Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!ucsd!ucbvax!bloom-beacon!LARRY.MCRCIM.MCGILL.EDU!mouse
From: mouse@LARRY.MCRCIM.MCGILL.EDU
Newsgroups: comp.windows.x
Subject: Re: A tirade about inefficient software & systems
Message-ID: <9011040724.AA03335@Larry.McRCIM.McGill.EDU>
Date: 4 Nov 90 07:24:46 GMT
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Lines: 85

> 1) The X Window System and GUIs built on X are expensive in terms of
>    CPU, memory and network resources.  Discuss.

Discuss what?  How your first sentence resembles "Have you stopped
beating your wife yet?"?

Also, "expensive" is a fuzzy enough term that the same accusation could
be leveled against well-nigh any windowing system of any sort.  In
fact, R4 contains a great deal of work designed specifically to lighten
the CPU load the server places on its machine.  (It's called speeding
up the server, but it amounts to the same thing.)

>    You should compare the X philosophy with competing network
>    windowing philosophies such as those based on PostScript.

Very well, let's compare X to a PostScript-based system.

The X approach distributes the memory load: the server is smaller at
the expense of making the clients larger.  I have never done any tests,
so I can't comment on the relative amounts of "smaller" and "larger".

Similarly, X distributes the CPU load.

X uses more network bandwidth, except in very atypical applications.
This is a disadvantage for the X approach.  (Again, I haven't done any
tests, so I can't say how much more.)

One difference which is not a philosophical difference, but is still
significant, is that X is available entirely free.

Another non-philosophical difference, which again is significant: X is
far more widespread.  (This is, I believe, due in large part to its
being available free.)

The first two are, IMO, advantages for X.  There is only one server,
but many clients, often on many machines.  When the client machine is
not the server machine, the client machine will typically be the
"bigger" (more memory, faster cpu, etc) of the two.  In the case where
the client and server are known to be on the same machine, and
portability is not a consideration, an all-singing-all-dancing server
may indeed be a better way to go.

(Network bandwidth does not worry me.  Networks are fast and getting
faster: 10Mbps is so common you find it in dorm rooms nowadays, with
fiber (80Mbps? 100Mbps?) in regular use, and higher rates coming down
the pike.  In any case, at least to my mind, it's heavily outweighed by
the advantages.)

> I personally think that X is seen as far too much as a panacea for
> Unix HCI, and it worries me that X has been seized upon as a
> ``standard'' with little thought for the implications in terms of
> hardware and networking costs.

HCI?  I'm not familiar with the acronym.

Seizing upon anything with little thought for the implications is
generally a bad idea. :-)

> The fact that I hear people from Sun tell me ``Our implementation of
> Open Look is great --- but you need 16 Meg in your SPARCstation''
> alarms me.

And well it should - but it falls on the other side of the line :-)
OpenWindows is one of the server-does-everything servers, and not only
that, but it *also* supports the X model.  Of *course* it's huge!

> The fact that a 68030 isn't enough CPU for a single user of an X GUI
> under V.4 alarms me.

It's not?  The 3/60 on my desk seems to do just fine, and it's only a
68020....

> The fact that X applications potentially generate significant numbers
> of packets and eat up CPU cycles even when there's no user
> interaction alarms me (ever watched a LANalyser when xeyes is
> running?).

That's a cause for bug reports, not for slamming the window system
design philosophy.  I'm sure it's not in the least difficult to code a
NeWS application badly so that it eats up bandwidth like mad too....

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu