On Fri, Mar 04, 2005 at 11:49:59AM -0600, Bob Jamison wrote:

Yeah... like give the main page a name like index2.php or whatever, to get it to work correctly, and play nicely with PhpWiki.

*Nod*

Over the past couple of weeks, I have learned to like Wordpress a lot. Don't forget the other benefits, like a searchable DB of articles, auto RSS feeds, etc.

I guess I can see the value for RSS feeds, but I can't imagine a searchable db to be a feature that people would really need... If so, with the current system you can always just load the archive.php page and search it via the web browser or whatever.

Also, I might qualify the skill level as pertaining to -current- contributors. Having a (non-techie)-friendly news engine could encourage new contributors who might be a bit intimidated by the current structure.

This has always been the argument made for adopting a CMS that I've seen in the past, but in my experience it is a bad rationale, for the following reasons:

* First and foremost, the limiting factor is not technical know-how, but the dedication to regularly post news.

* Second, if a system is working adequately, and it is changed for the purpose of getting involvement from imaginary people, this is a high risk. There's no way of being sure those people will appear!

* Third, it assumes there are non-technical people who won't contribute to the site unless things are "easy". In my experience, people can and *do* learn, especially if a process is documented and involves technology that is reasonably standard. This is especially true in open source projects, where learning cool new things is sometimes a big motivator to begin with.

* Simply _writing_ news requires technical knowledge. Technophobes are probably going to be challenged even being able to understand what developers are talking about and why it is newsworthy.

* Geeks that are motivated enough to *want* to update a project's website are probably also motivated enough to learn html, cvs, etc. Probably they already know it anyway. If they're not, then they may also not be motivated enough to learn a CMS.

* CMS are not magic. They have quirks, take some time to learn, and sometimes have their own limitations. There is no guarantee that adopting one would actually make things easier. It's also possible it could add complexities.

Basically, avoid the trap of designing for the lowest common denominator on assumptions alone, especially in technical projects like Inkscape. There's no guarantee that the non-technical people would actually participate anyway, and it risks alienating the technically motivated people.

CMS's can sometimes backfire, in that why they can make easy tasks easier, they can simultaneously make hard tasks harder. As an example, consider if you needed to make a global change throughout all past news articles. In the current flat file system, you would load the index.php and archive.php into text editors and do a search and replace. Any one with a basic understanding of computers should be able to handle that. Or you could use a cmdline tool like 'replace', or a sed or perl script, for more complex changes. I'd suspect most developers in this project could handle doing something like that.

In a system where the news items are stored in a database of some sort, now you need to be able to understand SQL, and create a set of query and update scripts to locate the records needing changed and update them; this could involve several fields or tables. You'd probably also want to replicate the database to test your script out, which requires some administrative knowledge of the database. Probably you'd find it easier and faster to just manually load each file in the site and convert them by hand.

The above might not be a *common* thing that we'd need to do, but should illustrate how adding technology to "simplify" a process could actually cause things to become more complicated. I've seen something like the above in *every* CMS conversion I've been a part of, and this was a strong reason why I deliberately set up the news in Inkscape to be so trivially simple-minded.

Besides, it's a fun toy. That's why we're here, is it not? ;-)

Heh, well I've satiated my desires to play with web tech a while ago. ;-) To me, the news is simply a tool to keep the community informed and to illustrate that the project is alive and progressing well. The important things to me is that it gets updated with regularlity, and that it doesn't take undue time or frustration for me to do it. However, if a change can ensure that others participate with it, that's great.

Bryce

On Fri, Mar 04, 2005 at 05:53:05PM +0000, Jonathan Leighton wrote:

Okay, I understand what you are saying here.

I probably would not have raised it if it wasn't for Bulia posting about RSS feeds, but from my point of view, as a user of WordPress already, it seems to make a lot of sense.

I'll see about creating some sort of rough prototype of how it could work. It shouldn't take more than an hour I don't think.

How are the current posts stored? In a database?

Flat files in CVS.

Bryce

On Fri, Mar 04, 2005 at 01:30:26PM -0500, mental@...3... wrote:

...

I've been through several website adoptions of CMS's (i.e. Zope) that turned out very badly, so am rather skeptical of the value of CMS systems in general,

Isn't Zope a web application framework rather than a CMS as such?

It has a CMS component called Plone. We have found that it has serious performance limitations, and does not play very well with non-Zope things (such as CGI scripts or PHP). Plone gets a lot of press, but it doesn't seem to be a very good solution, even if you're willing to go 100% Python for everything (as it requires). It cannot handle high traffic loads, for instance - we tend to lose the OSDL site for the day when we even get a modest slashdotting.

Bryce