RE: [ba-unrev-talk] An important interview
This is long, but I think it all needs to be said (or at least I feel a need
to say it). (01)
Jack Park quoted:
http://www.linuxandmain.com/modules.php?name=News&file=article&sid=141
"There are a few genuine legends in the Linux community, and among them is
an Aussie named Carsten Haitzler. Who? [snip] (02)
"When projects get too big people spend more time in politics (talking on
mailing lists and waiting for others) than actually doing something useful.
Generally, splitting something up, not autonomous units, and have them work
on their own and just end up working in unison ends up more efficient,
imho. This still means people have to agree how they interface, but again,
imho, the "benevolent dictator" method when one or a very small number
decide the important bits (the glue between the parts) and then let the
rest roll. I also don't see "the more the better" as better. Too many cooks
spoil the broth. Sometimes one or two really good people will easily beat
10 or 20 average ones only working on something in their spare time. I
personally prefer the "crack troops" style. Get five or six really good
people and they can do a lot. Hundreds of part-timers, imho, don't work as
well."
I completely agree on the superiority of small, capable, committed groups
over sheer numbers. This is why I keep pushing for tools that support
personal organization and collaboration within small groups as an essential
starting point for any augmentation effort.
Ideas and solutions originate in individual brains. Environments that
support collaboration of comparably capable people can foster and support
the generation of ideas, but ideas are still individual.
In the current state of the art for technical development, we are still not
sure of all the elements that allow a "skunk works" to achieve the
phenomenal results that it does, but it appears to be the case that no other
form of organization is as efficient at problem solving. Gather talented
people with experience in the problem domain and necessary technical skills,
explain the problem to be solved well enough so that what you hear back is
consistent with what you thought you said, and then get out of their way.
Management of such a group then involves removing any barriers to production
and checking from time to time to see that the problem(s) they are
addressing are still the ones that need solving.
This method works, and hardly anything else does. We can complain that "it
shouldn't be that way", but complaining won't change the facts. Given the
facts, I submit that tools should focus on providing the support that will
allow talented individuals and small groups to collaborate in a more
productive fashion. Some of those groups will tackle larger problems with a
base of how to foster productive cooperation, and the larger problems will
get better approaches if not solutions.
The problems pointed out in the site that Eric Armstrong referenced,
http://www.well.com/~doctorow/metacrap.htm which deals with metadata,
impacts every activity of a large group. Expecting to solve problems by
simply having tens of thousands of people chime in with opinions will not
work except that it may allow individuals to find others with whom they can
form a working group that can actually produce something. (03)
I reviewed a partial list of what have been classified as the "complex
problems" facing society that the tools we are discussing are to help solve,
and I think we are going about this in the wrong way. (04)
Question: What elements do these problems have in common that qualifies them
as "complex problems" in the sense that we discuss them? What are the
elements of tools that would assist people in attempting to solve these
problems? (05)
I submit that one of the main elements that all of these problems have in
common is that they all have large social or sociological components to
them - a major part of the problem is getting all of the stakeholders
involved to agree on whatever solution is proposed. These problems have
technical components, but for the most part the technical aspects are
solvable with sound (possibly large scale) engineering practices if it were
possible to get any sort of effective agreement as to what would constitute
a solution to the problem that all stakeholders could live with. (06)
In many instances all current efforts are opposed by some power group or
other, and until that changes, no solution to the problem will be
*permitted*. The fact that some power group or other has a vested interest
in maintaining the problem, or is unwilling to take the actions required for
a solution to be implemented is a sociological problem, and without a
resolution of the sociological component of the "complex problem", no
technical solution can possibly succeed. (07)
Utopian ideas are always predicated on the idea that "if everyone only acted
thus and so, there wouldn't be a problem". The statement is often perfectly
true but not relevant because the reality is that "you can't get there from
here". Approaches that try to contradict reality will not work no matter how
wonderful the intentions, nor how great the idea sounds, nor how wonderful
things would be if only things were different. One would have thought that
we would have figured that out by now, but that is apparently a utopian
idea. (08)
On the other hand, when the workings of reality are correctly understood and
actions taken in accord with those understandings, we get workable solutions
that can be implemented. We are just now finding out that organic farming,
once we understand and use enough elements of the system together, is far
and away superior to the techniques we have been using. The technique of
agroforestry that Eric Armstrong reported appears to do just this. (09)
The issue for solving complex problems then, is largely one of finding
approaches that we haven't yet found to issues of getting people to
cooperate in the discovery and implementation of solutions to the
difficulties that we face, and then developing technical solutions that are
rooted in reality rather than wishful thinking, and do not require the
solution of even larger sociological problems. (010)
When a solution is such that it can be implemented by a small group or even
a single individual whether the masses support it or not, then those that
can learn and are more willing to adopt a new way than to continue to have
the problem, then large problems can have local solutions which it is more
difficult for the opposing or neutral majority to obstruct. Home schooling
is one example. It is possible to bypass the disaster that is public
education entirely rather than trying to get the people who are responsible
for the problem to solve it. (011)
"There is nothing more difficult to carry out, nor more doubtful of success,
nor more dangerous to handle, than to initiate a new order of things. For
the reformer has enemies in all who profit by the old order, and only
luke-warm defenders in all those who would profit by the new order. This
luke-warmness arises partly from fear of their adversaries, who have the law
in their favor; and partly from the incredulity of mankind, who do not truly
believe in anything new until they have had actual experience of it."
