Re: [webmin-l] Ramblings after reviewing the pole

[Barry <we...@i1...>]

Unknown Questions wrote:

>
> *~* Barry Caplan says in the next posting to the effect that Cobalt 
> didn't quite get its easy / idiots guides in place even though they 
> identified it as a major requirement


Actually, I meant Cobalt didn't complete its mission of creating a large 
line of server appliances, each of which could integrate with the others 
in some unifying fashion. I am sure the documentation was never quite 
complete either, but I was thinking larger then that...and frankly, our 
driving principle in the UI was that since no one was really going to 
RTFM anyway, everything should *just work* and they shouldn't need to RTFM.


To make something really simple requires a really deep understanding of 
user needs and patterns, and information architecture, in order to match 
it to the system at hand. Frankly, there is no real reason, for task 
oriented approaches, that the user needs to know about system level 
stuff at all.

It actually could be quite liberating to design a product that way - 
treat the systems already there as a core part of the overall system, 
but not the system itself. Let the system include, and even be driven 
by, parts that are closer to the end user.

one thing I think was subtle and not really realized by many folks in 
the marketplace was that at Cobalt we never really marketed ourself as a 
Linux company even though we were arguable the most successful Linux 
company of all.

Instead, we marketed ourselves as a solutions company, and the problems 
we were offering solutions to were couched in terms closer to the end 
user's point of view, and not Linux specific at all. That we chose Linux 
toimplement the solutions was a great decision, but we could have done 
the same thing using some other *Nix, or even Windows. We didn't want 
our customers to have to know what was under the hood anymore then (fill 
in your favorite auto manufacturer) wants its customers to know how to 
build a car or keep it tuned up, or worse, manufacture the actual metals 
form ores.

We felt the much bigger market was going to expect and even demand that, 
even though the Linux sysadmins were our early adopters and very visible.

Yet Webmin sometimes feels as though the limited sysadmin market is the 
*only* market it is aiming for, and that is to its detriment. Many parts 
of it come off feeling like the equivalent of having to learn how to 
smelt ore to build your own steel when all you want to do is drive a new 
car. If cars came that way, then very few of us would have cars, and 
frankly, far fewer webmin users exist then could if this was done with a 
different approach...

IMHO.

Best, Barry