and see
through the execution and development of a microformats, potentially
entirely outside of this list, simply by religiously adhering to the
principals by which we govern ourselves and allow ourselves to be
governed.

For all the times that Andy has asked Tantek "what gives you the
right?" there is an equal opportunity to say, "I give myself the
right" to take these ideas, these practices, the fundamental goes and
assumptions of this community and to strike out on my own, to pursue
that which I know is right and is valuable to a community that those
who reside on the list are unfamiliar with. For all Andy's struggles
to have his way, there was a larger goal of using simple principles to
semanticize the web that he could have, at any point, taken elsewhere
and not forked the community, but done his work in an environment that
suited him better.

I know why Tantek did why he did and I support him in his decision.
But I also support Andy's ability to pioneer his own efforts, not
necessary under the microformats name, but under the same principles.
And should he be successful, well, he certainly would have some
valuable bargaining chips to lay down when he offers his opinions to
the us and to the cabal, wouldn't he?

That's all for now.

Chris

[1] http://en.wikipedia.org/wiki/Rhizome
[2] http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Chris_Messina_%28open_source_ambassador%29
[3] http://en.wikipedia.org/wiki/W3c#Recommendations_and_certifications

On 1/4/07, Joe Andrieu <joe at andrieu.net> wrote:
> Hey folks, especially Tantek,
>
> I just wanted to reconnect in this forum with something I believe in
> strongly, even though my latest posts about governance might seem to be
> a bit contrary:
>
>   "We reject: kings, presidents, and voting. We believe in: rough
> consensus and running code."
>
> That's a quote from David Clark in 'Rough Consensus and Running Code'
> and the Internet-OSI Standards War[1].
>
> There's also a brief article in a similar vein "Rough Consensus and
> Running Code
> KISSING THE FROG OF HTTP" [2].
>
> So far microformats has done great with the rough consensus and running
> code. I think that's a large part of why it works.  We could probably
> improve on our rejection of kings and presidents, although it is hard to
> figure out how to do that without creating other problems.
>
>
> Cheers,
>
> -j
>
> [1]
> http://www.computer.org/portal/cms_docs_annals/annals/content/promo2.pdf
> [2] http://www.w3journal.com/4/s1.people.html
>
>
>
> --
> Joe Andrieu
> joe at andrieu.net
> +1 (805) 705-8651
>
>
>
> _______________________________________________
> microformats-discuss mailing list
> microformats-discuss at microformats.org
> http://microformats.org/mailman/listinfo/microformats-discuss
>


-- 
Chris Messina
Citizen Provocateur &
  Open Source Ambassador-at-Large
Work: http://citizenagency.com
Blog: http://factoryjoe.com/blog
Cell: 412 225-1051
Skype: factoryjoe
This email is:   [ ] bloggable    [X] ask first   [ ] private