Re: Multiple Major Modes

[Joe Brenner]

Lennart Borgman <lennart.borgman@gmail.com> wrote:
> Joe Brenner <doom@kzsu.stanford.edu> wrote:
> > Lennart Borgman <lennart.borgman@gmail.com> wrote:

> >> >> >> to test it. (BTW, what file extensions do Mason files normally have?)
> >> >> >
> >> >> > The currently recommended Mason file extensions are *.mhtml for 
> >> >> > internal
> >> >> > components, and *.html for external ones:
> >> >> >
> >> >> >   
> >> >> > http://www.masonhq.com/?FAQ:ServerConfiguration#h-what_filename_extensions_should_i_use_for_mason_components_
> >> >> >
> >> >> > That means, of course that the file extension alone gives you no way 
> >> >> > to
> >> >> > distinguish between a plain html file and a top-level Mason file.
> >> >>
> >> >> Thanks. I am a bit surprised by the second convention.
> >> >
> >> > The reasoning seems to be that the client is asking for html, and that's
> >> > what you're going to deliver to them... the fact that Mason is used to
> >> > generate the html is an internal detail there's no point in bothering
> >> > them with.
> >>
> >> Hm, I see. It is maybe a pretty good reason on that side. But do
> >> clients really care about the URL "file extension"?
> >
> > Well, I do.  If the URL ends in ".asp", I think "oh, no...".

> Oh, you are not the type of client I was thinking of... I am glad to
> hear you care though ;-)
>
> I was more thinking about web browsers.

No, not web browsers, exactly.  But a common use case for Mason is
re-engineering sites that were developed statically and turned out to be
maintenance nightmares.  There the published interface is URLs ending in
*.html, and changing that because you're changing the back-end wouldn't
make any sense.