Hung up on wikis

[Corey Burger]

While hanging out on #ubuntu-motu today, Mark Shuttleworth mentioned 
that he is trying to get the wiki standardised on moinmoin markup. I
mentioned that the doc team was meeting this weekend about that as
well.

Corey


On Fri, 11 Mar 2005 11:12:30 +0000, Matthew East
<matthew.east at breathe.com> wrote:
> Right glad this mess has been sorted out.
> 
> On the subject of why the wiki engine is rubbish, and needs to be
> replaced, I will add my support.
> 
> > I personally favour a completely new wiki for one major reason: If
> > anyone has looked at recent changes on any wiki, you see a lot of
> > chaff or noise. If we have a wiki *just* for presentation quality
> > docs, then the *only* thing that will be appearing is doc related and
> > thus much much easier to quality control. I cannot underestimate how
> > much effect this will make upon the quality of our work. Coexisting
> > within the current wiki would cause major issue.
> 
> I don't think this is absolutely fundamentally important, but it is a
> good idea. There are about a million wiki pages, and lots of the
> discussion ones get confused with the documentation. If this were not to
> change, I would not be gutted, but i agree that it would be an
> advantage. One side effect might be that referencing between the two
> concepts would be more difficult.
> 
> As I understand Corey's suggestion, things like DocTeam discussion would
> be left in the existing wiki, whereas things like howtos would go to a
> new documentation based wiki. This might cause problems with linking. An
> example would be the DocBook wiki, which is both documentation, and
> related to the team's work.
> 
> I think the point is that the current wiki is only confusing because of
> the structural problems it has.
> 
> > Now on the technical side of things:
> >
> > 1. Remove parenting
> > -This concept is too rigid, but a nice idea
> > -We can manual parenting and categories to achieve the same result
> > with more flexibility
> 
> Couldn't agree more. It is precisely this which means the wiki is a
> sprawling mass of pages, most of which are in the wrong place.
> 
> > 3. One markup language
> > -Probably the biggest issue. The 4? currently supported are confusing.
> > Another reason for moving to a new wiki. If we force a move to one
> > language, every page will have to be checked to make certain that it
> > is in the new/chosen markup
> 
> OMG it is terrible. THe most dodgy things for me include
>  * The fact that sometimes random question marks creep into the middle
> of words
>  * The fact that the nesting of titles is extremely difficult.
> 
> No doubt there are solutions to these and all other problems, but they
> are by no means obvious.
> 
> > 5.Better history
> > -Contrast this:
> > http://en.wikipedia.org/w/index.php?title=Shanghai_Pudong_International_Airport&curid=354666&action=history
> > with this:
> > http://www.ubuntulinux.org/wiki/FrontPage/diff
> > -Ouch is about all I can say
> 
> Yes the history is pretty abysmal.
> 
> > 6. Wasted space on page
> > -The current wiki wastes a lot of space. I run at 1280x1024, which is
> > much higher than most people. Even so, the actual text begins 1/4 the
> > way down my screen, once you take into consideration all the panel,
> > menus, etc.
> > -This is kind of out of our hands. Another reason to move wikis
> 
> Doesn't bother me, even on 1024x768.
> 
> For me the biggest issue was not mentioned by Corey but was by Enrico.
> The cache problem is so frustrated. First, it doesn't remember your
> login, even with firefox's password manager. Ok, fair enough, some sites
> don't. But secondly, even when you log in, and refresh the page, often
> it still remains locked and has some crazy padlock signs all over the
> place. You have to navigate away from the page and back again in order
> to edit. Sometimes it works, sometimes not.
> 
> Grrrrrrrr
> 
> /me chews at wiki
> 
>