[
Lists Home |
Date Index |
Thread Index
]
From: Bedros Hanounik [mailto:Bedros.Hanounik@tarari.com]
>Regardless, the main concept is the same. Intel (a major player)
>introduced a new technology(PCI), got feedback from the industry and
>gave it (and the control of its future) back to the industry to benefit
>in THEIR OWN platform. Including DEC which used PCI in Alpha processor
>platform.
Did they do that before or after they fielded it? Again, regardless
of who feeds it, a standard should be timely and for proven technology.
>I'm not anti MS, I'm anti proprietary technology. The openness of PC
>platform made the PC so affordable almost everybody can buy one.
And that was perhaps 20 years ago, circa 1983. It was a good thing
although the MAC almost beat it. Anyone still have a Sinclair? I
still have an Amiga 1000 and the boxes it came in. Nice machines
but losers. (Ok, the Sinclair was toy, but it was fun.)
>Proprietary hardware platforms died or going to die. Proprietary
>software platforms (such as MS windows) will do the same soon. Now it's
>the software platform turn's to be open and stable so we don't waste so
>much money and time redeveloping or porting our products, and hence
>cheaper software.
Ummm... only if the framework is ubiquitous as well. The costs are
in the semantics, not the markup. Otherwise, I generally agree and
that is why we should start ramping up to create a rich client
standard. Convergence of the computing and entertainment platforms
over public pipes makes that an economic necessity. But there are
multiple contenders and anyone who takes on this task will have to
cope with that and pick a winner. Not fun.
>Even if you rely exclusively on Microsoft platforms,
>it'll be too costly to maintain, because every couple years they change
>their own ways of doing things.
So did Sperry/Univac. So did IBM. So did Dec. So has Sun. It is
good to remember that we are the first generation to have been in
this business long enough to see it come down to five year cycles
of technology chasms and actually have enough time in to be able
to see a technology repeating itself. From about 1950 to 1980
approximately, computer platforms and languages were in distinct
boxes and commercial professionals were dedicated to those boxes.
Now we can see these come and go fairly rapidly by comparison. The
real trick is to distinguish innovation from a redux for the sake
of selling new product. We can do that better now, but I caution
you that anything really revolutionary will always be scoffed at
because it is unrecognizable. In that sense, innovation plays
within the boundaries of limited changes that provide a significant
advantage but don't go so far as to be unmarketable. Relational
databases and object-oriented programming sat on the shelf for a
long time until the implementations and the advantages became
obvious. I am not a believer in a 'disruptive technology' that
comes from nowhere. Usually, a disruptive technology is just
a technology whose time has come because its maturity make it
sustainable and the environment is ready to recognize it. It
only looks new to the inexperienced. That's ok. Buzz sells.
That is why the principle of independent invention applies
to this rich client markup notion. We have at least three
different examples, one old by comparison and two recent.
MID was friendless even though it did become an ISO project.
XUL and XForms have some traction. XAML is guaranteed to
get traction by dint of the pervasive platform. So rich
clients will be there. Now the question is one of the timing
of a standard for them. We'll need that because as I said,
public pipes and multiple rich delivery devices won't thrive
without it. The ISO ISMID project could be reevaluated and
that is a simple and direct route because this time, the W3C
is the wrong place to do it. If not ISMID, then it will be
something one of the cabals creates. Selah.
So the challenge is to get MS and others to the table. They
will come eventually and it is a matter of how long we have
to endure competing similar but incompatible solutions. XML
only provides portable data, not even bare inteoperation and
this experience is where the truth of that will become apparent
to everyone just as it did to the MID team in 1993. Common
syntax and even common vocabularies only gets one halfway.
len
|
