tech-userlevel archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: File types [was: Re: wide characters and i18n] wrote:

> Well, what *I* use xattr's for is:
> o providing MIME type and, in the case of plain text files (whatever "text
> file" means today), charset information;
> o user tags to make indexing and searching easier.

Makes sense to me.

> > If you can figure out how to push it down to the file system
> > level, I think you'll be doing well; I don't think you can,
> > realistically.
> That doesn't mean I can't dream about it happening automagically the way I
> want it, right?

On the contrary, if you can figure it out, I'm all for it! I
just don't think it's doable, but I've been wrong before, and
I'd like to be this time.

> You have here a good argument in favour of making apps xattr-savvy.

Well, character sets make for a reasonable argument for
associating meta-data with files.  But files + directories have
the same theoretical power as files + extended attributes, so
there's some unnecessary complexity introduced.

If I were to bet, I'd bet that extended attributes are here to
stay.  But without standardisation (de-facto or de-jure)
across file systems and OSes, their use is going to remain
limited, often application specific, and sometimes OS
specific.  (Work creation schemes for programmers?)

The folk are creating recommendations for the
use of extended attributes:

Maybe a miracle will happen and Microsoft and Apple will join
them.  Or maybe history will repeat itself and Apple and MS
will product conflicting, multiply versioned recommendations
that are semantically different enough that they can't be
mapped 100% back and forth.

I think I've exhausted everything I've got to say on this
topic now, other than mentioning again that anyone thinking
about MIME types and UTF-8 might want to consider including
normalisation information if their file is in a normalised
form. :-)



Home | Main Index | Thread Index | Old Index