Re: [wmii] New Website?

From: John Nowak <john_AT_johnnowak.com>
Date: Fri, 19 May 2006 09:14:14 -0400

On May 19, 2006, at 12:56 AM, Kris Maglione wrote:

> First of all, we won't go into the amount of time I spend on
> implementing it. I'm writing this because I want an excuse to
> write code in ruby.

Fair enough. No excuse needed.

> If you have any doubts as to how bad it is not, look at the code.
> It was written as a prototype and that's what it is. It's by no
> means better than 95% of the shit out there (though it's certainly
> better than a few php wikis out there).

I was trying to pay Anselm a round-about compliment. In either case,
it doesn't matter: If you want to rewrite it, you should rewrite it.

> The wiki syntax is a more difficult issue. Look, for instance, at
> the plan 9 wiki syntax, which is designed to be as readable in acme
> as on the web. It's mostly a matter of whether the look of the HTML
> or the wiki syntax is more important. Considering the acme client
> that's already on the wiki, this is obviously a consideration.

Markdown's main goal is to be readable as plain text, i.e., it should
have full meaning even if never parsed. I don't have access to the
edit pages on the plan 9 wiki to compare. A lot of thought has gone
into Markdown. Depending on what you mean by "the look ... of the
wiki syntax", it may be worth a look. If you just mean a clean syntax
for markup, with the readability as a document in and of itself does
not really matter, Markdown is not the right choice (too verbose
mainly, perhaps too much freedom). If you want the syntax for the
wiki to double as distributable text file documentation, you'd be
hard-pressed to do better than Markdown. (I'd be interested in the
results of your efforts if you try however.)

> The revision history is something I can live with or without. I'm
> not really worried about spamming or abuse, but it would be pretty
> easy to lose a lot of information in two commits even with only 10
> or so people editing it, especially given the lack of collision
> detection. Anyway, it would be trivial to implement and it's more a
> matter of whether it's 'in the spirit of the wiki'.

As far as I'm concerned, clicking save should randomly change 0.1% of
the characters each time you do it. :-)

- John
Received on Fri May 19 2006 - 15:14:20 UTC

This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 16:05:39 UTC