Mark Gordon said:
You are missing the point, which that one line states. Claiming
that telegraphy is not significant because it is not speech, is
illogical. Usenet is not speech either.
If anything, the fact that both of them (Usenet and telegraph)
can and commonly are converted to speech would support the
opposite of Mr. Dovey's original point. The fact is, there *is*
a comparison between the language used on Usenet and the
language used with telegraphy (and both also relate to the
language used in speech).
What you are posting does not support Mr. Dovey's claim, it proves
it false.
One of my *real* example above was of a group of people using a monitor
to read the written text and having to internally vocalise it so such
things *are* relevant even without the hypothetical bind reader, who is
only hypothetical because the blind readers I have come across were on
other groups.
You're still missing the point. Usenet is not speech. Telegraphy is
not speech. That doesn't mean that neither of them are ever converted
to speech. It also does not mean that there is no relationship between
the language usage patterns that develop on Usenet and on Telegraph
networks. (It does not mean they are not related to the patterns used
in speech either.)
However it is completely relevant.
Nothing you say below makes it relevant. The fact is that what you
describe would *also* be required if you were a telegraph operator.
His statement is simply true, but his point is false because the
logic he used was not valid. And your statements are exactly
what I said above, interesting; but you haven't shown any
validity to his claim that telegraphy is not speech has any
significance at all.
Sometimes the only way I can decode
such things is to literally sit down and try filling in the missing
letters at random and for someone using text to speech this would be
even worse. I never even attempt to read the stuff where numbers are
substituted for letters because in stead of minutes it would be more
like hours for any significant piece of text because I just do not see
the connections, numbers being processed differently from letters in my
mind.
So if you can show me where that would be different between
Usenet and telegraphy, you'll have a point which is indeed
significant.
Compared to things like the above using 'u' in place of 'you' is just
mildly jarring, although 'u' can still double the time it takes me to
read a sentence since it is a break from the patterns that were learnt
with great difficulty.
BTW, the last estimate I saw was that 10% of the general population are
dyslexic and anecdotal evidence suggests that dyslexics tend to be good
at computing so I would not be surprised if the percentage is higher
than 10% amongst C programmers.
Now, what I'd rather know is how idioms affect those who use
software to convert text (whether that is on Usenet or sent via
telegraphy) to speech? I highly suspect that the software works
just about like my brain does... my brain, which has a very
limited capacity, has only managed to "learn" a few of these
idioms. It recognizes 'u' quickly, which may be very different
than your experience. But when I see almost any of these kewl
spellings longer than maybe 4 letters, my brain has a very
*fast* method of handling them... it skips ahead one work and
continues on as if that "word" had not even been there. If the
context makes the meaning obvious, then I understand what was
said; otherwise I don't.
I just don't waste any time trying. That of course works
because, of the things that I'm missing, it is obvious that
99.9% are not worth catching. The percentages may be a very
different for you, which forces you to actually take the time
with many of them. And in the process, you may waste a lot of
time, but you also pick up on a small, but perhaps useful, 0.1%
that I miss entirely.