Re: [kDev] kendraTools: information structure...

[Neil Harris <neil.harris@xxxxxxxxxxxxxxxxxxx>]

Daniel Harris wrote:

> Hi Neil and All,
>
> Thanks. I took a look at the Wikipedia page...
>
> At this stage of kendraTools I see no reason to define a strict 
> language and probably never to do so. Each application will make it's 
> own definitions; have their own input restrictions and search 
> criteria. Have a look at the address application (?) that I've just 
> coded for kendraTools:
>
> http://www.kendra.org.uk/tools1/addresses.php
>
> It's very simple (and matches my coding ability) but I hope it shows 
> that we can data mine this mass of relationships any way we like.
>
> Also, take a look at some other ideas for links that other 
> applications may include at some point:
>
> http://moof.org.uk/forum/index.cgi/DevelopmentPlanBook1Links
>
> So, I suggest that we concentrate on the applications and make sure 
> that the "information store" can hold as many different types of 
> information, constructs and relationships as possible and yet still 
> have a simple structure.
>
> Does this make sense?
>
> On the is/was are/were constructs: those are to do with time and 
> perhaps it's better to have an additional relationship to these object 
> stating when this relationship took place...
>
> Cheers Daniel

Dear all,

Here are is a distillation of some recent thoughts, based on discussions 
yesterday.

--------------------------------------------------------------------------------
The software

0  The system is not designed to solve a single Kendra project: it is a 
framework for making it trivial to generate solutions for Kendra 
projects in general.
0a The goal can be described as "bootstrapping the semantic web"
0b More concretely, this means hosting collaboration fora for describing 
things and facilitating exchanges based on these descriptions, from many 
points of view, keeping the diversity of views that is central to the 
Kendra concept

1  The data store stores _everything_ under a name which is a Unicode 
string (the "Universe of discourse")
1a [there will also be simple namespaces to deal with special / new data 
types]

2 these things can be (initially)
* structured data objects with fields made from simple types (including 
names of other objects)
* relational statements of the form "x relation y"
* English or other language plain text comments
* page/report templates -- definition by example, etc.
2a Some operations are restricted to admins for the time being.

3 The system will initially support very simple syntax for declaring 
templates for new kinds of object, and creating and editing objects.
4 Every user has an account.
5 All assertions are tagged by which user asserted them.
6 The system is designed to emit W3C-compliant Semantic Web pages, as 
and when that is well-defined.   (Just like a Wiki serves standard HTML, 
as well as supporting its own internal format.) These pages can be 
exported as standard Web pages that will appear to be part of the static 
public Web. 
6a RDF export will probably suffice initially.

7 Nice web design will need to be added for credibility among general 
user community. (So we need some nice artwork etc, but in the short term 
good clean HTML 4.01 + CSS will do for development).

8 Full source + database dumps to be available on servers, so mirrors 
can be set up by anyone.

9 Need to be able to import data from public domain sources (NIMA 
database etc), peer with copyleft content (MusicBrainz etc.), and allow 
copyrighted data owners to participate, without giving up the rights on 
their data...

Reference points:
* Wikipedia (easy editing, simple namespace, 70,000+ encyclopedia articles)
* Mindpixel (currently 1,000,000 very simple assertions)
* Open Mind Initiative (doesn't say)
* Commonforum (based on MoinMoin wiki)
* MusicBrainz, IMDB...

See also: http://www.wikipedia.org/wiki/Wikipedia:Size_comparisons

-----------------------------------------------------------------------------------
The server cloud

A migration path for the software...

1  single server, running a single copy of the server
2  multiple cooperating servers, running on multiple boxes, run by 
Kendra, as proof of concept
3  as 2, but with servers run by other trusted organizations, with a 
central Kendra "mothership"
4  allow anyone to act as a content peer? (requires self-governing 
community critical mass to manage potential problems)
5  the "mothership" becomes unneccessary.

-----------------------------------------------------------------------------------


The data

1  different licences may be needed per project
2  what dump format to support?