On 13 February 2012 22:07, Stanton McCandlish smccandlish@gmail.com wrote:
On Sun, Feb 12, 2012 at 4:34 AM, Pavel Tkachenko proger.xp@gmail.com wrote:
I even think templates should only be created by devs after discussion, otherwise it results in what we see now.
You can't be serious. Half the useful work at I do at en.wikipedia is template coding, to do all sorts of useful things. If I had to apply though some anal process of developer supplication to get something like {{Shamos 1999}} or {{Rp}} or {{Cue sports nav}} set up, I would simply quit the project. I don't mean that in a histrionic WP:DIVA way, I mean it would just frustrate me too much to not be able to fix things, expediently, that need to be fixed and which have to do with usage of the tools and writing a wiki, not with the underlying coding of the tools themelves. It's exceedingly presumptuous of developers of tools to suggest they should be in control of how people use the tools and work internally around their limitations. Which are often severe (see {{bq}} and {{gloss}}, especially the latter's subpage of test cases showing how awful MW's parsing of definition lists is when not overridden with explicit HTML). Note: All of what I'm referring to is at en.wikipedia.org.
Yes. The fundamental reason people do unspeakable things with ParserFunctions is precisely so they don't need to go through a gatekeeper. This is a feature, even if the ParserFunctions syntax isn't.
(Every domain-specific language that is allowed to become Turing-complete turns into brainfuck.)
- d.