Here's the last installment of my comments on the existing policy
document....
11.2, describing .la files:
"[they] contain a lot of useful info ... (e.g. dependency
libraries for static linking)"
Would "dependency information" be better?
11.2, penultimate paragraph reads:
Packages that use libtool to create shared libraries should
include the _.la_ files in the _-dev_ packages, with the
exception that if the package relies on libtool's _libltdl_
library, in which case the .la files must go in the run-time
library package. >>>This<<< is a good idea in general, and
especially for static linking issues.
What does the indicated "This" refer to -- that packages should
include the .la files in the -dev or run-time package?
11.3 The explanation of soname is wrong; for example:
polya:~ $ objdump -x /usr/lib/libxml.so.1.8.11 | grep SONAME
objdump: /usr/lib/libxml.so.1.8.11: no symbols
SONAME libxml.so.1
So this paragraph needs rewriting somehow.
11.7.5 What does the following mean?
However, programs that require dotfiles in order to operate
sensibly (dotfiles that they do not create themselves
automatically, that is) are a bad thing, and programs should be
configured by the Debian default installation as close to normal
as possible.
(It's the last part I don't understand.)
11.8 Logrotate: should it be a policy directive ("packages should
rotate their logfiles using logrotate") and written in a more
formal style?
11.9 There's a paragraph about changing permissions and security
policies (beginning "You must not arrange that the system
administrator..."). Is this any longer true now that we have
dpkg-statoverride?
11.9 Statically allocated ids:
If you need a statically allocated id, you must ask for a user or
group id from the base system maintainer, and must not release
the package until you have been allocated one. Once you have
been allocated one you must make the package depend on a version
of the base system with the id present in `/etc/passwd' or
`/etc/group', or alternatively arrange for your package to create
the user or group itself with the correct id (using `adduser') in
its pre- or post-installation script (the >>>latter<<< is to be
preferred if it is possible).
What is the "latter"? Is it the latter alternative ("or
alternatively ...") or the postinst instead of the preinst? I
would guess that it means "postinst is preferred to preinst", but
I may be wrong here.
12.1 The list of arches is probably out of date. Maybe policy shold
not be so directive here, perhaps referring to the output of
dpkg-architecture -h?
12.2 The last para reads:
If a package wants to install an example entry into
`/etc/inetd.conf', the entry must be preceded with exactly one
hash character (`#'). Such lines are treated as `commented out
by user' by the `update-inetd' script and are not changed or
activated during a package updates.
This isn't very meaningful as it stands. Either the whole
paragraph should be removed or a better explanation of what it's
talking about should be given.
12.5 3. Web document root and web applications
Pardon my ignorance, but what's a "web application" and what are
examples?
12.6 "Mailboxes are generally [mode] 660 user.mail unless the user has
chosen otherwise."
Should this be "unless the system administrator has chosen
otherwise."?
12.6 All MTA packages must include a newaliases program, so there
should be a para reminding that all MTAs must Provide, Conflict
and Replace mail-transport-agent.
12.6 last para:
check for the existance of /etc/mailname: "If it does not exist
>>>it<<< should prompt the user ...". What is "it"? I think
it's probably the pre/postinst.
12.8 There's a footnote (7) which says "Rationale: clarifies the
language..."; surely this shouldn't be in the document!
Julian
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Julian Gilbey, Dept of Maths, Queen Mary, Univ. of London
Debian GNU/Linux Developer, see http://people.debian.org/~jdg
Donate free food to the world's hungry: see http://www.thehungersite.com/
