A challenge to the ASCII proponents.

[Alan Kennedy]

JanC:

>>> The verb "gignooskoo" (trying to write it with Latin letters ;)

Alan Kennedy wrote:

>> Why limit yourself to that nasty little us-ascii alphabet? >;-)

Ben Finney wrote:

> Because it will display reliably on any computer.

I'm not so sure. Depends on what you mean by display I suppose. I have
a luddite classics professor friend who would deride the assertion
that the above is an accurate representation of the original greek
word.

> > Here it is in a format where almost everybody will be able to see the
> > original greek verb on their screen.
> > [instructions to cut and paste to a file, then open in a limited range
> > of programs, on computers possessing the appropriate font]
> >
> > So, the challenge to the ASCII proponents is: put the greek word
> > "gignooskoo" on everybody's screen, originating from a usenet message,
> > in the original greek, where "oo" -> greek letter omega.
> 
> Challenge accepted:
> 
> Open any drawing program.  Draw, in order from left to right, the Greek
> characters gamma, ipsilon, gamma, nu, omega, sigma, kappa, omega.
> 
> Done.  The desired word now appears on the screen.

OK, now that we've solved that problem, let's move it up a level. 

Now we want our greek to be indexable and searchable, so that, for
example, I can go to google and have it returned as a hit for the word
"gignooskoo". (apologies to greek people and greek scholars for the
poor rendering, if you've found this message at all).

And we want our greek to be accessible to visually disabled people.
Hacks like displaying bitmaps instead of glyphs work for visual
rendering. But what about non-visual renderings? Aural renderings?
Braille renderings?

> Oh, what's that -- you say that's cheating because the user has to use
> particular programs?  Perform manual steps?  Have some existing
> knowledge about the process?  That the process may fail for any of these
> reasons?

Not necessarily cheating. Just not scalable (to say, "The Illiad").
And not searchable. Or accessible. Yes, I could automate the process,
by say generating a series of vector commands, which results in
drawing the glyph on the users screen. But it still isn't searchable.

As for particular programs: we all use a limited set of software that
fits our personal paradigm for information modelling. But my process
only involved OS and software-independent concepts, listed below under
"existing knowledge". Also, I think browsers are pretty universal
these days. Note also that your proposed process requires the
availability of drawing software.

Perform manual steps: There'll always be manual steps. I count 12
mouse presses to follow my process. How many for yours?

Existing knowledge: All I need was knowledge of copy&paste, file
creation and file viewing. I might need Pretty basic and universal
computer knowledge, in these days of GUIs.

> Those are attributes of the "simple" process of manually manipulating
> XML content you gave.

Fair enough, if copying and pasting is a complex and error-prone
operation. But it's not. And even the copying and pasting would be
eliminated if I could have the usenet transport protocol encode its
data and metadata in XML.

And yes, it would also be necessary if I could encode protocol
metadata in UTF-8. But I can't. HTTP and MIME, restrict me to 8-bit
character sets like iso-8859-1.

> Not every computer is capable of automatically displaying Greek
> characters.  Even for those which can, there's not yet a universal way
> to instruct them to do so.  Hence, it is not possible to have any
> computer automatically display a word with Greek characters.

> But you already knew that, so why the silly challenge?

To raise the stakes once somebody has "ante'd up" >;-)

-- 
alan kennedy
-----------------------------------------------------
check http headers here: http://xhaus.com/headers
email alan:              http://xhaus.com/mailto/alan