-- Machiavelli in The Prince (1513)
Notice the date - I see nothing to convince me that this problem has
changed.
Assertion: Any proposed solution to any problem that requires all people
involved to agree on a single solution at the same time just because it is
technically sound is doomed to fail.
By contrast, there are examples of working solutions that were never
designed but emerged from the actions of numerous individuals whose personal
interests overlapped in at least that one area. Adam Smith wrote about "the
invisible hand" in "The Wealth of Nations" and so this idea has been
discussed mostly in terms of markets and economics. However, such things as
language develop because all of the people involved find it to their benefit
to be able to communicate. Common law is a result of the actions of
individuals with a common desire for ways of conducting their affairs
peacefully, dealing with those that are not peaceful, and resolving disputes
by means other than violence. Conventions for use in email, netiquette, and
newsgroup protocols evolve and are (loosely) enforced due to the actions of
numerous individuals who have a shared interest in communicating over the
internet.
This is an aspect of sociological reality. It isn't necessary to like it,
but disagreeing won't change the fact that it is true.
So, rather than concoct all manner of grandiose schemes or fanciful
philosophies that require reality to be different in order to work, I
suggest that, if we are serious about developing tools to augment human
intelligence in resolving "complex problems", that we concentrate on aspects
of reality that can be validated and develop tools that allow individuals
with common personal interests to solve their problems with respect to
collaboration and productivity.
Language and law evolved slowly because the experiments took a long time.
Even after most of the principles were agreed upon, it took longer for them
to be captured in any sort of "standard reference" so that anyone who wished
could learn what was known about the common agreements. If we develop tools
that allow the same sort of evolution to proceed at speed within groups that
are interested in resolving *some* problem, working out the compromises and
best approaches, and capturing both the results of the effort and the nature
of the process in a form that is then accessible to all others interested in
the solution, we will have contributed to mankind's ability to solve
"complex problems" where there is a will to do so (where there is no will,
no solution is possible). (012)
Since we are talking about augmentation of individuals and small groups, we
are not necessarily talking about huge amounts of resources. This effort
doesn't require the approval of the planet, only enough agreement amongst
those interested in solving the problem of creating tools that are useful in
this context. The initial solutions do not have to scale to millions of
people because millions of people aren't going to use them at once and get
anything done.
Question: What are the elements (features) of a software tool that will
support this sort of activity well enough to allow individuals and other
groups to get on with the problem of solving "complex problems"?
Some of these we know, and some can be extracted from the problem statement.
* An individual must be able to use the tool on his own machine(s) to
capture, organize, and manipulate information, turning it into a useful
repository of personal knowledge.
* An individual must be able to publish some or all of the results of his
thinking to a wide (public) audience.
* An individual must be able to join with other individuals with interests
in the same problem domain to manage and evolve their shared information and
knowledge jointly using the same sort of organization that works for him as
an individual.
* Any individual may belong to multiple groups. The individual must have
complete, simple control over what he shares and with whom.
* Groups must have joint control over who is a part of the group. The group
must be able to remove anyone who "doesn't play nice with others".
* The tools must stay out of the way as much as possible - provide maximum
benefit for minimum extra energy. The benefits should arise while doing the
work that needs to be done rather than because of doing extra work.
* The tool must provide retrieval of the information and its relationships
is as many ways as can be done easily enough to justify the work. (013)
I had started to develop design elements, but decided that this is far too
early in the cycle to be doing that.
Summary:
I assert that:
* Any proposed solution to any problem that is not based in reality will not
work, no matter how many other supposed merits it may have.
* "Solutions" are produced by individuals with common personal interests
working in small groups with an intention to produce a workable result.
* To be effective, any tool to augment human intelligence must support the
individual and the groups that he chooses to join because of his own
personal interests.
* The OHS group presumably constitutes or contains at least one such group.
* We need to start by building tools that will support the concepts given
above for ourselves, if only because "you must operate where you are, you
cannot operate where you are not".
* If we who have an intense interest in such tools cannot agree on what
should be built *for ourselves* and get it done, there is no point at all in
lamenting the "complex problems" that remain in the world and the fact that
we don't know how to build tools to solve them.
* If we do build a set of tools that aids us in collaborating on the
"complex problem" of building a set of tools with which we are (mostly)
satisfied for ourselves, we will have made a significant, and perhaps the
only possible, step toward tools that help in solving the "complex problems"
of the world. (014)
Thanks, (015)
Garold (Gary) L. Johnson (016)