I don't think you'd call what I do with my PC's as "careful tuning" and when
I got this notebook, with the preinstalled Win95, I loaded MSOFFICE 97-PRO,
Free-Agent, CorelDraw, a few other graphics handler programs and the PLUS
package so I'd have System Agent.
I also needed some CD-handling software to drive my CD recorder, and some
sound editing software. All that went in the first day and I've not had a
single serious problem to complain about. That's when I stopped complaining
about Microsoft. Their work is as good as any I've seen, and it costs VERY
little.
The app's I run aren't all from MS. Some are, though, and they seem to work
well, too. Some are Windows 3.1 apps which I've seen no need to update, yet
they work just fine.
There are a lot of complaints, some probably legitmate, but I'd say that by
and large, the majority of complaints result from failure to RTFM.
Dick
----- Original Message -----
From: Zane H. Healy <healyzh_at_aracnet.com>
To: <classiccmp_at_classiccmp.org>
Sent: Thursday, April 06, 2000 12:39 PM
Subject: Re: !Re: Nuke Redmond!
> >> Then there is the issue of *stability*.
> >>
> >I'm not sure what you mean by this. This particular system ( the one I
use
> >for my email, etc) has been "up" without a hitch for three years without
a
> >problem. I've never seen reason to cuss it. The key is that I don't try
to
> >make it do stuff for which it wasn't intended.
> >
> >The folks I see having problems with their MS-OS-based systems generally
are
> >the ones that (1) hand around the "chat" rooms (where their computers get
> >"social diseases"), (2) try to squeeze more performance out of their
> >computers by violating the components' specifications, (3) try to get
their
> >computers to do other sorts of things for which they (or their software)
> >weren't intended. Now, that's not to say it doesn't happen otherwise,
but
> >from where I sit, that's what I see.
>
> OK, I've observed of people useing Windows, there is always the occasional
> person that doesn't have problems. I've also noticed that the *only* time
> I hear about this is when someone else points out how unstable Windows is!
>
> Having said that, I'm aware that if you *very carefully* tune the system,
> you might be lucky and get a system that will stay up, even under a heavy
> load. I'm also aware that you'll have better luck if you don't have *any*
> 3rd party software on the system.
>
> I'm one of those people that just has to look at a Windows system and it
> crashes. Now let's look at your reasons why that could be.
> 1. I don't mess with chat rooms, I don't have time
> 1a. I don't pirate software (inferred)
> 1b. I don't even allow any systems I have running Windows to know
> that the Internet exists! Well, this one isn't quite true, as
> I've now got VirtualPC running Win98 on my Mac, and let it know
> the net exists so I can easily read the PDP-8 doc's on Highgate.
> However, I just have to push a button and I've got a clean
>
> system. Still none of the computers running it native have
> known about the net.
> 2. I don't overclock or even push these systems
> 3. I don't trust Windows enough to try anything like that
>
> BTW, this is what I call stability, this was just prior to a power outage
> on Sunday (I really need to break down and get an UPS).
> 4:15PM up 239 days, 16:56, 4 users, load averages: 0.06, 0.07, 0.08
>
>
> >> Someone mind explaining why if I install software on a Microsoft system
or
> >> make *very* minor changes I've got the reboot the _at_*& #$)@ thing?!?!
> >>
> >I've never wanted to become an expert on *NIX and its kin, but IIRC, if
you
> >make any changes to the system you not only have to restart the system,
you
> >have to recompile several modules, including, in some cases, the kernel.
I
> >remember attaching a serial I/O card to a LINUX box once, wherein I had
to
> >recompile (*GETTY) and restart several times. Adding a port to windows
> >normally didn't require a last time I did it. Of course, it is sometimes
> >necessary to restart the system if you have to add hardware, since nobody
> >recommends doing that without first shutting off the power.
>
> I'm not talking about adding new hardware and drivers, I'm talking about
> doing what should be a simple software install, or changing the
> configuration of software already on the system. The only time you should
> have to recompile the kernel is when adding a new one, new hardware (and
> these days that's not a requirement), or making some very serious changes
> (networking is about all that comes to mind). Also several times I've
> found it easier to get Linux to support hardware than Windows.
>
> I believe that it is to difficult under Linux to do these kind of things,
> however, progress is being made.
>
> BTW, MS isn't alone in the reboot crime. The Macintosh has this problem
> also, and is getting worse about it.
>
> >Well, the cost differential was larger than the cost of the PC machines I
> >used to demonstrate what a poor choice the uVAXII was as a platform
during
> >my last stint in the aerospace industry. THE JPL guys liked the uVAX-II
so
> >they used it to replace the Apple-][ that was originally designed into a
> >military-oriented project. I wouldn't argue that the uVAX-II didn't do
> >better than the Apple-][, but their ESDI interface didn't outperform
SCSI,
> >which they claimed it did, and the high-res graphics cards we were told
to
> >use in the uVax-II cost as much as the entire uVAX-II with all the other
> >peripherals. A comparable card from the same vendor but designed for the
> >PC/AT cost only $600.
>
> One should always buy the proper hardware for the job. If the
pple ][ was
> able to do the job the MVII was overkill. However, what percentage of
> MVII's are still in active use, compared to PC/AT's?
>
> There is also the real killer, what platform is the application you need
to
> run available on?
>
> >Not all cases are so extreme, but it's the extremes that tend to be
> >remembered. It's also no surprise that DEC seems to have gone out of
their
> >way, during the early days of widespread internet use (1985-1988). to
make
> >their LAN boards incompatible with anyone else's. They also tweaked
their
> >protocols to weaken their own networking system so people wouldn't be
> >tempted to mix and match.
>
> Back then as I recall just about *everyone* was incompatible! I've no
> trouble making DEC LAN boards from that timeframe work with various
systems
> running 100Mbit, and I've got enough variaty I probably should have
> problems in this area.
>
> >I guess it just says that when there's a tool that gets the job done, it
> >makes sense to learn how to use it as opposed to sticking one's nose in
the
> >air because it seems too "unsophisticated". What's more, people pay for
the
> >process of getting the job done. They don't want to pay for doing it the
> >"hard" way.
>
> Actually despite what I've said, I like VB, and to some extent agree with
> this statement. However, I don't use it, as it's only available on
> unstable platforms. It's a great language for simple little specialty
> apps. Last I looked it wasn't a great language for large complex apps.
>
> Zane
> | Zane H. Healy | UNIX Systems Adminstrator |
> | healyzh_at_aracnet.com (primary) | Linux Enthusiast |
> | healyzh_at_holonet.net (alternate) | Classic Computer Collector |
> +----------------------------------+----------------------------+
> | Empire of the Petal Throne and Traveller Role Playing, |
> | and Zane's Computer Museum. |
> | http://www.aracnet.com/~healyzh/ |
>
Received on Thu Apr 06 2000 - 14:33:14 BST