You have valid points, although I think you
took my language more strongly than I intended.
(I stopped short of saying there would be a
religious war, for instance.)
However, I've come to the conclusion that my
final point (which, unfortunately, came to me
last) is the most important. If we don't set up
some kind of international standard now, each
group of non-English speakers will go away and
do their own (incompatible) thing.
If we let this eventuality come to pass, the
problems you pointed out in your post would be
multiplied. In addition to your input problems,
you might have to switch to a different OS,
install different software, and maybe even
connect to a different ISP.
(Tho' not if you run Linux, of course.:-)
Well, you might still have to use a different
ISP.)
If we establish a common technology, we can
deal with the difficulties of language. But if
we don't, we'll have to deal with both
the difficulties of language, and
technology.
Every widely used language has a
transliteration method to ASCII characters. They
might be ugly, and they might be unpopular, but
they work.
ASCII is just easier.
I take your point. Better to use a simple
lowest common denominator that everyone can use,
rather than a monstrosity that isolates people.
However, I still have to disagree. If you are
likely to want the contents of a non-English URL,
you're going to need a non-English viewer of some
kind, and if you already have that, then the input
method is only a tiny step beyond.
Secondly, the vast majority of the world do not
use English. Granted, the English/Roman alphabet
covers many more languages than just English, but
I would still be surprised if the Roman alphabet
is recognizable by a majority of the world.
Thirdly, there would be terrible political
consequences to forcing everyone to use ASCII. It
would very quickly be perverted into an East
versus West issue.
Finally, if no standard for non-English
languages is defined now, it will happen later
nevertheless, except that each country will be
setting up their own separate network for their
own use. This would isolate people more,
not less.
The big problem with your reasoning is that it
it's a double-edged sword. The difficulties
you're predicting as an English speaker are
exactly the difficulties that non-English speakers
are trying to deal with right at this moment.
For instance, I'm of Japanese descent. I can
only imagine the difficulties of some older
Japanese people who know no English, when they
have to type in domain names.
"Microsoft," for example, would be written
phonetically, but it would come out like this in
Japanese:
ma i ku ro so fu to
Even after having described the English
alphabet, how do you explain that the mangled name
above (it's the closest you can get in Japanese)
is to be written like this:
microsoft
Want an even worse example? Try slashdot:
su ra sshu do tto
Actually, that's too phonetic. If I wrote the
equivalents of the actual Japanese letters, it
would come out:
If I were a parent, I'd prefer to have a
mechanism that locks out certain channels, rather
than the V-chip. I would find it much easier
that way, to keep track of what shows my children
have access to.
Seems like an obvious thing, so I wouldn't be
suprised if such a TV exists, but then, I've never
seen one.
"You may charge a fee for the physical act
of transferring a copy, and you may at your option
offer warranty protection in exchange for a
fee"
It is not a physical transfer, it is an
electronic transfer. Physical transfer is disk, CD
etc.
Hmm, this is interesting wording. Is it
talking about a physical copy, or a physical act?
A "physical copy" is pretty clear - a CD or a
box of disks.
But would a "physical act" include labour and
the cost of maintaining the internet link?
Another question for those familiar with the
GPL: does the license explicitly say anything
about charging for the labour involved in
development or packaging (in this case, the cost
of assembling and maintaining a distro)?
Let's be fair. If you're going to use
quotation marks, insert a quotation! I said
"Linux hardware support should focus on
certain pieces." And it was just a suggestion,
besides.
I specifically chose that word because I don't
for a moment believe Linux should abandon all but
a select few kinds of hardware.
I once took great joy in getting Linux working
on a 25MHz 386 with 4MB of RAM. It was awesome.
The machine wasn't very useful, but it was just
awesome that I could get the thing to work. But I
don't think Linux should focus on that
kind of hardware. (Not that it does, currently.)
So, maybe it's time all the interested parties
in Linux-land got together and decided on some set
of hardware specifications that Linux should focus
on.
Once it's decided that Linux hardware support
should focus on certain pieces, it will be much
easier to make configuration simpler for common
users.
Probably, we'll need some kind of non-partisan
standards committee that sets the rules.
Does such a group already exist? (Post links!)
I thought I heard of something like this before,
but I can't remember now. If so, maybe all that
needs to be done is to bolster the profile of this
organization.
It's a valid question, even if it assumes
the computer revolution was shaped like other
revolutions, by a handful of dogmatic leaders
advocating specific principles.
Is any revolution shaped by one
individual (or even a handful)? I'm not certain
that's very often true. Just how much influence
can a single individual have, in changing the
world?
My (completely unenlightened) guess is that
more often than not, revolutions occur when a
significant proportion of the population
already feel that things should be
different. But most people don't want to do
anything about it.
Then, somebody who has the balls stands up and
says it out loud. If there isn't sufficient
support among the people, the individual would get
shunned, ridiculed, or nailed to a cross.
But if there is enough support, the
individual is hailed as the leader of a
revolution. Really, the individual is not
much more than a figurehead.
If there is, indeed, a revolution taking place,
I think we just haven't found the right figurehead
yet.
I'm reminded of the movie, 2010, when
the deteriorating political situation required the
American astronauts to be "recalled" to Discovery.
It's just so eerily believable...
You know, there's a reason it's such a
'convoluted' command, It keeps people from
accidently executing it!.
I think that's hardly the point.
The point is that Ctrl-Alt-Delete is totally
nonsensical from the general user's perspective.
Why on earth should that mean "reset?"
My choice of solution would be a reset button
that you have to hit two, or maybe three times, in
close succession.
You wouldn't even need to document it; I
guarantee you that, when a single push doesn't
work, every single user will respond by
hitting it repeatedly, and before long, they'll
realize that you need to hit it more than once.
Do the creatures pass the Turing Test, being
able to carry on an extended conversation of at
least human ability?
Firstly, the Turing test is intended to
identify intelligence, not mere life, so it
doesn't really apply to this discussion.
Secondly, it is still hotly disputed whether
the Turing test is really any good, even for its
intended purpose.
If you think about it, all it really proves is
that some "thing" is capable of fooling one or
more people into believing that it's intelligent.
Unfortunately, there are all sorts of silly
reasons why people can be fooled, most of which
probably have to do with psychology and empathic
reactions.
There is the telling fact that in contests
where the Turing test has been applied, there have
actually been human beings who failed the
test. Which, of course, is merely a measure of
how poorly the person is able to fool his peers
into believing that he is intelligent.:-P
Seems like whenever there's an article that has
something to with pagers or cell-phones, someone
says something to the effect of "I don't carry any
of these things because it's an intrusion of my
privacy, and besides, I like interacting with
people the old-fashioned way." Often, it's said
in a holier-than-thou way that I find really
annoying.
Now, I'm not accusing you (JJ) of saying this;
the way I read your post, you were just stating
your own opinion, which is quite fair enough. And
you did add a bit about how you interact with your
co-workers, which is positive for this discussion.
But all too often, people say things which
basically boil down to "I don't like pagers and
cell-phones!", which is not particularly
insightful or illuminating. Giving a personal
opinion is all well and good, but this opinion has
already been said by zillions of people zillions
of times, on Slashdot and in countless other
forums. Why not try and add something more
original to the discussion?
(After all, if you don't want to be disturbed,
you can turn it off.)
Sorry for the slight rant.
(For the record, I don't carry a pager or a
cell-phone, but I have nothing against them.)
I really couldn't imagine having a
'Development Appliance' for development work since
every developer I know does things a little bit
differently
Not sure I agree with this.
You only need a general purpose computer for
development if you are running the
program on your own machine. But there's nothing
stopping you from having a "development machine"
to do your coding and a "testbed machine" to do
testing and debugging. In fact, as long as the
interactions between the machines are nice and
streamlined, this might be a much nicer way to do
things. One really nice bonus would be that
there's absolutely no chance of crashing
your OS.
For example, as a developer, I've never needed
a 3D video card as part of my development
environment, and that's unlikely to change, unless
Microsoft goes and makes something silly like "3D
Visual Basic". (Not that I'm a VB programmer; I'm
not.)
Sure, the development machine needs some
flexibility, but there's definitely a lot of junk
you'll never need on such a machine.
If you ever watch a proficient abacus user,
it's totally mesmerizing. For some people it's
faster than using a calculator because of the
ergonomics of the device. With practice, you can
click the little discs around much faster than
you can press buttons.
Of course, fancy things like sines, cosines,
exponents and logarithms are rather hard with an
abacus... But you have to start somewhere.
The typical "CS is software, CE is hardware" is
rather vague and almost stereotypical, but I think
it's reasonably accurate.
As far as software courses go, the core
courses here include Java, data structures and
algorithms, an OS course, and a software
engineering course. 4th year elective courses
include AI, database, distributed systems
programming and more software engineering
courses
I myself graduated from CS at Waterloo a few
years back. (For those who don't know, Waterloo
is, deservedly or not, arguably or not, Canada's
most reputable university for computer-related
studies. It's a couple of hours drive from
Toronto.)
I agree that CE is a bit broader than CS, but
I'd disagree that CE is simply a superset of CS
(as somebody else, not you, has stated).
At Waterloo, the difference is not just one of
focus, but of administration: CS is part of the
Mathematics faculty, and CE is a part of the
Engineering faculty. I understand Waterloo is one
of the few universities that has an entire faculty
devoted to math.
The CS curriculum there includes very math-like
subjects such as computing theory and abstract
machines like Turing machines and finite state
machines. And obviously, there are more pure math
courses.
P.S. From one Waterloo alumnus to another
(soon-to-be), get ready for lots of letters from
UW begging for money...
I know, the headline is a little inflammatory,
it's designed that way...:-)
although the chances are so very small, we
would be risking a lot - an entire ecosphere.
What we call ecological "destruction" is not
usually "destruction" in the common sense.
Usually, an ecosystem just changes, it doesn't get
destroyed. I suspect that true ecological
destruction can only occur due to massive
changes... and I don't even mean on the scale of
the atmospheric pollution we humans are pumping
into the air. I firmly believe that the terran
ecosystem will adapt to accommodate it. (Which is
NOT to say that we shouldn't reduce emissions!)
True ecological destruction will occur when the
sun burns out, or say, a passing nebula renders
the entire solar system poisonous, or a black hole
knocks the Earth out of orbit. Think of it: there
was once a collision with a massive meteorite, and
there were huge changes in the ecosystem (enough
that I grant you could call it
"destruction") but it eventually recovered from
the catastrophe, and we are its result.
The only way of not "destroying" an ecosystem
is not to go there at all. In fact, our
spacecraft may have already carried terran
microbes to Mars. Come to think of it, dust and
other flotsam that drifts away from Earth could
"contaminate" Mars, without any action on our part
at all! Should we wrap our planet with a giant
sheet of plastic, to hold in our terrestrial
germs?
Ecological preservation makes a lot of sense,
but only to a point. Beyond a certain point, it
just becomes an unreasoning attachment to a status
quo... Remember, an ecosystem is
supposed to change; that's what makes it
alive.
Disclaimer: IA-in-No-way-whatsoever-ABiologist.
If I said something idiotic, feel free to set me
straight.
Binary patches would work best if the
compiled code was predictable and modular.
Unfortunately, it depends so much on what CFLAGS
the guy compiling it used that day, the phase of
the moon, and the exact pattern on their t-shirt,
that it would prove complicated.
I'm not sure I follow this. If you're going to
provide binary packages, why not binary patches?
Sure, there are hundreds of different possible
settings when compiling an app, but if you're
going to provide binary packages, you've still got
to choose one of them.
What I'd like to see is a package
patching system. For example, you give it
foo_1.2-1.deb, apply the patch, and out
comes foo_1.2-3.deb.
I'm sure it can't be as simple as a binary diff
between the two files because there's compression
happening, but some multi-stage process of
unpacking, patching, and repacking would probably
work.
And then, if it were integrated into APT, it
could maybe combine an FTP site with a local CD to
reduce download times considerably.
A couple of people have mentioned the Arrow,
presumably because this article appeared in a
Canadian publication. However, it should be noted
that nowhere does it mention any Canadian
involvement in this scramjet research.
Also, I think comparisons with the Arrow are
not very valid.
The Arrow was not a research project on new
technology, it was an engineering project to
produce a working fighter incerceptor that can
enter active service.
It was a military vehicle, which meant that the
market for it was relatively small. Furthermore,
the Canadian requirements on the aircraft were so
specialized that nobody else in the world was
interested in such a beast.
On the other hand, the market for a working
scramjet vehicle would probably be huge.
I'm not going to start speculating why the
Arrow project was nixed, because that would
probably ruffle a lot of Canadian feathers and
start a flame war (it's probably Canada's biggest
conspiracy theory).
Personally, I don't find it fishy that they
cancelled the project; I do find it fishy
that everything related to the project was
destroyed.
Your thought-experiment suffers from a
sample size which is far too low to make it
relevant
For the record, I agree with you. It's almost
meaningless to conclude anything based on
the information we have, but I'm just not going to
let that stop me from having an opinion.:-)
However, for the sake of argument...
It's like me looking at my apartment of 6
rooms and noticing that, as far as I can see,
there's only a computer in my dining room. "This
is suggestive," I might say, "that dining-room
like conditions really are the only situations in
which computers can exist."
OTOH, staying with that analogy, moving your
computer to your bedroom won't cause it to
malfunction. But move a human being to any other
planet on the solar system (without any
technological aids), and his life expectancy will
be greatly shortened.
Slashdot Mirror
You have valid points, although I think you took my language more strongly than I intended. (I stopped short of saying there would be a religious war, for instance.)
However, I've come to the conclusion that my final point (which, unfortunately, came to me last) is the most important. If we don't set up some kind of international standard now, each group of non-English speakers will go away and do their own (incompatible) thing.
If we let this eventuality come to pass, the problems you pointed out in your post would be multiplied. In addition to your input problems, you might have to switch to a different OS, install different software, and maybe even connect to a different ISP.
(Tho' not if you run Linux, of course. :-)
Well, you might still have to use a different
ISP.)
If we establish a common technology, we can deal with the difficulties of language. But if we don't, we'll have to deal with both the difficulties of language, and technology.
--
Every widely used language has a transliteration method to ASCII characters. They might be ugly, and they might be unpopular, but they work.
ASCII is just easier.
I take your point. Better to use a simple lowest common denominator that everyone can use, rather than a monstrosity that isolates people.
However, I still have to disagree. If you are likely to want the contents of a non-English URL, you're going to need a non-English viewer of some kind, and if you already have that, then the input method is only a tiny step beyond.
Secondly, the vast majority of the world do not use English. Granted, the English/Roman alphabet covers many more languages than just English, but I would still be surprised if the Roman alphabet is recognizable by a majority of the world.
Thirdly, there would be terrible political consequences to forcing everyone to use ASCII. It would very quickly be perverted into an East versus West issue.
Finally, if no standard for non-English languages is defined now, it will happen later nevertheless, except that each country will be setting up their own separate network for their own use. This would isolate people more, not less.
--
The big problem with your reasoning is that it it's a double-edged sword. The difficulties you're predicting as an English speaker are exactly the difficulties that non-English speakers are trying to deal with right at this moment.
For instance, I'm of Japanese descent. I can only imagine the difficulties of some older Japanese people who know no English, when they have to type in domain names.
"Microsoft," for example, would be written phonetically, but it would come out like this in Japanese:
Even after having described the English alphabet, how do you explain that the mangled name above (it's the closest you can get in Japanese) is to be written like this:
Want an even worse example? Try slashdot:
Actually, that's too phonetic. If I wrote the equivalents of the actual Japanese letters, it would come out:
Don't ask. It's a quirk of Japanese spelling.
--
The RIAA must buy out the internet.
This is the only reasonable solution.
--
If I were a parent, I'd prefer to have a mechanism that locks out certain channels, rather than the V-chip. I would find it much easier that way, to keep track of what shows my children have access to.
Seems like an obvious thing, so I wouldn't be suprised if such a TV exists, but then, I've never seen one.
--
Hmm, this is interesting wording. Is it talking about a physical copy, or a physical act?
A "physical copy" is pretty clear - a CD or a box of disks.
But would a "physical act" include labour and the cost of maintaining the internet link?
Another question for those familiar with the GPL: does the license explicitly say anything about charging for the labour involved in development or packaging (in this case, the cost of assembling and maintaining a distro)?
--
Couldn't any human being claim prior art if this were to be patented?
Only if you are an existentialist, and you claim to have invented yourself.
--
Let's be fair. If you're going to use quotation marks, insert a quotation! I said "Linux hardware support should focus on certain pieces." And it was just a suggestion, besides.
I specifically chose that word because I don't for a moment believe Linux should abandon all but a select few kinds of hardware.
I once took great joy in getting Linux working on a 25MHz 386 with 4MB of RAM. It was awesome. The machine wasn't very useful, but it was just awesome that I could get the thing to work. But I don't think Linux should focus on that kind of hardware. (Not that it does, currently.)
--
So, maybe it's time all the interested parties in Linux-land got together and decided on some set of hardware specifications that Linux should focus on.
Once it's decided that Linux hardware support should focus on certain pieces, it will be much easier to make configuration simpler for common users.
Probably, we'll need some kind of non-partisan standards committee that sets the rules.
Does such a group already exist? (Post links!) I thought I heard of something like this before, but I can't remember now. If so, maybe all that needs to be done is to bolster the profile of this organization.
--
It's a valid question, even if it assumes the computer revolution was shaped like other revolutions, by a handful of dogmatic leaders advocating specific principles.
Is any revolution shaped by one individual (or even a handful)? I'm not certain that's very often true. Just how much influence can a single individual have, in changing the world?
My (completely unenlightened) guess is that more often than not, revolutions occur when a significant proportion of the population already feel that things should be different. But most people don't want to do anything about it.
Then, somebody who has the balls stands up and says it out loud. If there isn't sufficient support among the people, the individual would get shunned, ridiculed, or nailed to a cross.
But if there is enough support, the individual is hailed as the leader of a revolution. Really, the individual is not much more than a figurehead.
If there is, indeed, a revolution taking place, I think we just haven't found the right figurehead yet.
--
I'm reminded of the movie, 2010, when the deteriorating political situation required the American astronauts to be "recalled" to Discovery. It's just so eerily believable...
How territorial we humans are.
--
You know, there's a reason it's such a 'convoluted' command, It keeps people from accidently executing it!.
I think that's hardly the point.
The point is that Ctrl-Alt-Delete is totally nonsensical from the general user's perspective. Why on earth should that mean "reset?"
My choice of solution would be a reset button that you have to hit two, or maybe three times, in close succession.
You wouldn't even need to document it; I guarantee you that, when a single push doesn't work, every single user will respond by hitting it repeatedly, and before long, they'll realize that you need to hit it more than once.
--
Do the creatures pass the Turing Test, being able to carry on an extended conversation of at least human ability?
Firstly, the Turing test is intended to identify intelligence, not mere life, so it doesn't really apply to this discussion.
Secondly, it is still hotly disputed whether the Turing test is really any good, even for its intended purpose.
If you think about it, all it really proves is that some "thing" is capable of fooling one or more people into believing that it's intelligent. Unfortunately, there are all sorts of silly reasons why people can be fooled, most of which probably have to do with psychology and empathic reactions.
There is the telling fact that in contests where the Turing test has been applied, there have actually been human beings who failed the test. Which, of course, is merely a measure of how poorly the person is able to fool his peers into believing that he is intelligent. :-P
--
Seems like whenever there's an article that has something to with pagers or cell-phones, someone says something to the effect of "I don't carry any of these things because it's an intrusion of my privacy, and besides, I like interacting with people the old-fashioned way." Often, it's said in a holier-than-thou way that I find really annoying.
Now, I'm not accusing you (JJ) of saying this; the way I read your post, you were just stating your own opinion, which is quite fair enough. And you did add a bit about how you interact with your co-workers, which is positive for this discussion.
But all too often, people say things which basically boil down to "I don't like pagers and cell-phones!", which is not particularly insightful or illuminating. Giving a personal opinion is all well and good, but this opinion has already been said by zillions of people zillions of times, on Slashdot and in countless other forums. Why not try and add something more original to the discussion?
(After all, if you don't want to be disturbed, you can turn it off.)
Sorry for the slight rant.
(For the record, I don't carry a pager or a cell-phone, but I have nothing against them.)
--
I really couldn't imagine having a 'Development Appliance' for development work since every developer I know does things a little bit differently
Not sure I agree with this.
You only need a general purpose computer for development if you are running the program on your own machine. But there's nothing stopping you from having a "development machine" to do your coding and a "testbed machine" to do testing and debugging. In fact, as long as the interactions between the machines are nice and streamlined, this might be a much nicer way to do things. One really nice bonus would be that there's absolutely no chance of crashing your OS.
For example, as a developer, I've never needed a 3D video card as part of my development environment, and that's unlikely to change, unless Microsoft goes and makes something silly like "3D Visual Basic". (Not that I'm a VB programmer; I'm not.)
Sure, the development machine needs some flexibility, but there's definitely a lot of junk you'll never need on such a machine.
--
I can't imagine Sealand has very numerous internet connections. Probably there's a single trunk from the U.K. or France.
Could the U.S. pressure one of those governments to stop the telco or ISP that's supplying the connection to Sealand?
Of course, this would simply mean that Sealand would shop around and find another provider, but what else can the RIAA do?
--
Hmm, still sounds a bit rude, though.
--
these SiBuckyballs (suckyballs?)
That doesn't sound very flattering. I'd prefer "slickyballs."
(Yeah, sounds rather rude, but then "suckyballs" could be construed as rude, too.)
--
I'm only half kidding.
If you ever watch a proficient abacus user, it's totally mesmerizing. For some people it's faster than using a calculator because of the ergonomics of the device. With practice, you can click the little discs around much faster than you can press buttons.
Of course, fancy things like sines, cosines, exponents and logarithms are rather hard with an abacus... But you have to start somewhere.
--
To create a rudimentary programming language, all you need is about two dozen symbols. ...
Nonsense. Everyone knows all you need is NAND.
I think we should communicate with our alien neighbours using only NAND.
--
The typical "CS is software, CE is hardware" is rather vague and almost stereotypical, but I think it's reasonably accurate.
As far as software courses go, the core courses here include Java, data structures and algorithms, an OS course, and a software engineering course. 4th year elective courses include AI, database, distributed systems programming and more software engineering courses
I myself graduated from CS at Waterloo a few years back. (For those who don't know, Waterloo is, deservedly or not, arguably or not, Canada's most reputable university for computer-related studies. It's a couple of hours drive from Toronto.)
I agree that CE is a bit broader than CS, but I'd disagree that CE is simply a superset of CS (as somebody else, not you, has stated).
At Waterloo, the difference is not just one of focus, but of administration: CS is part of the Mathematics faculty, and CE is a part of the Engineering faculty. I understand Waterloo is one of the few universities that has an entire faculty devoted to math.
The CS curriculum there includes very math-like subjects such as computing theory and abstract machines like Turing machines and finite state machines. And obviously, there are more pure math courses.
