forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ross Gardler <rgard...@apache.org>
Subject Re: Wiki
Date Mon, 13 Nov 2006 20:07:57 GMT
David Crossley wrote:
> Gav.... wrote:
>>> From: David Crossley
>>> C. Grobmeier wrote:
>>>> Gavin wrote:
>>>>> Here is one such thread :-
>>>>> http://marc.theaimsgroup.com/?l=forrest-dev&m=112277290808391&w=2
>>>>> Can't seem to find the one where I was involved at the moment,
>>>>> Will have another look later.
>>> Thanks. That links to some of the past discussion too.
>>>
>>> We need to review those threads, some great ideas evolved.
>>>
>>>> Thank you very much. I read it and understand, that you don't want a
>>>> wiki cause it's to early for this project. You fear that docs are in the
>>>> wiki but not in the core docs, that the PMC can't oversight the entries
>>>> and so on.
>>>>
>>>> I see a wiki more as a sandbox. It contents small artifacts of users who
>>>> don't want to get involved in the mailinglist or of users who don't want
>>>> to start creating a patch. Sometimes a wiki makes things easier. It
>>>> works for Commons and i think for Struts too.
>>>>
>>>> But i -really- don't want to bring this old discussion back to life
>>>> cause i am new to this project and hope to see your reasons more clearly
>>>> when i am deeper into it. I just was curious cause many projects use a
>>>> wiki but Forrest don't.
>>> Well one reason is our history and purpose. We are
>>> our own documentation system. Open your text editor,
>>> tweak our doc sources in site-author/ and review
>>> them with 'forrest run'. However we have received
>>> very few documentation patches.
>>>
>>> The above linked to an important proposal discussion
>>> which has not yet been implemented. Either use Lenya
>>> or piggyback on Cocoon's Daisy.
>>>  [PROPOSAL] A CMS for our Docs
>>>  http://thread.gmane.org/gmane.text.xml.forrest.devel/13450
>>>
>>> I glanced through that again. If someone could help to
>>> summarise it and make a new proposal, that would be
>>> fantastic.
>>>
>>> The Forrest project seems to be gaining some new energy
>>> these days, so perhaps we can move forward.
>> It was a very long group of threads that started off as a proposal
>> For Daisy to become the CMS of choice, it turned into how can we
>> Make Lenya + Forrest work together, Lenya as the CMS of course, then
>> Forrest would publish Lenya published (committer approved) docs.
>>
>> Hows that for a summary ? (er, only joking :) )
>>
>>

>> We have DOCO (DOCU ?) although I forget how far this has got and even
>> If this was its purpose.
>>
>> A summary would be good and I'll try and do one shortly, but as for making
>> it a new proposal, I am unsure as to what to propose, does it need yet
>> Another discussion considering advances in Lenya, Forrest etc have changed
>> How we might go about such an integration. The thread obviously chooses
>> Lenya for the CMS part, are we happy with that or do we need to go through
>> Some more choices again? Are Lenya in a position to be able to help again
>> Where necessary?
> 
> I waited for a while to see if some other people would comment.
> None, so i guess that means that people feel the same way as
> me about this topic: not enough energy for integration exercises.
> It sounds to me like there are not enough active developers here
> to enable it.

Sorry for my delayed response - been really busy in none Forrest things 
lately. Let me share my experiences and thoughts...

First of I wrote an HTMLEdit plugin that allowed Forrest sources to be 
edited from within a Forrest webapp. This worked really well and I used 
it quite a bit. But it was never community supported and I outgrew it 
and needed a CMS.

I then wrote the Daisy plugin, it works pretty well, in fact Cocoon 
currently uses it for their docs. However, it never attracted any 
interest from a wider community. The Daisy project were not interested 
as they didn't want to be dependant on Forrest for their static 
publishing. Cocoon adopted it, but have never contributed to its 
development and they now appear to be moving to a Maven based build 
process (including docs).

Another CMS related solution, although specifically for educational 
materials is the Eclipse plugin in the whiteboard. This code was brought 
over from my Burrokeet project which integrated standards compliant 
learning content into a Forrest generated site. However, I moved left 
this project a couple of years ago and very little has happened on 
either Burrokeet or Forrests Eclipse plugin.

There are devs here who are also devs at Lenya, hence the proposal of 
creating a Lenya plugin. Unfortunately nobody has found the time to 
realise that objective.

So, current integration of Forrest and a CMS is as good as dead in my 
opinion. But that doesn't mean it can't be revived there are plenty of 
good examples of how to go about it in our SVN history, of course these 
also contain examples of mistakes to avoid).

> If you want to pursue this, then perhaps just investigate
> a small part of it, e.g. one suggestion was to ask Cocoon
> if we could have a small section of their Daisy to use
> as a doc scratchpad, but leaving our main documentation
> as it is now.

This can be up and running with minimal effort since we already proved 
it works within Cocoon. But there are problems with this approach, 
performance being one of them (I can explain the technical reasons for 
this if anyone wants to pursue it, lets have a new subject for this though).

I'd love to explore all this in the context of my RT on Forrest 2. I 
don't have an immediate need for CMS integration, but I do have had some 
  thoughts about it with respect to the problems I identify in that 
thread (many of which come from my work integrating with Daisy and other 
applications at clients).

> Even with just that, there are still many issues to be
> investigated, e.g. restricted edit access, backup of sources,
> how to maintain oversight of diffs, define the exact purpose, etc.

important considerations, much of which I have answers for ;-)

> I don't want to squash your enthusiasm, but we don't want
> you spending effort if the rest of the project is not behind it.
> So it is a good idea of yours, to explore the high-level first.
> 
> Personally i would rather work with the doc system we have now
> and concentrate on the 0.8 release.

Yes, this is the same concern I have wrt my Forrest 2.0 RT. Having said 
that, a better doc system may help address the docs issues identified by 
others in that thread.

Ross

Mime
View raw message