[Design] Human Knowledge Database - Thinking Bigger

[Arlo]

	This may or may not have been thought of or discussed before, but I'd
like to hear what people think, or at least have people point me in the
right direction.

	It seems to me, that in a world where organizing knowledge is so
important, there really isn't a good way of doing it.  I realize
Chandler is trying to fix this, but if all Chandler becomes is a single
app that solely is useful only if people are using the program, then it
has failed.

	Chandler needs to develop a de facto method for saving knowledge.  Its
methods need to become a standard that can be used by anyone for
anything.

	I've been thinking quite some time about the state of Documentation in
the linux world.  It seems to me that it is becoming more and more
difficult to find answers to problems, or more importantly the correct
answer.  The beauty and chaotic nature of the web is that publishing is
easy and inexpensive.  However, one often finds information that is
incorrect, or misinformed.  Posted knowledge needs to be auditable as
well as organized.  More and more users find themselves going to forums
for solutions.  How many times do we need to see the question "How do I
enable my mousewheel?" asked?

	What if there was a single source you could go to for your answers?  A
place with a sophisticated way of organizing data.  Some sort of fusion
between forums and How-Tos where users are ranked and audited by each
other.  Imagine being able to find your answer and knowing that it had
been verified by people that actually know what they're talking about. 
Now I realize that there is a certain disdain for an absolute source of
knowledge, but this is where the idea of Chandler's repository comes
in.  
	Imagine having every article tagged with certain labels.  The linux
articles would have relevant distro tags, as well as architecture
specific; at an even higher level everything would be labeled linux. 
Now servers could just select which parts of the knowledge repository
they wish to mirror.  gentoo.org might just mirror things related to
linux and gentoo; xfree.org might mirror the xfree articles.  But
somewhere all the articles would be accounted for.  And when the client
searches the knowledge base, they would be presented with the results of
their query from different repositorys seamlessly.  Clients could be
local programs or web based.  If someone were unable to find their
answer they could submit a request to the community that would be
answered shortly.  False information would be marked by people, and new
information updated.  You could also have resources for every article
and related information.  
	Ideally, a complete newbie would be able to come to this source 
knowing nothing but the word linux, and find everything they needed to
know. As they read along the related information links, they would learn
everything from how the filesystem is organized, to what xfree is, and
what a distro actually means.  You might also include a sort of profile,
including more links for those that know less, and removing extraneous
information for advanced users that don't need to be presented with a
"what is xfree?" link.

	Now I realize some people may be thinking that it's no problem to
google for "enable mouse wheel xfree".  But there are a few key ideas
here.  For one thing, this will remove a lot of redundancies among the
web by creating a single location to find knowledge, making it easier
for someone just starting out.  Also the idea of organizing and labeling
data will allow sophisticated ways of organizing and presenting the
resources.  The other key is the ability to verify the knowledge through
the open source concept of many eyes and the ability for anyone to
modify.  These freedoms will be checked by the community who will be
able to change other people's work.  Unlike a wiki, you would have some
sort of point or vote system where a good article would gain more points
and thus become harder and harder to destroy.  Eventually certain
branches would be so well recognized that they would be impossible for
anyone to modify.  Also, among users you would have a point system for
intellectual thought, clarity, and the like, so a well-ranked user would
be less contested than others.

	Now this does not only apply to the open source world.  Think about
having a car mechanic organization, where people could go search for
knowledge specific to certain cars.  Or say a scientific repository of
animal data.  The creators of a branch might even be able to set
settings such as the rank needed to create a page, or to modify, or the
number of verifications needed for a modification.  On a given page you
might have a notification saying "one modification pending, awaiting 4
verifications (requires rank 7)".  If you were above that rank you could
view the modification and verify it.  I can imagine this sort of
technology encompassing all human knowledge.  It would work side by side
with what we now know as the Internet, still allowing references to the
web, but at the same time creating a cleaner and more reliable interface
to knowledge.  Scientists have wanted this sort of technology for some
time and have looked for solutions, as have every other organization
that deals with the need for easy access to and modification of
knowledge.  I think the technology behind Chandler can eventually
fulfill this need.

-Arlo