On Tue, 7 Jan 2003, Derek Simkowiak wrote:
> 1. A driver is "configured", not "defined", in the config file. I.e.,
> changes like:
Except that you can have more than one "instance" of the same driver, so
in some sense, the config file "defines" how many instances you are
using. You can have several routers that use "driver=redirect", for
example. So you aren't just configuring *the* redirect router; you are
configuring *a* redirect router.
> "Within a driver definition, there are [...blah, blah...]"
> "Within a driver configuration, there are [...blah...]"
... but yes, that still makes sense.
> 2. Underlying drivers and a particular config thereof are not the same
> thing. I.e.:
>
> "[Section Title] 3.8 Running an individual router"
> "[Section Title] 3.8 Running an individual router config"
That doesn't make sense to me. It is not the config that runs, it is the
code of the driver. So I suppose really, it is "running the code of an
individual router with a specified configuration". That's a bit long...
> This kind of consistency throughout the manual would have
> prevented my particular "instance" of confusion. :)
Consistency is always hard to achieve, especially when a document keeps
getting edited again and again over the years. Thank you for making all
these points; I have noted them and will take them into account when I
next update the manual.
I don't want to change the terminology too much, because the about-to-
be-published new book uses similar terms.
> Finally, once we decide on the correct terminology, I would be
> happy to go over the manual making corrections. If the documentation
> maintainer will simply email me privately as to the file formats and
> processes, I can start submitting diffs (or whatever).
Thank you for the offer. The best format to submit document changes to
me is diffs on the spec.txt file, because the base source of the manual
uses a private markup notation from which all the various document
formats are generated. Let me see what I can come up with when I do the
next edit (probably not for a couple of months) and then you can suggest
changes to the new edition. (A lot of documentation changes are piling
up - not only for new features - which is why working on the current
document may not be the best approach, as I am planning to do an
extensive revision.)
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
