Survex (interface expectation)

[Andy Waddington on Survey stuff]

> If [no] manual and no joy from context sensitive help, how do YOU
> find out how a Windows program works ?

As I said, I read books written by [not Microsoft], or, more commonly, I
swear a lot, and assume that something I could do quite happily on software
which fit into 16k rom on a BBC micro is completely beyond software for a
similar purpose on a Windows system. The "View" word processor, for example,
later developed into ViewPS by Neil Raine at Acorn, produced, in 1986,
a word-processed (caving) journal on a BBC Micro + Apple laser writer which
still looks better than anything I can get out of Word (TM). I don't imagine
it is because Word can't do the things I want - just that the interface is
so counter-intuitive and overloaded and the help so inadequate, that it is
impossible to find out how to do it. I realise that Word is without doubt
the worst piece of software in the universe, and that it is unfair on
Windows to tar it all with the same brush - some stuff is usable and very
rarely even quite good, but it is the stuff like Word that creates the
"interface expectation" isn't it ? Windows users don't read the f*** manual
because they are used to struggling without one, and find that even if they
get one, it doesn't help at all. In that situation, most of us would not
bother to read a manual, either...

Don't get me wrong - I'm not slagging off Windows as being poor quality here
- this was a question of interface expectation, and many modern desktop
OS'es create the idea that a manual is an obsolete concept whilst not
providing enough help to find your answers. KDE is not a huge amount better
and suffers from imitating a lot of Windows more popular features. RISC OS 4
is the other desktop I use a lot, and I don't have the same problem here
because I have very detailed manuals from previous releases, and the OS has
pretty much stopped evolving, so I only use it for old software with which I
am quite familiar. A new user (hah!:) might find that it was just as
difficult to learn to use - I didn't seem to get more than a couple of
sheets of "how to install" with this modern release. Manuals seem to be
things you are expected to pay extra for, and most people don't nowadays.

The problem is how to get over to users of Survex that this is *not* an
appropriate idiom for cave survey software. Users need to understand about
cave surveying - we don't provide a "Wizard" to hand-hold you every step of
the way because users who don't understand what they are doing are going to
produce crap survey data anyway. There must be lots of software that isn't
the run-of-the-mill application type that "everyone" knows about, and which
is designed for specialists to use. How do those sorts of applications get
over this problem of users assuming that a manual is unnecessary ?

In my (contentious ? Moi ?) opinion, users who don't want to read a manual
and can't use a command line interface, should probably be actively
encouraged to keep out of the data processing side of cave surveying :-)

Andy