Re: [TCLUG:14414] Usability (was Ghost for Linux)

[Kevin Bullock <kbullock@ringworld.org>]

On Tue, 7 Mar 2000, Eric M. Hopper wrote:

> 	*shrug* You can make all the jokes you want to about CD
> cupholder people, but these are real, live people who have things they
> want to do on a computer too.  In the heat of tech support, it may be
> hard to believe that they can be in any way productive or useful, but
> there they are.

It seems so easy for "us elite folk who know about computers" to look down
on "those lusers who don't know a CDROM from the hole in their ass," but
really, how much do all of us know about, say, the chemical engineering
processes that make possible all that we do on computers? How much do we
all know about the mechanical engineering that goes into our cars? The
sysadmins and *n?x gurus are as much Eloi as [we|they] are Morlocks.

Short answer: [Let's] Get over [y]oursel[f|ves].

> 	I don't know if you've tried this, but next time the Outlook
> user has their problem, you might try explaining POP3 using as few
> technical terms as possible.  Once they understand what's going on,
> maybe they'll do the right thing.  I've actually had a lot of success
> with this technique.

This is what I do whenever I help people. It works very well. The basic
concepts of protocols, interfaces, etc. are very understandable once you
strip off the in-language and acronyms that make our jobs quicker.

> 	Yeah, the person may mess up when they use it, and do something
> stupid.  Suddenly they'd be confronted with having to learn how things
> really work.  It's that kind of thing that causes people to become
> Morlocks.

Isn't this how we all started anyways? :)

> > Now, on to the question of duplicating boxes, I believe RH has tools
> > to do and enterprise install, or something like that.  This may be a
> > better solution than Ghost, and easier to deal with than cpio, and
> > more robust than dd, however I have not used it.  YMMV.
> 
> 	I didn't know about this tool.  If it's nice and friendly, and
> easy to use to get some common things done, then it might be a reason
> for Ghost not to be ported.  On the other hand, the more tools that are
> ported, the better, even if they are eventually abandon.  Of course, it
> may be counterproductive to convince a company to port a tool that we
> know will be recieved very poorly, but I suspect Ghost would find some
> immediate users in certain places.

My sense is that there's enough momentum behind this kind of a tool to
create one that would fit into the POSIX philosophy, yet still make life
easier for duplicating lots of drives/boxen. Ghost was created within a
particular context - that of DOS (and later Windows). *n?x comes from a
completely different context, with default behaviors defined differently.

<hypocrisy level=mild>Instead of pontificating on a listserv about why
Ghost sucks vs. why it would be useful, form a project to outdo
it.</hypocrisy>

Long live the meritocracy of software! (But please don't extend this to
people.)

Pacem in Terris / Mir / Shanti / Salaam / Heiwa
Kevin R. Bullock