CHAT: The ridiculously stupid, offtopic HCI thread.
|From:||Keith Gaughan <kmgaughan@...>|
|Date:||Friday, September 3, 2004, 11:50|
Philippe Caquant wrote:
> --- Ray Brown <ray.brown@...> wrote:
> I perfectly respect Ben Sheiderman, whoever he might
Then read the guy's books. He's one of the big names in HCI, and it'd do
you no harm. Methinks you've never actually had to build a
reasonably-sized piece of software, whether against a deadline or not.
> The fact is that I never had to cope with his theories.
In fact, you probably have, it's just that you don't know it, and you
There's an ad running for Guinness (the beer) here in Ireland. The
concept is that people always need to find something to compain about,
or find fault in things no matter how good they might be. It seems to
me that's what you're doing.
Of course software has flaws, and that's all you're seeing. You haven't
seen the leaps and bounds that have been made since the 60s and 70s in
terms of HCI, and you shouldn't because that's the point of it all:
making the interface as easily learnable as possible. Screw intuitive
because next to nothing's intutive, but we can make things learnable.
> I have to cope with ordinary computers and
> software. I'm not working in a university, but in a
> normal office, with normal people. What planet are you
> living on ?
I'm living on Earth: how about you?
I work in an office where as part of my work I have to build interfaces
for people who are functionally illiterate. When I tell you I know more
about HCI than you do, I'm not kidding.
>>>But I, as a
>>>user, don't give a damn about what the smart
>>>specialist thinks. I got a fucking tool,
>>I assume you're not familiar enough with colloquial
>>English to realize
>>what you've just written!
> Ah, an interesting point that had escaped me so far. I
> just tried to look American, and I heard that the
> first rule to speak American is to say "fuck" every
> fourth word, as an average. But true, "fucking tool"
> sounds, ehm, weird.
> Look, we got an AIX system, and it was delivered with:
> vi, emacs, ed and INed, whatever it may be. This
> system is recent.
Yup, they're mostly Unix tools that date from the '70s. Nobody uses ed
anymore, but emacs and vi are commonly used because vi has a small
footprint making it useful for admins who need a powerful editor they
can fit on a floppy. Emacs is completely programmable and almost an
OS in itself. These are, and pay attention now P-O-W-E-R T-O-O-L-S.
They're mostly not meant to be used by your granny, but instead by
people who need the power they provide.
> If they are modern good tools, why
> weren't they included in the package ?
That makes no sense. Read what you've written again.
Anyway, what IBM wants to bundle with AIX is IBM's business: take the
matter up with them.
> Why do I have
> to make an inquiry to know what the supposed good
> tools are, and how to get them ?
You mean, look at the manual? Possibly because they're P-O-W-E-R
> And, believe it or
> not, most of the people working today on Unix still
> use vi !
Yes, because it's bloody powerful, and you can move mountains in a few
keystrokes in it. When you're editing vast amounts of text, this is a
benefit. If all you do is write grocery lists with it, the extra work
needed to learn its esoterica isn't worth it. That's why your average
non-power-user Unix user uses the likes of Kate and Gedit.
> And they are proud to know that Esc-ZZ means
> something, and ^z means something else. That's why I
> said "stone age".
And you know, both vi and Emacs where actually huge leaps forward when
they came out.
They're proud because these esoteric keystrokes allow them to do just
about everything without taking their hands off the keyboard, meaning
they're able to be more productive.
vi was first developed by a teenager who wanted a better text editor
than what was out there at the time. The state of the art in HCI might
have moved on since there, but that doesn't mean that vi doesn't have
a place in the world anymore.
> The fact that "millions of people" are satisfied with
> vi isn't really an argument to me.
Maybe you're missing the point: vi is a P-O-W-E-R T-O-O-L. One tool
doesn't suit everybody. Anybody can grok notepad, but I could never be
as productive with that building stuff in ColdFusion without HomeSite.
Equally, my sister would just be confused by the array of options built
into it. It's not that she's dumb or anything, far from it, it's that
the tool doesn't suit her.
One size does not fit all. Stop trying to pretend it does.
> Millions of people
> are satisfied with George Bush or Jacques Chirac.
> That's their problem. My problem is having a good (and
> intuitive) tool to get my work done.
And the whole world isn't under the thumb of Bush and Chirac, so what's
Anyway, if you're not happy with Chirac, then remove him from office
through democratic means. Last time I looked, France *was* still a
>>> That's not uncommon: PHP and Perl do it. You're bitching about nothing
>>> there. Most languages designers attempt to keep things familiar and only
>>> change things if they can't think of any other clean way to do it.
>> Yes, I did try to explain this to Philippe a few weeks back. I agree he's
>> bitching about nothing, but he won't be convinced.
> OK, my problem is, for ex, the following: as I'm
> else), I try writing small programs or functions in
> 75% of the cases, when it doesn't work, it's because
> of syntax problems:
<hits head against the wall>
> I know perfectly what I want to do
> and how I would write it in a language I know, but
> it another way
That means you don't know the language. Let's just say you you started
using French grammar with English. Now, by your reckoning, that should
be ok because you know what you wanted to say and how you would say it
in a language you know, but all those horrid anglophones decided it was
no good and you had to say it another way.
You're like somebody who'd go up to my box and type something like:
$ write a game for me
sh: 'write' not found.
$ make game
make: could not make 'game'.
but worse, because you should know better.
> (BTW, I'm just reading Flanagan's
> using the same symbol, "+", both for concatenating and
And that's why many dynamically-typed languages use a seperate operator
especially if they're weakly typed.
> - about type conversions, for ex. WHY had they to do that ?).
If you've ever written anything in a statically-typed language that
doesn't have type-inference, you'd know.
glue code, though it's got a lot of hidden power due to the fact it
has closures (which rock). When you're doing that kind of thing, you
don't want to worry about type-casting, you just want it to do stuff.
> Anyway, all the time I spend on these
> syntax problems I consider as lost for productivity.
So learning new things is a waste of time. Riiiiiight!
Keith Gaughan -- talideon.com
The man who removes a mountain begins by carrying away small stones.