Rick Stevens wrote:
> Regardless of your feelings about proprietary software, the vast
> majority of Windows users--through ignorance, stupidity or laziness--
> use Skype. There's no getting around it. Sorry, but that's just how
> it is. No one said life was fair or was going to conform to your
> expectations, despite how noble they may be.
By accepting that "everyone else uses Skype, so I will too" you're
contributing to the problem, because the people who want to talk to you
over VoIP will think the same because of you.
By actually REFUSING to install Skype and explaining why (also pointing out
practical issues, e.g. that 32-bit multilibs are a PITA), you may actually
get people to install software speaking open protocols, which in turn may
get their friends to do the same and so on and you may help spreading open
standards instead of proprietary lock-in.
So installing Skype is entirely the wrong thing to do.
> Taking just the excerpt you provide above, it is utter nonsense that
> those "malicious features" are _common_ in proprietary software. Some
> yes, but I'll posit that most spyware comes from some criminal
> enterprise, installed by people who don't lock down their systems. Does
> that appear anywhere in that article?
You seem to underestimate the amount of "phoning home" proprietary software
does without even telling you. Almost all proprietary software "phones
home" to check for updates, and who knows how much "statistical"
information they send while doing that. Without the code, who can check?
> No one is taking anyone's freedom away. I don't see anyone holding a
> gun to your head to use Skype. Use an H.323/SIP program and relay of
> your own choosing. Just be aware that not everyone is going to be
> available on it.
But are YOU? If not, you're forcing your friends to use proprietary
software.
> You can't IM Yahoo chat users unless you have a Yahoo account, whether you
> use AIM, Pidgin, Kopete or some other client.
But that's not a problem because Free Software can speak that protocol. If
the Skype protocol gets successfully reverse-engineered and implemented in
Free Software, it may become a worthwhile alternative (using that Free
client only), but they're doing all they can to prevent that, and so far
they have been successful at their quest for lock-in.
I'll also observe that, because of people like you who don't care about
freedom, releasing a proprietary freeware client for GNU/Linux on x86 is an
effective way to stifle development of actual Free Software, all those
people who only care about free beer will happily use the proprietary crap.
Without the binary-only GNU/Linux port, they'd work together with the
people actually caring about freedom on a Free GNU/Linux client. So you
have to question a company's motives when they release a proprietary
GNU/Linux port of their application: chances are they don't want to support
GNU/Linux at all (and they really don't, the only way to support GNU/Linux
is to release Free Software, not proprietary binaries), they just want to
prevent Free competition (by removing the motivation for a significant
subset of the community). As weird as it sounds, it's actually BETTER for
us in the long run when there's NO GNU/Linux client at all for something,
because that motivates people to write one, which will usually be Free
Software.
> There's no one program, system, or protocol that is appropriate for
> every possible use, and insisting that there is, is short-sighted to say
> the least.
But there are protocols which are not appropriate for anything, and
proprietary protocols are part of that.
If SIP is not good enough, we need to help getting a better open protocol
worked on, not use a proprietary one.
> Portable? Almost nothing binary is portable unless it uses some lame
> intermediate "virtual machine" thing like Java (which isn't a new
> concept at all...we old timers can recall UCSD P-System Pascal in the
> '80s). They were called "compiling interpreters".
Indeed, and that's another argument against proprietary crap.
That said, they could easily provide a 64-bit binary if they cared. They
refuse doing this, making their software effectively useless for 64-bit
systems. Multilibs are an obsolete hack.
> As for 64-bit, this is true. But there's a LOT of stuff (open source
> even) that don't have 64-bit binaries available unless YOU rebuild it
> yourself--and not everyone is capable of doing that.
Most Free Software you actually want to use is packaged in Fedora or RPM
Fusion, and almost all the software in Fedora and RPM Fusion Free is
available on x86_64. In fact, we actively fix it if it doesn't work there.
(For example, I did so for z88dk.)
If the software you're thinking of is not packaged yet, where's your
wishlist entry? Or even better, your review request?
The proper solution is to get the software packaged, not to try to run some
third-party binary. Users should never have to run unpackaged binaries nor
to build stuff from source themselves.
Kevin Kofler
--
fedora-list mailing list
fedora-list@xxxxxxxxxx
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines
