Josh Cohen <jcohen@avogadro.com> writes:
> As far as whats going on here, the sites arent shutting anyone out.
> The problem here is that the linux users need to stop whining
> and get off their lazy asses to do the work that the *paid* programmers
> do (at microsoft for example).
Right. Let ten thousand formats bloom; when there's a simple, single
standard format, it's possible to have multiple independent
implementations, but not when there are thousands or tens of thousands
of formats whose definitions consist mostly of the bugs and low-level
design mistakes of the existing implementations.
Increasing the amount of programming work needed to interoperate is a
simple and highly effective way to reduce the number of viable
implementations, ultimately leaving only the market leader standing,
eliminating wasteful competition.
That's not what the Web is about, though; rather the opposite. The
point of the Web is that we don't need ten thousand formats and
protocols --- we should have one format that's good enough for
everything and simple enough for everyone, one protocol that's good
enough for everything and simple enough for everyone, accommodate
everything else as second-class citizens, and unify everything with
universal namespaces so the second-class citizens aren't any better at
cooperating with each other than the universal formats are at
cooperating with them.
> The burden is on the browser/platform advocates, not the web site
> developers to make the site viewable on X platform.
I guess the burden is on the people who want to communicate --- web
site developers and web site visitors. If the visitors want to
communicate, they can write software to handle the million weird
proprietary formats, or they can use existing software that does; if
the developers want to communicate, they can use standard formats more
visitors can understand.
> In the case of MSFT,
> the advocates happen to be the development employees. I can say from
> first hand experience, the IE team spends a lot of time before any
> release (though not always successful) running around and doing
> really tedious work. How would you like it to be your job
> to sit on the phone all day trying to get a developer of X plugin
> to make his stuff work on the new browser before it ships?
> Most of the time, we made hacks in IE to make it work with
> a given popular site or plugin.
> One way or another, we made SURE that important things, wether
> they be big sites or popular plugins, worked on the browser and OS.
So it takes Microsoft-class leverage to get plugins to interoperate
with your browser? That sort of undermines the point you were making
earlier, doesn't it?
> SO, the spec, format or whatever for flash has been made public
> (at least according to this thread), why doesn't someone stop
> whining and go write a better plug-in for NSCP on linux, or
> opera, or Konquerer, or whatever newfangled linux browser they want.
Yeah, and for Word, and for Excel, and for QIF, and for every
half-assed stupid unspecified bug-ridden pseudo-XML format with
embedded scripting that some moronic Microsoft contractor pulled out
of his ass two days before a deadline. Or you could just give up and
use Windows, and that's what the proliferation of formats leads to ---
switching platforms being prohibitively expensive.
> "Scratch that Itch" I say, make a better one, do a great
> deed for the open source movement!!!
> If open source is "oh so fucking wonderful" here's your chance...
> just get out of my fucking face and go back to your science project of an
> OS...
I'd like to remind the audience that JoshCo co-wrote one of the most
eloquent pieces on how wonderful open source software was for users.
But that doesn't have much to do with how easy it is to reimplement a
thousand different pieces of software from scratch.
> >Hopefully all of these devices using derivatives of Mozilla will help
> >swing the tide the other way.
>
> Get Real!!!
> Very few "devices" are using Mozilla, or have announced plans to.
I can't imagine who would want to. It's huge.
> AU systems browsers, or Sybian based browsers...
Sybian-based browsers? Are these browsers that run on that
thousand-dollar supercharged dildo? :)
> If people really wanted aol to switch, the advocates would do the
> (unpleasant) work of making mozilla/netscape truly componentized
> so that AOL could really just flip a switch and use it.
>
> Maybe an interesting challenge would be to try and build
> a set of components based on the mozilla code that
> implement the same COM interfaces as the IE components.
Yeah --- all they'd have to do is reproduce the thousand or so
undocumented bugs in the IE component without introducing any of their
own, and they'd be set. Realistically, though, AOL is going to have
to do some work.
FWIW, Galeon and Konqueror already embed Mozilla as a component.
> Even IE has done a better job, Windows, Mac, Solaris, HPUX.
> While Linux may be a competitor in spirit, Solaris and HPUX
> are far more realistic competitors *today* in that they
> directly take away NT sales, especially in the server room.
> However, despite this, customers told MS they wanted
> IE for the platforms they have in their shops, hence
> the solaris and HPUX versions. Linux just doesn't exist
> in the same numbers in big business at this point...
It could also be that MainSoft's MainWin doesn't support Linux, and
that's what the IE ports were done with; and MainSoft's non-support of
Linux is rather curious, given the level of demand I've seen among
developers... could it be that their continued viability depends on
Microsoft not doing to them what it did to Bristol? (Not that Bristol
has a Linux version of Wind/U either.)
> Anyway, maybe when good old AOL which owns sweet netscape
> ports their applications to Linux, and decides those
> customers are worth catering to on the desktop...
What's the chance that AOL will pick a fight with Microsoft like that?
Peaceful coexistence is likely to be far more profitable, particularly
given the level of vindictiveness we heard loads of testimony about
during the antitrust suit.
This archive was generated by hypermail 2b29 : Sun Apr 29 2001 - 20:25:22 PDT