Re: [LogiLogi-list] LogiLogi And Wiki

In-between replying tends to make mails longer and longer, so I do not
reply to every single statement, but more so to your core issues.

> system design clearer, then I'd say fire away! But it seems there is no 
> strong contra-distinction; not only does it not really shed light on 
> your design but it runs the risk of alienating those of us who've 
> appreciated wiki for years.

That is not the intention, and I re-formulated the text a bit to
mirror this: http://en.logilogi.org/MetaLogi/LogiLogiIsNoWiki

> > Wiki-like-editing is just a nice  feature, not what it really is about.
>  
> "What it really is about" cannot be operationaly defined, nor "nice 
> feature". I'm talking about system design, not aesthetics and 
> preferences. Do you see what I mean?

To make it plain: Most of the core features of LogiLogi Manta
(peer-groups, adding links to locked texts, rating and tag-links)
can function without support for Wiki-editing at all, and to be
honest I seriously thought about not adding it in for the first
release of Manta and have all articles to be only editable by 
their author.

> "Wiki" is a very clear and unique method of enabling editing. CMS covers 

Here you seem to agree with me on the definition.

Note also that my text was meanth for people visiting LogiLogi, and
thinking 'nice, another Wiki', while that would make them miss the
point about LogiLogi. 

Words and concepts carry associations with them, and where these are 
not very helpfull or precise, different ones are needed for clarity. 
We don't call a chicken a bird all the time.

> >Yes: Extracted from my post to the Citizendium-list:
> >http://en.logilogi.org/MetaLogi/DifferentLevelsOfAudience
>  
> Oh good, thanks for this. (I hope you're keeping a set of writing on 
> your design philosophy.)

I will overtime, as questions arise (in the spirit of the 
Logi-philosophy: http://en.logilogi.org/MetaLogi/LogiS)

>    " The LogiLogiManta <http://en.logilogi.org/MetaLogi/LogiLogiManta>
>    system supports having articles for different levels of audience
>    (different contexts in general) and falling back to other levels if
>    none exists at the requested level.
>    History/Students/Aristotle will for example be a good place for a
>    student-level article on Aristotle from a historical perspective.
>    Linking from History/Students/Parmenides to Aristotle will result in
>    the link shown above as the requested link. If no Aristotle page is
>    available that is tagged with both History and Students, then
>    LogiLogi <http://en.logilogi.org/MetaLogi/LogiLogi> will look for
>    History/Aristotle. If that one cannot be found, Aristotle is used.."
> 
> As an example of what seems concrete to you but is actually abstract to 
> the reader: " Linking from History/Students/Parmenides to Aristotle will 
> result in the link shown above as the requested link" It's very very 
> hard to make concepts like that clear. Believe me, describe things like 
> microwave tracking through airspace or instructions for installing a 
> complex system there are often little ideas that seem impossible to 
> capture with simple sentences. That's part of the challenge.

In my expecience of trying to explain LogiLogi to people I found the
same problem. Now that's why I mainly try to implement it as soon as
possible, so people can actually see and use it as a working system,
which should be much less difficult for them and me...

So unless you look at the code I do not expect you or anyone to be
able to grasp the system in full detail from my descriptions... Those
are meanth to give people an idea of what motivates LogiLogi, and of
what is possible with it from a users perspective in a few months.

> /Operationally/ I have no sense of how the system achieves that, i.e. 
> look and feel and function. "supports having articles for different 
> levels of audience (different contexts in general) and falling back to 
> other levels if none exists at the requested level." ... "supports" and 
> "falling back"; these physical metaphors are only that.

Falling back means here: Doing a database-query History AND Students
AND Aristotle, if no results, reducing it to History AND Aristotle as 
tags, and so on.

> Peek my mess: http://bentrem.sycks.net/wiki/

There is quite a lot to read there, but no UML or non-philosophical
descriptions there either...

> >No, the idea is that in using Agile Development techniques and a platform  
> >that supports it (Rails), coding something is almost as fast as creating a 
> >full blown spec for it.
>  
> Ah. Good for you.
> But ... nothing like "self-documenting code"? So there's nothing for 
> someone on the outside to see?

I did and do discuss design issues with mostly face 2 face. And the 
code is not self-documenting code, but quite well documented manually...

(I will for you and others generate a nice html-representation of the
documentation inside the code later today, and put it online...)

> >http://svn.sourceforge.net/viewvc/logilogi/trunk/doc/diagram.png?view=markup
> *Woot=wooot!* Lovely! And what is it that generated the diagram? A 
> component of SVN? I'm impressed!

Not SVN, Rails Application Visualizer: http://rav.rubyforge.org/

> >I must admit that it was not Thunderbird but me. As your e-mail was 
> >formatted
> >in HTML and my mail-reader (a stone-age, Linux console based reader by the 
> >strange name of mutt) could not handle that,
>
> Oh gosh, sorry. I only started using HTML mail routinely about 4 months 
> ago ... I'm a great foot-dragger when it comes to backward 
> compatability. That's twice in the past couple of weeks it has tripped 
> someone up.

Ok, thx for switching back.

> I think I have memories of mutt. But know: I spent years using Pine and 
> Lynx. Heck, I still kind of like Pico.
> ;-)
>
> <cut out>
>
> *Gadd ... "community" is "Politics". Are we to restrict ourselves to 
> discussing awks? or abstractions?*
>
> Speaking of which (do you have connections to the Indymedia community? 
> they're quite Political, I guess) I just re-read documenation on SMS today.

No, although a friend of mine is is sometimes involved with that site.

Wybo

-- 

::Student:
- History, Informatiekunde (computer linguistics, IR, webtech) and Philosophy
- Member of the Center for Metahistory Groningen (http://www.rug.nl/let/cmg)

::Free Software and Open Source Developer:
- http://www.LogiLogi.org, innovative system for cumulative, shared commenting,
                           publication and idea sharing: Web as it should be...
- ComLinToo, a computational linguistics toolset written in Perl
- Lake (LogiLogi.org Make), a make-replacement using makefiles in pure C++

::Being:
- In the world, go figure (http://nl.logilogi.org/HomE/WyboWiersma)