deprecated because: Development has ceased??? Maybe development is *complete*

[Matthias Andree]

Am 10.09.2011 07:35, schrieb Conrad J. Sabatier:

> Well, I'm certainly willing to do what I can, for as long as I can.  I
> maintain a handful of other ports, so I'm not unfamiliar with ports
> maintenance.  As long as I'm capable of doing so, I'd be glad to.  If
> at some point, some change in the base system or ports renders further
> maintenance extraordinarly difficult or impossible, well then of
> course, I would have to relinquish those duties and let these two ports
> climb the stairs to the Attic.  :-)

Thank you, and good speed with your new ports. :)


> Well, I' sure you know that installing from source "by hand" is often
> much more difficult than using ports.  All sorts of odd little "road
> bumps" often crop up that have to be dealt with, and many users simply
> may not have the necessary skills.

And if they don't have the skills to self-support such installations,
they shouldn't be using dead software, because they must be able to rely
on us for some support.  For these users, it's better if they find
themselves another ports that fits their purpose.

> That whole area still just seems rather fuzzy and grey to me.
> Opinions as to what constitutes "support" seem to vary widely.

It's a fuzzy term indeed.  The point is avoiding making promises (by
packaging dead ports) that we can't live up to -- and for me, it was an
implicit given all the time that "we're talking about unmaintained ports".

> My *personal* feeling is that as long as a port continues to build and
> run and doesn't require any modifications to other ports in order to
> do so, and has no known serious vulnerabilities that would render it
> truly dangerous to use, then we should try to keep it around (yes, even
> if it means we have to host the distfiles(s) after the original site is
> gone, which I know many would disagree with).

I think that I could live with, although I'd extend the restriction
beyond just serious vulnerabilities, but those would be criteria other
people would subscribe to (such as critical bugs, like corrupting data.
According to Debian's definitions [1], it'd be somewhere around
"serious" or "grave" bugs.

[1] http://www.debian.org/Bugs/Developer#severities - note the site
negotiates language with your browser, be sure to configure the latter
properly.

> Again, just my personal feelings on the matter.  Having dabbled with a
> number of Linux distributions, I feel very strongly that the ports
> collection is one of FreeBSD's strongest assets (relative to Linux), and
> that we should strive to keep it as "complete" (for lack of a better
> word) and rich and diverse as possible.

Yes, but my personal feeling is "not at all costs".  If the trade-ins to
be made in order to have a port in get too large, time to reconsider
inclusion.  Which is what has happened more extensively than in the
past.  When I made first contact with FreeBSD (4.X with 3.X still
popular), it was just short of 6,000 ports. That has more than tripled
in the years since.  That is a humongous amount of software, and even if
we drop 1,000 ports, that's less than 5% today.

Of course someone's favourite may be one of those 5%, which hurts and
gets complaints, but in perspective it doesn't look that bad.

> While I haven't done an extensive search for alternative language
> translation software (truth is, I was in a bit of a hurry to get
> something useful installed as quickly as possible, and ksteak seemed
> the most appealing), I do think ksteak is one of the most pleasant to
> use, and also offers a fairly rich set of translations compared to some
> others that I've tried in the past.  I'd really hate to see it go
> before it really is necessary.

Looks like these two guys (steak and ksteak) didn't have sufficient
momentum to be ported over to Qt4.

> So, yes, I will officially volunteer to take over as maintainer of
> these ports.  I'll send-pr them shortly.

Thanks a bunch for helping us!