On 10/10/11 8:02 PM, Dave McGuire wrote:
On 10/10/2011 06:56 PM, Toby Thain wrote:
X11 just does not satisfy me as a way to define a
high-information-content user interface. It's certainly capable of
delivering a high-information-content user interface but it doesn't
define it.
I agree with the latter, but for the former, it does satisfy me. (that's
not to say that I think it should satisfy you) There's really nothing
that can be done on a graphics display that X cannot do in one way or
another, in an almost entirely device- and architecture-independent way.
It's goal is not to "define" user interfaces, but to provide a mechanism
for creating them.
That's too low level for most stuff, as the web has all but proven.
I don't even know where to begin with this. It's a total apples vs.
oranges comparison.
"Proven"? How, exactly? "The web" isn't a replacement for a
windowing
system any more than Windows is a replacement for an oscilloscope.
You're comparing apples and oranges, and while I don't know you very
well, I damn well know you know THAT. In fact, I've concluded that
you're just bored and know I won't let completely stupid and/or patently
false statements just float by.
Further, that low-level "mechanism, not policy" architecture of X is one
of its big points of existence. An X display can look and act like
ANYTHING.
And that's the problem. For most uses, the detailed design of widgets
doesn't matter. You just want a CHECKBOX here, a TEXT field here, and a
SUBMIT button there. Good enough for Amazon, at least, which is a
non-trivial UI. Plenty of other examples too.
There are too many respectable people on this list who DON'T know much
about this stuff to play with the possibility of misleading them here.
X11 was never designed to be used from one continent to another, and the
WWW was never designed to be the exclusive way that humans interact with
computers. That whole "wasn't designed for it" is precisely why neither
of them work very well in that capacity.
Remote computing on a intranet is interesting, less wasteful than big
desktops, but it hasn't exactly thrived as a model. For it to do so,
would resuscitate the idea of a "workgroup" or "departmental" server
with enough grunt to virtualise desktop apps.
Until the www shows itself too embarrassingly incapable of this, you are
going to be stuck with "cloud" app providers like Google.
The web by itself isn't too fancy but couple with
Javascript, AJAX,
SVG and other high-information-value media of interchange, and boy is
it fun to work in! With just a little care, apps scale up enormously
and deploy with so little effort on my part.
Yes, some of the stuff they're abusing the protocols and bandwidth to do
are pretty impressive. Using them, however, TO ME, feels like riding a
bike with square wheels. X is a pretty "heavy" protocol in terms of
bandwidth utilization, but it's positively miserly when compared to XML
or JSON!
Errr... that's not quite as obvious as you imply. :)
The web offers massive opportunities for compression, that X11, RDP,
VNC, etc, can't match, in having a client side processing environment
and complete flexibility in data formats...
So your logic is that it's ok if it's bloated and inefficient, because
we can just compress it? That sounds like the Lispers' arguments that
the performance problems associated with Lisp have been "solved" because
computers are faster now.
(Lisp doesn't have any performance problem, but that's another thread...)
No, I am saying that the web is architecturally efficient, at least
potentially, because it allows complete flexibility in protocol and
front-end decoding. The protocol you like - doesn't.
And it's too bad those "opportunities" for compression aren't being
taken advantage of, save for the web servers with the Apache compression
option enabled. And there are, what, maybe six of those? The fact is,
that stuff CAN be compressed, since it's all fat inefficient text, but
it almost never IS compressed on the wire.
I'm not talking about gzip. I'm talking about compression *in general*
which means designing your vocabulary and operations to suit the domain
in question. X11 cannot do this, it's at least one level removed from
this flexibility - but systems like NeWS *can* too (at least to my
recollection), and they made X11 look crummy at the time.
Next...will JSON and AJAX (etc, whatever the kids come up with next
week) really deal with compressed objects, or is this argument dependent
upon bandwidth providers handling that compression for you, by having to
build compression support into all of the end-user "routers" that are
terminating lines everywhere? At the server/browser level, then? (which
is probably the only practical place to do it, now that I've typed that)
JSON is a compact format but not the only one possible.
Gads, man. I'm not talking about using X11 to somehow "replace" the
friggin' WWW. I'm talking about using it from DOWNSTAIRS.
Wifi sucks. :)
--Toby
-Dave