[ydl-gen] Re: pine: was: Re: JOB POSTING: Application Evangelist

[Eric Dunbar]

On 8/28/05, Danny Hembree <danny-hembree at dynamical.org> wrote:
> On Sat, 2005-08-27 at 09:32, Eric Dunbar wrote:
> > >   The track record is irrelevant. Their license prevents people from doing
> > > the right thing and this is why many distros (including Red Hat) have
> > > dropped Pine.
> > >
> > >   If I'm a distro maintainer, I don't want to *have* to wait for a third
> > > party developer if I found a security fix now. Why should I wait? Also, if
> > > the fix that I have is a feature enhancement that the UofW folks don't
> > > want or agree with, then I'm not allowed to ship the binaries.
> > >
> > > Does this ring a bell:
> >
> > At the risk of inflaming the religious sensitivities of the devout
> > open source-only bigots (I don't pull punches on this topic, do I ;-):
> >
> > I suspect that the UW team's track record is significantly better than
> > that of the typical equal quality GNU project. Plus, the anecdotal
> > comments on this list suggest that reality is very different from the
> > Microsoft-schooled FUD GNU open source advocates (I am NOT targeting
> > this at you Chris, merely at all the FUDers that I've seen post on the
> > "merits" of GNU vs other OSS)!
> >
> > Pine is hands down the most user friendly CLUI mailer available to
> > people. I first was introduced to it in 1993 and STILL today I see it
> > in use (and I use it on my server). It's a functional piece of
> > software that has yet to have been replicated under the GNU licence.
> >
> > Anyway, that's my match and ten litres of jet fuel on this topic.
> 
> The point of the GNU license is to prevent an individual or group from
> privatizing the project. That means cutting out all the folks who worked
> on it and exploiting the user/developer base that has been built. The
> University of Washington doesn't need or want GNU style developers, they
> have graduate slave labor. Other institutions have similar projects that
> are funded by taxpayers and/or captive rate payers. These institutions
> can sell off or give away the project and you'll either have to quit
> using it or pay someone license fees. The list of projects that this has
> happened to would be a good research topic. Netscape, perhaps Mozilla
> and Debian springs to mind, Fetch, CuSeeMe, ...

I don't think that you'll have to worry about that with pine. All of
the aforementioned softwares were fairly unique at the time and it
only makes sense to get a return on investment. pine is unique in that
there is no other comparable CLUI e-mailer, BUT, unlike Netscape and
CuSeeMe, is not exactly rocket science to put together. Plus, if
they'd wanted to sell pine they would've done so when their was a
market for their product.

To fail to explicitly include pine and support it is a distro doing a
*disservice* to the distro's users. I have yet to see a distro include
an acceptable alternative and it's not exactly the easiest piece of
software to install for people who aren't familiar with the inner
workings of yum or apt-get or...

PS Debian??? AFAIK Debian is GNU and has never been anything but.
Perhaps you were thinking of something else?

> The GNU license protects users and developers from this by insisting
> that the source code and all works derived from it remain open and
> available. I've found that those who rant against this are either new to
> the industry and don't know the history, or are looking to privatize a
> project and find the GNU license to "restrictive".

Unfortunately many of the "supporters" of the GNU have the zeal and
blindness of the religious -- too many can't see that GNU, or even the
whole OSS concept is not the be all and end of software development,
and, that their writings marginalise OSS (fortunately, it is the
zealots that are marginalised, though, since "news" is becoming
Fox-i-fied, it is the crackpot that draws attention to herself).

> As to the merits of one system over the other, well, thats the big
> debate of our time isn't it. I take a very practical point of view and
> avoid metaphysical arguments, what works. The ubiquitous presence of GNU
> software in every major project speaks for itself. Where would we be
> without gcc and emacs ? Do you remember the proprietary compiler wars?
> Compile many time and run nowhere used to be standard operating
> procedure.

I'm not entirely sure how to interpret this paragraph so I'll focus on
the "practical point of view... what works" and throw my own two cents
in.

>From a practical POV (with the consideration of "what works") pine
ought to be _included_ in many distros. It *works* and its licence
allows for its distribution.

> I'm not arguing against the UW and the way they are handling pine, just
> pointing out the pitfalls and trying to explain the reluctance others
> have about using it. It's almost an economic certainty that those who
> don't have the protection of GPL will eventually get bit.

That's making a pretty big assumption. As I wrote above, pine is past
its saleable prime. Yes, it's a damn good piece of software, but it's
not beyond being replicated by another team (I'm sure people would
take that dog of a software, mutt and do something good with it if
pine were to disappear). CuSeeMe was a trailblazer at its time and
Fetch was so much better than anything else you could find so it was
only natural and reasonable that their developers (the educational
institutions) receive some remuneration. There will now be relatively
little cost to maintaining pine (especially since they have an army of
low-cost grad student labour), and, it probably is worth more to UW as
a recruitment and training tool than if they were to sell it off.

PS Has UW provided any accessibility guarantees?

Eric.