Re: web inherits UNIX console?

[David Holland <dholland-tech%netbsd.org@localhost>]

On Tue, Oct 04, 2011 at 08:12:50PM -0500, David Young wrote:
 > > They are.  But rather than have a half-dozen editors, a few of which
 > > you can replace and most of which are specialized to be especially good
 > > at editing for a particular purpose, you propose to replace them all
 > > with one editor which you can't replace at all and which, since it is
 > > necessarily one-size-fits-all, is at best mediocre for most tasks.
 > 
 > One editor plus your preferred key bindings plus command evaluation plus
 > type-through filters equals any specialized editor that you could name,
 > AND a consistent user interface.
 > 
 > Another way to think about this is that the one editor contains only the
 > editor universalities, and not the specialties.

Yes, and therefore has no ability to do special-purpose things that
appear in many of these editors as they appear in practice, such as
handle single-line input intelligently.

 > Avoiding modes is important.  That's why Emacs is not the solution.
 >
 > [...]
 > 
 > You think that if emacs was less modal, it would necessarily be less
 > powerful?

Yes. In order to be able to do anything useful, you have to be able
(at a minimum) to specialize to address the things you're intending to
work with. Is it a photo or a source file? I can provide you with a
program that does not have separate modes for editing photos and
source files, but you won't be able to do much with it.

Similarly, suppose I give you a stateless function that can accept a
value of any type -- any type whatsoever -- and return a value of the
same type. What functions can have this type? Well, it turns out that
if it really accepts any type whatsoever, it knows nothing about what
it's got and cannot do anything with it, so the only such function
that can exist is the identity function.

A Unix shell is not modal in the sense that Emacs is, but it exhibits
a number of similar properties, like the notion of current directory.

 > > Looking at my screen just now, I see
 > > 
 > > - Three IRC sessions
 > 
 > Three documents.  You and the people in the same channel are editing
 > a record of a conversation together.  Some type-through filter
 > intermediates.

Streams are not documents. Confusing the two has led many people down
the garden path.

 > > - My music player
 > 
 > I don't know what you use.  I use iTunes: rows and rows of songs.  In
 > columns: song name, album, performer, duration, genre, track number.
 > Looks like a document to me.  One or more sorts and filters apply.
 > There's some "dressing up" of the columns (rows alternate colors,
 > columns line up).

That's not a document, it's a database query. The database is not a
document either; it's a collection of document metadata.

 > > Take it how you like.  Most of that research - all that I've seen, at
 > > least - has not been looking at good interfaces for individual experts;
 > > it's been been looking at how best to accommodate most people, almost
 > > all novices (explicitly all novices, sometimes) without being too
 > > horrible for any of them.  Not how to be good, but how to be tolerable,
 > > if you will.
 > *snip snip*
 > > Human-factors research doesn't look at that.  It looks at what's good
 > > for _non_-experts to use, what is most usable _without_ needing
 > > adaptation of the user and the interfaec to one another (we call these
 > *snip snip*
 > 
 > Sounds to me like you are badly informed.  I will not hop into the
 > "let's reject human factors" rabbit hole with you.

It is easy to receive the impression that "human factors" means
"making computers user-friendly!!11!1!" because that's what many
people have concentrated on for a long time. It is also what various
disproportionately visible and loud strands of thinking concentrate
on. It is also the case that such work tends to look at the impact on
new or naive users, because doing studies on expert users mostly tells
you only that expert users work better with things they're used to,
flawed or not, than with new stuff, better or not, which isn't
interesting.

This doesn't mean that there isn't useful or interesting stuff to look
at, but it does mean that if you want to show it to people and be
listened to (instead of being ignored like you're a recorded message
from freedesktop.org) you need to provide the substance, and also
avoid certain types of ritualized invocations which make people assume
they know what you're about to say and tune it out.

I'd also like to note that recently I ran across this article:

http://www.theglobeandmail.com/news/technology/tech-news/canadians-dont-know-enough-about-technology-to-prevent-risks-report/article2182867/

which is not of much interest except for the following curious
excerpt:

 : The rapid pace of technological evolution also poses a problem, he
 : added, as people struggle to keep up with new programs on the
 : market. The safeguards put in place today may no longer be effective a
 : few weeks later.
 : 
 : The fact that most technology is relatively easy to learn only adds to
 : the problem, he said.
 : 
 : "You don't need to have a certain level of knowledge in order to work
 : with technology," he said.
 : 
 : "Anyone can use it. As a result, that in and of itself can allow a
 : person to get into areas that are otherwise difficult to manage."

Computers are not user-hostile enough any more!

-- 
David A. Holland
dholland%netbsd.org@localhost