HP3000-L Archives

April 1998, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Bruce Toback <[log in to unmask]>
Reply To:
Bruce Toback <[log in to unmask]>
Date:
Thu, 2 Apr 1998 13:56:26 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (70 lines)
Gavin writes:

>Certainly NT is much better in many areas than 95, but when it comes down
>to whether it is fundamentally more robust and stable when used in a
>real-life developer workstation environment, I'm not sure it's really any
>better than 95 was.  Install programs still vomit files all over the disk,
>happily replacing whatever critical shared DLL files that they feel like,
>and none of NT's security or kernel features seem to be used to prevent
>applications from doing whatever their developers felt like to your system.
>
>So, I'm interested in whether this is just me or not.  How are other people
>finding the reliability of NT as a *workstation* platform over time?

I haven't found the reliability of NT as a workstation platform. Is it
supposed to be there?

Yes, it's possible to get the system into a state so that it will work
for "weeks" at a time. However, that's not a stable configuration.
Anything you do differently is likely to result in another pile of hours
down the tubes as you try to figure out what went wrong and fix it.

I know I'm not alone in dreading adding hardware or software to my NT
systems. One prominent HP 3000 developer refuses to add any A/V software
like QuickTime or RealAudio to his machine because it took him three
weeks to get sound working the first time. I don't want to add another
disc drive to my machine, even though I have the drive and need the
space, because I don't have a spare four hours in which to troubleshoot
the result. I certainly don't want to replace the drive with a bigger
one: I'll never get the Registry back to the way it was.

Installing software is another nightmare. Typically, I allocate fifteen
minutes to install a new software package, plus two hours to figure out
what went wrong and fix it.

The Registry is the most poorly thought-out concept ever put into a major
operating system. It has all the disadvantages of Unix's zillion little
configuration files, with the extra added detraction of being invisible.
It was obviously conceived in the rarefied air of some academic exercise,
without any thought given to the fact that in the real world, things get
messed up.

It's very hard to make an HP 3000 unbootable by manipulating files. There
are a small number of absolutely critical files, and if these are there,
the system will come up far enough to allow the rest to be built,
restored or reconstructed. A Macintosh will boot from any disc that has a
system and a finder: two files are sufficient to make the system usable
enough to repair. NT, like Win95 and earlier versions, can be killed by
any number of minor cuts and abrasions.

So in the very narrow sense of being a machine on which it's possible to
develop application software, NT is adequate, and a big improvement over
Win3.1. For developing system-level software, or as a developer's
ordinary desktop system on which software is constantly being added,
updated, tested and removed, it's awful. As far as I'm concerned, my NT
system is a black box that runs DevStudio, not a general-purpose
computer. I don't have time to keep it running in any other capacity.

-- Bruce,
   (staring morosely at this 4GB IDE drive that's
    still uninstalled)



--------------------------------------------------------------------------
Bruce Toback    Tel: (602) 996-8601| My candle burns at both ends;
OPT, Inc.            (800) 858-4507| It will not last the night;
11801 N. Tatum Blvd. Ste. 142      | But ah, my foes, and oh, my friends -
Phoenix AZ 85028                   | It gives a lovely light.
[log in to unmask]                   |     -- Edna St. Vincent Millay

ATOM RSS1 RSS2