Information Life Cycle

[Paul O'Malley]

Hi,

Proposed patch that might help fixing bug 1.

We actually have to realise that users have needs, and users need to 
know where to get those needs serviced.

The Wiki does one set of tasks.
help.ubuntu.com does another
the forums do another.
IRC does another.

Recently several people on this list from mdz through to corey, mdke and 
others have all said something like this:
"Good information for users, helps users get what they need to achieve 
their goals."
Sourcing that information is the thing.

On IRC we have a bot which points in some cases to the wiki, some cases 
to the forums, and also the the help.ubuntu.com, and other locations.

I do not have all the answers as to how to achieve all this, however 
here are some of my thoughts.

We need to get or create a tool that can create a "meta database" of all 
info on all of the information propagation methods we have.

This tool effectively becomes the information front end to all things good.

Imagine our own search engine, it does the docs, and how does it choose 
the keywords, it uses section headings.

With this index in hand you can tag the urls that are on the wiki.
A compare and contrast exercise can then take place and where 
appropriate the wiki page (say for universe synaptic or universe) can be 
given a "URL Stamp" at the top which says for example, 
http://help.ubuntu.com/6.06/ubuntu/desktopguide/C/extra-repositories.html#AddingExtraRepositories  
for the version.

This means that information can be honed to the point where supported 
information get higher ratings, and dirty hacks get left behind.

However how do you define what information you want.
Perhaps something like this http://ubuntu.cc.com.au/pop.php which tells 
you what people most ask about could be used to help inform the decision 
being made.
Regardless of the accuracy of the actual factoids (which I will deal 
with in a moment), this list means that most people have at the moment a 
request for
https://wiki.ubuntu.com/RestrictedFormats

This document has also been the subject for discussion on this platform.

(I do agree with the idea of one lead and several smaller ones, but that 
is a separate issue.)

The next thing that needs to be done is that the searches on the wiki 
should become part of the information process.
What are people looking for?
How do we address those needs?

This is one reason to allow people search "help.ubuntu.com".
Again search logs become part of the information life cycle process.

The key to all this is to define the keywords words, and create an index 
of them.
The quality of the information is for another day.

(I hope this is not too hard to parse.)

Regards,

Paul