Glynn Foster wrote:
Glynn Foster wrote:
Userspace Utilities (compiler, archiving, editors, ...)
This is a difficult section to complete, and very much maps together with the
modernization/familiarity issues. I think the answer is close to what the SFW
consolidation (including PSARC cases in the pipeline) already contains, less
some components that don't really seem to be a good fit with the 'basic
operating system and desktop environment' mantra.
Is there anything off this list we're missing eg. locate? Is there anywhere we
can add more value to promote some of the top features of OpenSolaris eg. Zones
or ZFS, and bring them more user visibility?
I think locate (actually, the old fast find) is a very very long standing
RFE (on the order of 11 years).
As someone else mentioned, lsof has features that would be good to have,
although its implementation left a lot to be desired, last I looked at it
(and I recall the arguments that were on -code, months ago, they don't need
to be rehashed).
It might be also worthwhile, at this point, to bring in the Familiarity section
of the problem statement -
3.3) Familiarization
FAM-1: Provide a set of packages in the default install
and network repositories that users would typically
expect to see with a similar UNIX operating system
eg. GNU/Linux.
That's very vague, because it's so hard to gauge "What people expect", it
varies based on the value of "people". (the longer you've used something,
the more you come to expect precisely it, and nothing else, as I think we've
all seen in prior arguments about indiana).
FAM-2: Provide an environment such that migrating (particularly
GNU/Linux) users will be familiar with the utilities
available, while preserving the binary compatibility
story that people want and expect.
That's less vague, but there's a basic subset of features we lack that are
common to basically every Unix like operating system, other than us. (I'm
sure everyone has their own list of such things, and I'd bet they coincide
to a large degree too), search b.o.o for the modern_times keyword, there's a
reasonable list right there (and at least a handful that should probably
have the keyword, but don't).
I would expect with those added, "familiarity" would be largely there already.
In general, as I've said before in conversation with various people
(probably over and over again...), I think while users may think they want a
copy of GNU frob, or whatever, what they actually want is the feature
(maybe with the same, or a largely similar interface, maybe not). I don't
think it makes sense to make drastic changes (or however many different
copies of the same basic utility) when minor feature addition would have the
same result.
With PSARC 2007/047 /usr/gnu, various GNU userspace utilities have been included
into /usr/gnu where there is already an existing Solaris utility. Is this
enough? Is a GNU branded zone what we want, or better still to add functionality
to the existing Solaris utilities in a compatible way that appeals to 90% of
what's perceived as missing?
I don't have many answers here - I'm pretty certain we're a lot closer with the
recent work that's been done. Is closer rather than better where we need to be?
The goal here should be making things better, NOT "the same as $FOO", for
any value thereof.
There's many commonly missed things that (I'd hope and assume) were seen as
beneficial. In general, better should be the target, closer is a nice perk.
"Closer to $THING" should never be a goal in and of itself, you can be very
similar to something awful, that would hardly be anything to brag about
though...
-- Rich
_______________________________________________
indiana-discuss mailing list
[email protected]
http://opensolaris.org/mailman/listinfo/indiana-discuss
