In article <[log in to unmask]>,
Denys Beauchemin <[log in to unmask]> writes
>The latest crop of high-end systems are 450 MHz Pentium
>II with 64 or 128 MB of SDRAM on a 100MHz motherboard, 18 GB of UltraDMA
>disk drive, a third generation DVD-ROM or a 40X CD-ROM and an AGP video
>card with 4 or 8 MB DRAM. Starting Microsoft Word takes 4 seconds or less.
Just for comparison, I started Word 97 on my 60mHz Dell Pentium, 1993
vintage, with 32mB RAM, 500mB of old WD Caviar IDE drive Drivespace 3'd
to 900mB, and a 1mB graphics card I can't even get proper WIN95 drivers
for.
4 seconds.
Filled the memory with Excel, Access, etc., so it couldn't take
advantage of preloaded stuff. Still 4 seconds.
So what are you guys *doing* with the other 390 MHz? Have a look and see
if somebody hasn't come in via the 'net and corralled your machine into
the search for the next largest prime :-)
But apropos the future of the PC, you don't *know* where the driving
forces are going to come from.
When the PC first came out, we could see it could run a chess game, but
who would want to run an Accounts package on it?
Nobody, of course. Well, not until much later... But it didn't take long
for WP/Spreadsheets/Personal DBs to emerge as the 'big three', even
though it seems obvious now.
Now, games might have driven PC specs up, but a Nintendo or a
Playstation can leave the PC in the dust at polygon handling.
And Bill's done a great job driving the business market up, a little in
the vanguard, but user resistance has kicked in. Office 2000 won't have
the benighted Paper Clip, or any of his ilk, and IE5 (or will it be 6?)
is supposed to be a *lot* smaller.
I love the way the mainframe/mini/PC debate has now gone nearly full
circle. We bought minis because the mainframes were large inflexible
dinosaurs that MIS wouldn't let us play with. And when the minis got
mission-critical, and MIS got their 'dead hands' on those too, we bought
PCs.
And then they got so mission-critical (or some of them did) that MIS got
hold of those too. And found them hard to maintain. And now client-
server is dying a death of complexity, maintainability and insufficient
bandwidth, we have Windows Terminal Server. Which is just like
character-mode green-screen, except it's graphics-mode multi-coloured-
screen now. And isn't WTS just a perfect match for the NC?
Now, we are starting to see that 10mB networks and/or the extra layers
of protocol needed on each machine don't cut it for client-server,
compared with the tightly-couple approach.
Hence WTS on the one hand ('make it more tightly coupled') and 100mB and
up on the other ('let's get that bandwidth'). Kind of solving the candle
at both ends.
But somehow, we are supposed to believe, the Internet *will* have enough
bandwidth, and hey, we 'just need something a little more complex than
the browser metaphor', and all will be well. Beats me how; just do the
math. Is everybody thinking qualitatively when they should be thinking
quantitatively here?
So where are we when we have gone full circle? Instead of giant
mainframes, we are to have our ISP's processors. Not sure how they
differ, but never mind. Instead of full-blown PCs, we are to have cut-
down, ultra-thin-client, 'web browser' machines.
Sheesh. I guess we never wanted these local apps did we? I mean, how
many of you would ever buy a film on video, when you could rent it, see
it on scheduled TV, or even get it Video-On-Demand via subscriber TV?
Oh, that many? Ahh. OK, transport? You'd never take your car if you
could get the bus or a train, would you? You *would*? Well, I don't
know....
So I guess we aren't gonna give up those PCs either.
Now one of the main problems with PCs for corporate use is
maintainability. I have two choices of GUI front end to evaluate at the
moment. One needs downloading to every PC, and modifying on every PC
every time I change my app. The other lives at the centre and uses a web
metaphor.
The 'central' one is 'easier' right now, but it's just bandwidth again.
Once I have enough bandwidth to those remote PCs to support the kinds of
things being envisioned as making local hard drives redundant, I also
have enough bandwidth to use 'push' technology for my upgrades. And with
software support for a 'Zero Administration Policy', the 'pushing' will
be a no-brainer.
Today, 'Make' figures out what's changed, and selectively recompiles.
Tomorrow, Make will 'push' as well.
So, if the technology issues more or less cancel out, and if we want to
be masters of our own destiny, the PC isn't about to die, now or anytime
soon.
Doesn't mean to say a lot of people won't prefer the ISP/NC approach if
it's offered, and will happily trade the 'autonomy' of a PC for the
'security' of an NC.
Where do I think it is going to go? Well, despite business use, I think
PC development will continue to be home-entertainment driven, and it
will keep on going, getting faster/bigger/cheaper/easier.
We've only just begun to scratch the surface of the convergence of PCs
and telephony, let alone the joker in the pack, television.
Wish I knew how it was going to happen though. A little piece of that
would make anybody rich.....
--
Roy Brown Phone : (01684) 291710 Fax : (01684) 291712
Affirm Ltd Email : [log in to unmask]
The Great Barn, Mill St 'Have nothing on your systems that you do not
TEWKESBURY GL20 5SB (UK) know to be useful, or believe to be beautiful.'
|
