In message 51365.12.6.201.2.1291651405.squirrel@popaccts.quikus.com, "J. Fors
ter" writes:

Well, that's never met my bar for QA on backups.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

Back before PCs I was writing doctors office software. The system had one fixed
and one removable drive plus a 8" floppy drive. I wrote some software to backup
the removable disk on the floppies as the fixed drive had too much data to do a
two step backup. Several years later after I was at a new job and my old
employer wanted me to write a restore program as the fixed disk crashed. Luck
was with me and the restore worked. I would not have risked it otherwise, so the
restore program was not "needed" till that point. Hardware was more expensive
then and good backup systems did not exist at a price the market would bear. Now
we have so many ways to backup, but hardware is also more reliable that few
bother to backup a contact list on their cell phone. Insurance is just like
anything else the price must be close to or exceed the value expected to
receive. Can the cost of a good backup system exceed the value sure.

Stanley 

----- Original Message ----
From: J. Forster jfor@quik.com
To: Poul-Henning Kamp phk@phk.freebsd.dk
Cc: Discussion of precise voltage measurement volt-nuts@febo.com
Sent: Mon, December 6, 2010 10:03:25 AM
Subject: Re: [volt-nuts] do you like Labview in your labs?

Disk platters were very expensive. I checked the tapes were readable, but
not the files on them. A new car could be bought for about the price of 5
platters.

-John

===================

volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

It depends on how much money you have and if you want to eat or not.

-John

================

In message 55428.12.6.201.2.1291655181.squirrel@popaccts.quikus.com, "J. Fors
ter" writes:

No, it depends how much you value your data.

Poul-Henning

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

A single 2.5 MB platter was groceries for 3+ months. A Diablo 31 was the
price of 2-3 cars. A 4K core memory board was about the same as a disk
drive.

-John

===================

In message 52731.12.6.201.2.1291655614.squirrel@popaccts.quikus.com, "J. Fors
ter" writes:

I know.

I was there too.

But it still only depends how much you value your data.

We went on excursions to vendors machine-room to test restores.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

I had no way of checking restores.

-John

===============

In message 59300.12.6.201.2.1291656020.squirrel@popaccts.quikus.com, "J. Fors
ter" writes:

I guess you mean: "Checking the restores would have been to expensive
relative to the value of the data."  Good sysadmin'in means the
CEO gets to make that decision.

Anyway, I really don't see that as a valid reason to curse all
software not written by a $MEGACORP...

Poul-Henning

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

Look, if you were NASA, you could afford to test every nut, bolt, washer,
or whatever.  If you are a small, poor operation you just cannot. That is
reality. You sometimes have to do what you have to do. Sometimes you do
without.

In the US there is a freedom to succeed, but also a freedom to fail. They
go hand in hand. Whatever the outcome, you are essentially on your own.

I use older versions of widely used commercial SW because it has been
pretty much debugged. If something really is crap, it will not survive the
marketplace. Vista, for example.

That's what I do.You do as you please.

FWIW,

-John

==================

J. Forster wrote:

If you stick with a stable version of an opensource software product, and
don't upgrade it, programmers won't "screw with it".  On the other hand, if
you upgrade any software, open source, or not, programmers will have screwed
with it.  That's why it is called an upgrade.

So, you discovered the problem with using an incompetent programmer.  They
exist.  Any competent programmer would have tested the utility to make sure
that you could backup and restore everything the utility was supposed to
backup and restore.

Most of the opensource projects are built by teams.  Virtually every project
that I mentioned earlier were written by corporate teams.  Openoffice.org is by
Sun. Mozilla, by Netscape.  SELinux, by NSA. Octave, by the University of
Texas, and funded by DEC, SUN, IBM, ..., OpenSolaris is by Sun.  Python is
by the Python Software Foundation (A non-profit corporation).  The ubiquitous
scripting language PERL was written by the genius Larry Wall.

Further, if you used any of the earlier versions of windows (95,98,XP), the
IP layer, and PPP code were lifted verbatim from BSD unix.  IE was based on
the open source mosaic project.

I can't think of a single microprocessor/controller manufacturer in the last
15 years that didn't first do the porting work for their new machine to the
opensource GCC compiler.

You are really doing yourself a disservice letting the experiences you had
with one incompetent programmer keep you from enjoying the goodies offered
by the opensource software movement.  If you stay with the stable releases
of Debian, or Ubuntu, or Mint linux, I dare say you will never find a bug.

-Chuck Harris