[Grok-dev] Re: maintaining the Grok website

[Martin Aspeli]

Hi guys,

If I'm allowed to be a little blunt ... You need a CMS. That CMS should 
be Plone. Read on. :)

> We have some trouble maintaining the grok website in the current 
> structure.

I'm not surprised by this. Abstract text and scripts generating HTML 
went out last century. It works for the person who wrote the text and 
the scripts, but it's really hard to collaborate this way.

> Tutorials that appear on the list don't make it onto the 
> site. While the current approach served us well to get things up and 
> running quickly, I think we need to investigate how to reduce the 
> bottlenecks.

+1

> There are a number of possible solutions:
> 
> * we continue with the current approach for a while, but we appoint a 
> broader group of active volunteers that I can ask to publish something 
> online. (I'd be happy if I had help too in making sure tutorials aren't 
> lost forever :)

Such as?

> * we start using the wiki in a major way. This means we need to put in 
> clear links in the current website and then move our smaller tutorials 
> onto it. We need a group of active wiki gardeners to make sure it 
> remains up to date and to clear up any messes.

I have very little faith in wikis as a way of building public websites. 
Wikis are great for collaboration within a group of people. They are 
terrible for producing consistent, end-user oriented materials. The Zope 
wiki is the best example, but I've never seen a site that was maintaiend 
as a wiki that worked well as a public site.

For example, look at http://jquery.com - nice site, sells a clear 
message. Now click Documentation: boom, MediaWiki. This is hard to 
navigate, looks out of place and the articles there vary in quality and 
layout and style.

> * we explore the use of a CMS. I know Sebastian is building one with 
> Grok. Darryl has one too. Then we probably have enough people on board 
> with knowledge of two Zope 2 CMSes: Plone and Silva, so those are 
> options too. The latter options are attractive as they're more 
> feature-rich than the Grok-based ones.

Please don't take this the wrong way, but I think you're mad to want to 
invent a new CMS in Grok. If a great one existed already, of course 
that'd be good. Having a project that community was vested in will also 
help you maintain it and keep it up to date (like plone.org does to 
Plone, PloneHelpCenter and PloneSoftwareCenter). But this really stinks 
of NIH.

I don't buy that there's a "credibility argument" for not having 
grok.org run on Grok. Grok isn't a website building tool, it's an 
application framework. I have a whole lot more respect for a community 
that picks appropriate tools, rather than one that refuses to look 
outside their own sandbox.

I think the choice should be Plone. Plone 3 is a very strong release. 
The Plone community (which overlaps with the Zope and Grok communities) 
will support you. They will fix bugs if you find them. They want to get 
involved in Grok more directly.

But that's kind of besides the point: All you need is Plone 3 + a custom 
theme (which you can get someone in the Plone community to build for you 
from an HTML mock-up if you wish). It'll take you a week or two. Then 
you have a solid CMS that will meet all your needs, it will make it easy 
for you to keep content up to date, you get versioning, workflow, visual 
editing, consistent styling, user management, backup... and if you need 
more advanced features, like lists of components 
(http://plone.org/products), a community-contribution documentation 
section (http://plone.org/documentation), or an issue tracker 
(http://plone.org/products/poi) and other stuff ... those things exist 
for Plone and you can get them when you need them.

The use case for grok.zope.org is very similar to the use case for 
plone.org, and sites like plone.org is something Plone is really good at 
(probably because of plone.org).

I'm not saying Silva's a bad choice - I don't know it very well. In 
fact, use Drupal or Joomla if you wish. But pick an appropriate tool 
that is mature, supported and has a wide community around it. You'll end 
up shooting yourself in the foot with a custom-built solution. Spend 
your time writing documentation, working on the framework and promoting 
it, not inventing the seven thousand'th open source CMS.

/rant off

Martin

-- 
Acquisition is a jealous mistress