Boost logo

Boost :

From: Darin Adler (darin_at_[hidden])
Date: 2002-01-28 12:34:09


On 1/28/02 9:21 AM, "Bill Seymour" <bill-at-the-office_at_[hidden]> wrote:

> Not really. In standards parlance, to say that a feature is
> deprecated means "Be advised that this feature will likely
> disappear in future versions." That's a lot stronger than
> saying, "Using this feature is frowned upon by experts."

OK. Here's my suggestion. Re-read the discussion we've just had. Wherever it
says "deprecate" replace with something like "don't use because it does not
always work".

I don't think that my incorrect use of the word deprecate affects the
discussion much in any other way.

> I think we might be having a semantic problem here. Surely
> any OS needs some way for a user to uniquely identify a file.
> Whatever that way is is what I would call a path. Am I
> missing something?

Well, what you're missing is that pathname and path have specific meaning.
Documentation about Macintosh explaining FSSpec and FSRef doesn't say "these
are the Mac OS paths". It says "these are what you use on Mac OS instead of
paths".

So I can re-read the discussion we've just had. Wherever it says "path"
replace with something like "file location identifier".

I think we had two semantic problems!

    -- Darin


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk