[Regarding the single high power radio transmission toward possible ETs done long ago:]
Wow. So, why'd they stop? Afraid the Ur'Quan are going to stop by or something? heheh
Basically, yes. Some have argued that the odds are tiny that hostile aliens would harass Earth even if they wanted to, but the counter-argument is that we have no defense against any ET advanced enough to even get here in the first place, so why take any chances, no matter how small?
Also, there are counter-arguments to support the position that any ETs that went to the trouble to come here would actually be likely to be hostile.
Obviously no one really knows either way, but prudence essentially costs nothing.
Heck, in 20 years we may discover a more practical way to transmit over vast distances... and suddenly discover aliens are already trying to communicate with us.
Or maybe not. A lot can change in 20 years though.
Sure. Anything can happen. The sun could unexpectedly go nova tomorrow, quickly refuting our understanding of stellar dynamics and nuclear physics.
But the article is talking about prediction, and prediction really has no choice but to extrapolate from what we do know. Otherwise, it's not "prediction", it's "hoping".
So given what we know, the only way to pick up weak ET signals, such as Earth gives off, is to build a very large cryogenic Allen array in space. Like I said. Which unfortunately is not yet planned.
So presumably the author of the paper that this article is about, is simply assuming that ETs are beaming signals straight at us, which is too unlikely to "predict", although we could hope.
My post above basically is politely refuting the whole idea of "predicting" ET contact in 20 years. Without that space array, it's just not going to happen.
But maybe the author also predicts we'll put up such a space array (I kind of doubt it, but let's give the benefit of the doubt).
I'm a big fan of SETI, but they tend to downplay the fact that we're only likely to be able to pick up signals beamed directly at us.
We can't currently pick up ET signals equivalent to what Earth is broadcasting to space, even if they were coming from Alpha Centauri; they're just too weak.
This is an analog problem of signal to noise ratio, far more than anything else, so faster processing won't help a bit.
A cryogenic Allen array (to minimize thermal noise), especially in space far from Earth, or on the far side of the moon, would help a tremendous amount.
Usually discussions about SETI itself don't bring that up, because of issues of optimism and such, but it was easy to find web hits on the eseentially identical question: can ETs pick up Earth signals?
Similar
comments
by John Dreher, Staff Astronomer, SETI Institute,
although he goes on to assume that ETs would be able to pick up weaker signals than humans are able to -- assuming implicitly that ETs will have better analog technology than we do (maybe they do, but that doesn't help us to do the same).
What about ETs actually beaming a signal at us? Maybe they do so to all nearby stars, one by one. Maybe...would we do that?
"...it has been agreed by all relevant groups that we should not be actively sending out messages to try to reach other civilisations", says another
page
Ok, so we would not be so foolish as to attract undue attention from an unknown and possibly hostile galaxy, but maybe ETs will be more naive than that. Or a lot more confident (play ominous music here;-)
So, bottom line, this is a cool topic, but are we planning to build a cryogenic Allen array in space in the next two decades?
I think we should, but any predictions really should be based largely on that one issue.
P.S. the recent lab verification of photons having orbital angular momentum, able to carry arbitrary amounts of information per photon, implies a new medium we'll need to check for ET signals. Maybe that's what all advanced civilizations use.
I've taken the phrase "power user" to mean someone that knows enough to check whether the computer is supplied with electricity before calling tech support.
If they call themselves a "power user", yes. But I'm a professional programmer, so when I call someone that, I mean that I think they know the app quite well, including less obvious tricks.
Anyway, in this case I used the phrase to mean that indeed it was Word screwing up, not that this was a group of people who were simply misusing it because they could barely find the keyboard with both hands.
I don't suppose there's a term you prefer, to describe people who really know what they're doing with an app?
You could say the same [i.e. "swear at"] vi, or emacs, for that matter.
Naw -- While it's true that I've sworn at emacs because I didn't know how to get it to do something, and I've sworn at vi for not having a feature I wanted, this is rather different than swearing at Word for not doing what you tell it to do.
Word is buggy. I knew of exactly 1 serious bug in the original vi (it crashed if a global search/replace pattern wrapped around to the next line), none in vim (maybe I've been lucky), and only minor bugs in the various versions of emacs I've used (not counting the less-used infinite add-ons).
I'm sure that vi and emacs had more bugs than I personally have seen, but my experience is not unusual -- whereas every heavy user of Word becomes keenly aware of its bugs.
That's a significant difference. Bill Gates has made explicit statements about his beliefs and policies about bugs in his products; I'm not flaming, so I won't quote him directly here, but I really do think that the attitude reflected in those famous comments has a direct impact on products like Word.
When I use Windows at work, Word is powerful and pretty nice...if and when it works. It doesn't crash on me, but it does refuse to do what I tell it sometimes; power users get used to doing workarounds, so it's not that big of a deal if you use it every single day -- you memorize its idiosyncracies.
However, several times I've seen a whole group of Word power users (not clueless lusers) need to given up on a document and start over from scratch -- usually just on little things like the company business plan or 12 month road map (urk). The only workaround each time was to copy/paste the original document text into a new Word file, because Word was hopelessly confused by whatever little magic cookies it had left in the original document.
I.e. I know it's not just me being confused, I see this happen to everyone who uses Word heavily on big documents, sooner or later.
To be charitable, this may be the eventual fate of any huge app that grows by accretion from a small program to a hugely enormous giganto app, without being redesigned and recoded and refactored along the way.
So yeah, Word -- nice when it works, I guess, but it can be quite frustrating other times.
The reason why some gay men act effeminate is because being gay frees people from the normal constraints of society: they feel they have the freedom to act effeminate. Society and peer pressure constrains straight male behaviour
Certainly that's part of it for some people, but I've had gay male friends who grew up acting in an effeminate way starting in childhood, and began to wonder if they were gay because of that, and only later discovered a strong sexual preference for the same sex.
The moral being that there's many kinds of people in the world, and a single explanation rarely accounts for all differences. Including biology -- and also that there seem to be multiple biological reasons for these things.
What the community could really use is a few good artists that volunteer to make some of the ugly projects look good
There are many, many talented graphics artists in the world, but almost no one is asking for their help on these projects. Nor does the average programmer know where to go to find artists who are willing to help out.
We need an Open Graphics Art Project to connect together open source programmers with open art artists.
Same thing to a lesser extent with other professions like information architects (often found in the same person as a graphic artists, but not identical), usability/ergonomics, writing, game playability tuning, etc.
Perhaps all it would take is the right web site to help these people find each other.
I would thing that the best thing a library could have is a copy, be it a mirror, dvd, cd, of Project Gutenberg
I actually did do that, I made a multi-CD copy of the entirety of Project Gutenberg and gave it to my sister, a schoolteacher in central Florida, for her gifted classes, and a copy to give to her friend, who's a librarian in that area.
It was a very novel notion to them, so I don't know that it ended up getting used (I'm too many thousands of miles away to check), but I gave it a shot...
the likely attitude that Orson Welles might have had to another director taking one of the works with which he became most closely associated, can only be a matter of conjecture
I think we can predict quite well how he would have reacted to the coming remake of his masterpiece: "Halloween Part 7: Freddy Krueger versus Citizen Kane", starring Geraldo Rivera.
This is Slashdot, where any sufficiently advanced opinion is indistinguishable from fact....A sufficiently advanced opinion is a fact.
[...] At which point it becomes hindsight, and no longer quite so advanced.
In technical jargon (e.g. wave theory), the opposite of "advanced" is "retarded".
They HAD to find some reason to bring the suit up to $5 billion, you know...
They calculated the projected future losses caused by ill will generated in potential customers, losses from counter-lawsuits, contempt of court fees for frivolous lawsuits and fraud, etc, and the previous $3 billion they were asking for wasn't enough to cover it.
Surely you don't distrust common sense..."Wintel server 10 times less expensive to operate than Linux mainframe"...and that's only counting the hardware!
When you throw in the software, that brings up the mainframe cost another $80!
And it is irrelevent to consider the cost of the Windows software, just ask them.
Leave it to Microsoft to discover that mainframes cost more than servers.
dollargonzo wrote:
...are the primary data sets, but both take a lot of processing, and people who (like myself) actually bother to figure out the formats end up being quite stingy with the results.
Please, pretty please, then, open-source your
reformatted results!
I've looked into GIS several times over the years,
hoping to use data for highly nontraditional purposes,
but the formats are indeed a major pain,
so I've always gone away discouraged.
Most of these assumptions about future computing technology (AI aside of course) were underestimating rather than overestimating
Most of these assumptions never happened in the
first place! This article got pulled out of thin
air.
For instance:
In the late 80s I remember it being a well regarded popular 'fact' that 100MHz was the absolute limit for the speed of a CPU
Not! For the speed of a cpu with one particular
technology, sure. Which is true; each technology
has limits and needs to be replaced with a
better technology.
Same with disk sizes. But NO ONE
who had even the tiniest clue thought
that 100Mhz and 1GB were limits established
by physics.
And the
"world market of 5 computers" (A) was just
one person, (B) has been claimed to have
been only an accurate estimate for the coming
year, (C) has been discussed to death.
As for the coming death of Apple, BSD,
and Unix in general, these things have
been predicted endlessly by self-proclaimed
marketing pundits who typically don't
know anything about anything.
This article
is complete crap. And as for its
question:
What are your favorite beliefs-turned-on-their-heads in the history of computing?"
They haven't happened yet. Strong AI will eventually
be my favorite.
The parent article gave a broken link
(all dots and slashes removed from the URL):
One of many links: A Bullshit Detection Guide
The correct link is
http://www.geocities.com/Athens/Oracle/4855/bs.htm;
the page is titled "A BullSh-- detection Guide"
so I hadn't found it in a google search, either
(usually my first line of defense for bad
URLs)
I'd have given 2 cuckoos to tachyons, only 1 cuckoo to time travel
In special relativity, faster than light travel (FTL)
implies time travel quite directly.
So to treat the two subjects as being significantly
different means to be working in a theory other
than relativity.
Special Relativity (SR) is nice and simple but
fairly limited in scope, but agrees extremely
well with experiments within that scope.
Its extension to cover gravity, General Relativity (GR)
is extremely elegant, and also agrees well with
experimental observations, but is not integrated
with the rest of the infrastructure of fundamental
physics (quantum physics, quantum electrodynamics,
the Standard Model...)
So general relativity may eventually become obsolete,
even though currently it's currently a great theory,
and whatever replaces it may modify special relativity
too. So this isn't some kind of absolute statement.
Still, in the absence of a theory that
is trying to supplant relativity, FTL implies
time travel. Presumably the author of the book
knows this, despite listing FTL and time travel
as two different subjects.
For more info see these two sections of the
relativity FAQ:
relativity: time travel
and
relativity: FTL
,
hosted by and partly written by John Baez,
a quantum gravity researcher with impeccable
physics background (I've done some online study
under him; he's also a fantastic teacher).
[ given the highly optimized current crop of
parsers...]
would it be possible for someone to improve the performance another 100x?
If you're talking about XML parsers written in
Java and then interpreted rather than compiled
into native machine code, maybe, by rewriting
in C and then using a slew of optimization
techniques on top of that. Maybe. Probably not,
even then; if it's using certain inefficient
Java constructs, maybe 10 fold to 30 fold faster.
100 fold only if they weren't actually very
optimized.
If it's written in extremely optimized Java
compiled to native
machine code,
it'll still typically be faster yet if it's rewritten well in
C, no matter what Sun Microsystem's propaganda
says, but not always by very much, and sometimes
not at all. Java has certain language features
that tend to impose at least a little bit of
overhead compared with C, and sometimes a lot
of overhead. But not 100-fold kinds of overhead,
and maybe only 20%...it varies.
But if these are already written in C and
highly optimized, then
I strongly doubt it (except of course by getting
faster hardware).
In the version I did, I could've made it
somewhat faster...optimizing more
for the most common cases might have bought
a factor of two, with luck. Probably not,
but maybe.
Paying very close attention to fitting
the core of it into level 1 cache, using exactly 100%
of available registers even if it meant less
clear code, trying very hard to avoid branches,
and then trying even harder (e.g. multiplying by
zero rather than branching, or the equivalent,
depending on the cpu under test),
and then recoding the innermost loop in assembler
language....Typically such things give at least
another 30%, and sometimes you can even get 5 five fold...
but the level 1 cache isn't all that small and
the code isn't all that big...
I'd actually be shocked if a month's work by
a world class guru could make it even 10 times
faster. A 100 fold seems exceedingly unlikely;
I'd bet serious money against it (but only after
looking at the source code of the parser in
question first:-)
Once something has been highly optimized by
one competent group of people, a second group
of people might have new bright ideas, but
they are usually more like 20% faster kinds of
ideas, not 100 fold.
Unless an entirely new algorithm is invented,
but I don't see how that would apply here.
Basically it's very hard to make things fast,
and most of the time, a 20% increase
in speed is doing very well indeed...unless
something had unusual problems to start with --
which paradoxically has always been moderately
common, and is even more so now that machines
have gotten so fast, thanks to Moore's law.
People often pessimize rather than optimize,
these days.
Slashdot Mirror
Wow. So, why'd they stop? Afraid the Ur'Quan are going to stop by or something? heheh
Basically, yes. Some have argued that the odds are tiny that hostile aliens would harass Earth even if they wanted to, but the counter-argument is that we have no defense against any ET advanced enough to even get here in the first place, so why take any chances, no matter how small?
Also, there are counter-arguments to support the position that any ETs that went to the trouble to come here would actually be likely to be hostile.
Obviously no one really knows either way, but prudence essentially costs nothing.
The Ur'Quan Masters
Sure. Anything can happen. The sun could unexpectedly go nova tomorrow, quickly refuting our understanding of stellar dynamics and nuclear physics.
But the article is talking about prediction, and prediction really has no choice but to extrapolate from what we do know. Otherwise, it's not "prediction", it's "hoping".
So given what we know, the only way to pick up weak ET signals, such as Earth gives off, is to build a very large cryogenic Allen array in space. Like I said. Which unfortunately is not yet planned.
So presumably the author of the paper that this article is about, is simply assuming that ETs are beaming signals straight at us, which is too unlikely to "predict", although we could hope.
My post above basically is politely refuting the whole idea of "predicting" ET contact in 20 years. Without that space array, it's just not going to happen.
But maybe the author also predicts we'll put up such a space array (I kind of doubt it, but let's give the benefit of the doubt).
We can't currently pick up ET signals equivalent to what Earth is broadcasting to space, even if they were coming from Alpha Centauri; they're just too weak.
This is an analog problem of signal to noise ratio, far more than anything else, so faster processing won't help a bit.
A cryogenic Allen array (to minimize thermal noise), especially in space far from Earth, or on the far side of the moon, would help a tremendous amount.
Usually discussions about SETI itself don't bring that up, because of issues of optimism and such, but it was easy to find web hits on the eseentially identical question: can ETs pick up Earth signals?
"No", says this Seti League guest editorial "ET Detection of Earth TV Unlikely" that goes into a little technical detail.
Similar comments by John Dreher, Staff Astronomer, SETI Institute, although he goes on to assume that ETs would be able to pick up weaker signals than humans are able to -- assuming implicitly that ETs will have better analog technology than we do (maybe they do, but that doesn't help us to do the same).
What about ETs actually beaming a signal at us? Maybe they do so to all nearby stars, one by one. Maybe...would we do that?
"...it has been agreed by all relevant groups that we should not be actively sending out messages to try to reach other civilisations", says another page
Ok, so we would not be so foolish as to attract undue attention from an unknown and possibly hostile galaxy, but maybe ETs will be more naive than that. Or a lot more confident (play ominous music here ;-)
So, bottom line, this is a cool topic, but are we planning to build a cryogenic Allen array in space in the next two decades?
I think we should, but any predictions really should be based largely on that one issue.
P.S. the recent lab verification of photons having orbital angular momentum, able to carry arbitrary amounts of information per photon, implies a new medium we'll need to check for ET signals. Maybe that's what all advanced civilizations use.
See e.g. Photons Spin More Data
I'm interested, for more or less the same reason as this John guy, so do please let me know at the email address in the header above. Thanks!
If they call themselves a "power user", yes. But I'm a professional programmer, so when I call someone that, I mean that I think they know the app quite well, including less obvious tricks.
Anyway, in this case I used the phrase to mean that indeed it was Word screwing up, not that this was a group of people who were simply misusing it because they could barely find the keyboard with both hands.
I don't suppose there's a term you prefer, to describe people who really know what they're doing with an app?
Naw -- While it's true that I've sworn at emacs because I didn't know how to get it to do something, and I've sworn at vi for not having a feature I wanted, this is rather different than swearing at Word for not doing what you tell it to do.
Word is buggy. I knew of exactly 1 serious bug in the original vi (it crashed if a global search/replace pattern wrapped around to the next line), none in vim (maybe I've been lucky), and only minor bugs in the various versions of emacs I've used (not counting the less-used infinite add-ons).
I'm sure that vi and emacs had more bugs than I personally have seen, but my experience is not unusual -- whereas every heavy user of Word becomes keenly aware of its bugs.
That's a significant difference. Bill Gates has made explicit statements about his beliefs and policies about bugs in his products; I'm not flaming, so I won't quote him directly here, but I really do think that the attitude reflected in those famous comments has a direct impact on products like Word.
However, several times I've seen a whole group of Word power users (not clueless lusers) need to given up on a document and start over from scratch -- usually just on little things like the company business plan or 12 month road map (urk). The only workaround each time was to copy/paste the original document text into a new Word file, because Word was hopelessly confused by whatever little magic cookies it had left in the original document.
I.e. I know it's not just me being confused, I see this happen to everyone who uses Word heavily on big documents, sooner or later.
To be charitable, this may be the eventual fate of any huge app that grows by accretion from a small program to a hugely enormous giganto app, without being redesigned and recoded and refactored along the way.
So yeah, Word -- nice when it works, I guess, but it can be quite frustrating other times.
Certainly that's part of it for some people, but I've had gay male friends who grew up acting in an effeminate way starting in childhood, and began to wonder if they were gay because of that, and only later discovered a strong sexual preference for the same sex.
The moral being that there's many kinds of people in the world, and a single explanation rarely accounts for all differences. Including biology -- and also that there seem to be multiple biological reasons for these things.
True, people "borrow" art a lot, but are you just blowing off some steam, or do you have an idea for how an open art project can help with that issue?
There are many, many talented graphics artists in the world, but almost no one is asking for their help on these projects. Nor does the average programmer know where to go to find artists who are willing to help out.
We need an Open Graphics Art Project to connect together open source programmers with open art artists.
Same thing to a lesser extent with other professions like information architects (often found in the same person as a graphic artists, but not identical), usability/ergonomics, writing, game playability tuning, etc.
Perhaps all it would take is the right web site to help these people find each other.
offtopic?? Can moderators read?
Oldest trick in the book: stir up controversy by being jerks just to get free publicity.
I actually did do that, I made a multi-CD copy of the entirety of Project Gutenberg and gave it to my sister, a schoolteacher in central Florida, for her gifted classes, and a copy to give to her friend, who's a librarian in that area.
It was a very novel notion to them, so I don't know that it ended up getting used (I'm too many thousands of miles away to check), but I gave it a shot...
See article
Objective C borrows both syntax and semantics from Smalltalk (on top of C itself, of course). It's a nice small language.
I think we can predict quite well how he would have reacted to the coming remake of his masterpiece: "Halloween Part 7: Freddy Krueger versus Citizen Kane", starring Geraldo Rivera.
In technical jargon (e.g. wave theory), the opposite of "advanced" is "retarded".
That happened well over twelve seconds ago, where have you been???
They calculated the projected future losses caused by ill will generated in potential customers, losses from counter-lawsuits, contempt of court fees for frivolous lawsuits and fraud, etc, and the previous $3 billion they were asking for wasn't enough to cover it.
Surely you don't distrust common sense..."Wintel server 10 times less expensive to operate than Linux mainframe"...and that's only counting the hardware! When you throw in the software, that brings up the mainframe cost another $80! And it is irrelevent to consider the cost of the Windows software, just ask them. Leave it to Microsoft to discover that mainframes cost more than servers.
Please, pretty please, then, open-source your reformatted results!
I've looked into GIS several times over the years, hoping to use data for highly nontraditional purposes, but the formats are indeed a major pain, so I've always gone away discouraged.
So help the world out, publish your stuff!
Most of these assumptions never happened in the first place! This article got pulled out of thin air. For instance:
Not! For the speed of a cpu with one particular technology, sure. Which is true; each technology has limits and needs to be replaced with a better technology.
Same with disk sizes. But NO ONE who had even the tiniest clue thought that 100Mhz and 1GB were limits established by physics.
And the "world market of 5 computers" (A) was just one person, (B) has been claimed to have been only an accurate estimate for the coming year, (C) has been discussed to death.
As for the coming death of Apple, BSD, and Unix in general, these things have been predicted endlessly by self-proclaimed marketing pundits who typically don't know anything about anything.
This article is complete crap. And as for its question:
They haven't happened yet. Strong AI will eventually be my favorite.
The correct link is http://www.geocities.com/Athens/Oracle/4855/bs.htm
;
the page is titled "A BullSh-- detection Guide"
so I hadn't found it in a google search, either
(usually my first line of defense for bad
URLs)
In special relativity, faster than light travel (FTL) implies time travel quite directly.
So to treat the two subjects as being significantly different means to be working in a theory other than relativity.
Special Relativity (SR) is nice and simple but fairly limited in scope, but agrees extremely well with experiments within that scope.
Its extension to cover gravity, General Relativity (GR) is extremely elegant, and also agrees well with experimental observations, but is not integrated with the rest of the infrastructure of fundamental physics (quantum physics, quantum electrodynamics, the Standard Model...)
So general relativity may eventually become obsolete, even though currently it's currently a great theory, and whatever replaces it may modify special relativity too. So this isn't some kind of absolute statement.
Still, in the absence of a theory that is trying to supplant relativity, FTL implies time travel. Presumably the author of the book knows this, despite listing FTL and time travel as two different subjects.
For more info see these two sections of the relativity FAQ: relativity: time travel and relativity: FTL , hosted by and partly written by John Baez, a quantum gravity researcher with impeccable physics background (I've done some online study under him; he's also a fantastic teacher).
If you're talking about XML parsers written in Java and then interpreted rather than compiled into native machine code, maybe, by rewriting in C and then using a slew of optimization techniques on top of that. Maybe. Probably not, even then; if it's using certain inefficient Java constructs, maybe 10 fold to 30 fold faster. 100 fold only if they weren't actually very optimized.
If it's written in extremely optimized Java compiled to native machine code, it'll still typically be faster yet if it's rewritten well in C, no matter what Sun Microsystem's propaganda says, but not always by very much, and sometimes not at all. Java has certain language features that tend to impose at least a little bit of overhead compared with C, and sometimes a lot of overhead. But not 100-fold kinds of overhead, and maybe only 20%...it varies.
But if these are already written in C and highly optimized, then I strongly doubt it (except of course by getting faster hardware). In the version I did, I could've made it somewhat faster...optimizing more for the most common cases might have bought a factor of two, with luck. Probably not, but maybe.
Paying very close attention to fitting the core of it into level 1 cache, using exactly 100% of available registers even if it meant less clear code, trying very hard to avoid branches, and then trying even harder (e.g. multiplying by zero rather than branching, or the equivalent, depending on the cpu under test), and then recoding the innermost loop in assembler language....Typically such things give at least another 30%, and sometimes you can even get 5 five fold... but the level 1 cache isn't all that small and the code isn't all that big...
I'd actually be shocked if a month's work by a world class guru could make it even 10 times faster. A 100 fold seems exceedingly unlikely; I'd bet serious money against it (but only after looking at the source code of the parser in question first :-)
Once something has been highly optimized by one competent group of people, a second group of people might have new bright ideas, but they are usually more like 20% faster kinds of ideas, not 100 fold.
Unless an entirely new algorithm is invented, but I don't see how that would apply here.
Basically it's very hard to make things fast, and most of the time, a 20% increase in speed is doing very well indeed...unless something had unusual problems to start with -- which paradoxically has always been moderately common, and is even more so now that machines have gotten so fast, thanks to Moore's law. People often pessimize rather than optimize, these days.