P.S. From one Waterloo alumnus to another (soon-to-be), get ready for lots of letters from UW begging for money...
--
I know, the headline is a little inflammatory, it's designed that way... :-)
although the chances are so very small, we would be risking a lot - an entire ecosphere.
What we call ecological "destruction" is not usually "destruction" in the common sense. Usually, an ecosystem just changes, it doesn't get destroyed. I suspect that true ecological destruction can only occur due to massive changes... and I don't even mean on the scale of the atmospheric pollution we humans are pumping into the air. I firmly believe that the terran ecosystem will adapt to accommodate it. (Which is NOT to say that we shouldn't reduce emissions!)
True ecological destruction will occur when the sun burns out, or say, a passing nebula renders the entire solar system poisonous, or a black hole knocks the Earth out of orbit. Think of it: there was once a collision with a massive meteorite, and there were huge changes in the ecosystem (enough that I grant you could call it "destruction") but it eventually recovered from the catastrophe, and we are its result.
The only way of not "destroying" an ecosystem is not to go there at all. In fact, our spacecraft may have already carried terran microbes to Mars. Come to think of it, dust and other flotsam that drifts away from Earth could "contaminate" Mars, without any action on our part at all! Should we wrap our planet with a giant sheet of plastic, to hold in our terrestrial germs?
Ecological preservation makes a lot of sense, but only to a point. Beyond a certain point, it just becomes an unreasoning attachment to a status quo... Remember, an ecosystem is supposed to change; that's what makes it alive.
Disclaimer: IA-in-No-way-whatsoever-ABiologist. If I said something idiotic, feel free to set me straight.
--
Binary patches would work best if the compiled code was predictable and modular. Unfortunately, it depends so much on what CFLAGS the guy compiling it used that day, the phase of the moon, and the exact pattern on their t-shirt, that it would prove complicated.
I'm not sure I follow this. If you're going to provide binary packages, why not binary patches? Sure, there are hundreds of different possible settings when compiling an app, but if you're going to provide binary packages, you've still got to choose one of them.
What I'd like to see is a package patching system. For example, you give it foo_1.2-1.deb, apply the patch, and out comes foo_1.2-3.deb.
I'm sure it can't be as simple as a binary diff between the two files because there's compression happening, but some multi-stage process of unpacking, patching, and repacking would probably work.
And then, if it were integrated into APT, it could maybe combine an FTP site with a local CD to reduce download times considerably.
--
A couple of people have mentioned the Arrow, presumably because this article appeared in a Canadian publication. However, it should be noted that nowhere does it mention any Canadian involvement in this scramjet research.
Also, I think comparisons with the Arrow are not very valid.
The Arrow was not a research project on new technology, it was an engineering project to produce a working fighter incerceptor that can enter active service.
It was a military vehicle, which meant that the market for it was relatively small. Furthermore, the Canadian requirements on the aircraft were so specialized that nobody else in the world was interested in such a beast.
On the other hand, the market for a working scramjet vehicle would probably be huge.
I'm not going to start speculating why the Arrow project was nixed, because that would probably ruffle a lot of Canadian feathers and start a flame war (it's probably Canada's biggest conspiracy theory).
Personally, I don't find it fishy that they cancelled the project; I do find it fishy that everything related to the project was destroyed.
--
Your thought-experiment suffers from a sample size which is far too low to make it relevant
For the record, I agree with you. It's almost meaningless to conclude anything based on the information we have, but I'm just not going to let that stop me from having an opinion. :-)
However, for the sake of argument...
It's like me looking at my apartment of 6 rooms and noticing that, as far as I can see, there's only a computer in my dining room. "This is suggestive," I might say, "that dining-room like conditions really are the only situations in which computers can exist."
OTOH, staying with that analogy, moving your computer to your bedroom won't cause it to malfunction. But move a human being to any other planet on the solar system (without any technological aids), and his life expectancy will be greatly shortened.
--