I don't know who this "they" is, but RSS is just an XML dialect over HTTP. HTTP has push. That probably just means someone wrote a server that sould keep streams open and a client which could, too. RSS shouldn't be at all difficult to adapt for something like that; write an RFC, throw it on usenet, wait two weeks, and implement it in Mozilla. The rest will follow under the weight of whether people actually want it.
Moderators: don't moderate something insightful that claims X Topic would be revolutionary unless you know about X Topic, please. Push HTTP has been there since day one. It started push, not pull.
There is push HTML. It sucks. The reason RSS can't get spam is that you request things; they're not sent to you. You don't want to introduce a way for a server to reach a specific user.
why not make it work right and handle refreshing server-side?
Because, er, you hold open a few sockets that way. Sockets have overhead. You seem not to realize the kind of resources IRC and Telnet take up per user compared to HTTP.
How about XP 1.1.0? The first digit would stand for a really important revision. The second digit ("1") would indicate that this was a minor functional upgrade from 1.0.0, and the third digit would indicate the number of very minor changes or patches applied . ..
E911 site seems to disagree with you. The Wireless Communications and Safety Act of 1999 (frequently referred to as the 911 Act) is the bill that required national 911 compliance by September '02. It has been required for about a year and a half now.
Pretty much how you'd expect: by making fewer mistakes.
This seems a little like suggesting that a sewing machine isn't more accurate than a human seamstress, because it doesn't *know* about its misstitches. Comprehension isn't the issue. I have deleted legitimate email as spam by knee-jerk, and the math above (which leads to 1 in 625 deletions as in error) seems perfectly legitimate to me.
Or are you one of those that thinks that mathematical models aren't useful if they don't have a deep love for and brotherly understanding of what they're approximating?
I'm confused. Why is it that requiring a network to carry emergency services equates in the average slashdotter's mind to unwanted regulation? They're not taxing, they're not restricting, and frankly, I think the extra tenth of a cent per month each person has to shell out is responsible, when you're covering for things like fire, burglary and murder.
Consider that every telephone in the nation on the traditional network - even ones shut off for nonpayment! - must respond to 911. So, you're in a horror movie, out in the forest, being chased by a murderer, and the writer thinks it'd be cute to send you into a shack after a phone, only to have it be disconnected, so that your perfectly reasonable civilized response is useless.
In the real world, that doesn't happen. If the phone company shuts off your line, they must still respond to calls to the operator, to 911, and to repair (and they usually also respond to calls to the business office for obvious reasons.) This is a rational behavior and the law requires it as a safety measure.
I think it's quite the appropriate thing to require this of VoIP providers, just as they required it of cell phone providers. Save your battle cries and sabre-rattling for when they do bad things. Go yell at SCO or something.
Close. The premise is that ancient earth civilizations are the societies structured according to the whims each of one, or occasionally two gou'auld. The reason the cultures are mimiced in other planets is that thouse gou'auld hold sway there too; you'll note that the randomness only applies while they're trying unknown addresses which former gou'auld have lost control of. By comparison, there are some planets where the Gou'auld have not had influence, and even a few where alternate aliens have had input.
The only reason we didn't stagnate like those other planets is that we managed to banish our gou'auld controllers (which, over time, is starting to look more and more false; I believe we've had one all along, lying quietly among us.)
Especially given that elsewhere in the article they quote the addressable memory range at one quarter that
No they do not...
With support for up to 256MB of RAM and 256MB of ROM
That specifically refers only to RAM and ROM, which adds up to double what you claim.
Ah, yes, I forgot how useful ROM was for storing video memory, especially with its high speed write capabilities. So clearly I was a fool not to include ROM in mention of how much Video RAM was available. Silly me.
All palm did was say "let's give them a huge potential screen aperture so they can do weird stuff. 15 bits sounds good, hurf hurf, let's go."
You speak with such authority!
Tu quoque, I see. Occam's razor. Got a better explanation?
It's no more complex than that.
Back it up then.
Rephrased for demonstration purposes: "There's nothing more to say." "Tell us more."
Stop speculating on how absurd the maximum resultant screen could maybe someday be.
A gigapixel on a 3 inch screen will ALWAYS BE ABSURD unless it is to be projected.
No, it isn't. That's about 10,000 dpi, or roughly eight and a half times the linear resolution of a $30 inkjet printer. Most of the prototype digital ink projects are 1600-2000 dpi already. That sort of resolution has a big impact on the quality of deep color images and on text. That sort of resolution would have a tremendous impact on zoomed documents, which on a very small screen like a PDA is often a critical issue.
What's more, what good is an OS capable of addressing a 2 Gbyte display (as an example, 32,768x32,768x16bit) when it can only address 256Mbytes of RAM!
That's exactly what I just told you, and you tried to tell me I had to count the rom and that I was off by a factor of two. Now, you're informing me? Are you drunk, or do you post a bit at a time over the course of a day?
Garbage. Combining gigapixel display ability with 256MB main memory limit IS NOT FUTURE PROOFING!
Apparently nobody's informed you that not all screens are square, and that especially in industry dramatically lopsided aspect ratio screens are moderately common. Or maybe you're just too dense to realize that there's no good reason to make it any smaller. I do find it amusing, though, that my suggestion that this was a failure to handle future proofing in an appropriate fashion was met with "this is not future proofing." I wonder if you're also the type which believes that if you just set the bar high by today's manufacturing cost standards that nobody six years down the road will find a way and then a reason.
I did not say they would. I said, gigapixel 3 inch displays (implying those to be viewed with the naked eye without projection) are absurd.
It's quite annoying when someone takes a statement you make and tries to suggest that it's worthless because it's tangent to something they earlier said and therefore isn't germane. I wasn't rebutting you. You carried little content. I was bringing up a new issue. I don't give a damn what you said. Gigapixel screens won't seem absurd at all in ten years, when our reaction will be "of course it's a gigapixel, it's got to look at least as good as print."
It's funny how now you're suggesting that I should have taken into account what you didn't say about projection. I don't need to; I was simply not presuming to know what sorts of screens PDAs will use in ten years. I don't believe in projection; it needs a flat white surface (those fog screens may prove me wrong, but I wouldn't want my screen reservoir to run dry.) Maybe it'll be a foldable OLED screen. Maybe it'll be holographic. Maybe it'll be in our eyewear. Maybe it'll be traditional.
The point is, you're making the hasty assumption that in ten years, Palm devices will be limited to a three inch screen. Don't do that. They're only limited to that right now b
"I was basically unlucky. I got two jobs on the cusp of the companies shutting down, one of which moved me out of state. Neither had anything to do with my job performance; note that the companies disappear on the same date. Instead, please look at my resume from before the dot com bubble made hiring practices frequently unethical."
Despite all that Americans believe that the guilty should be punished. That people should be held responsible for for their actions. That evil should not be allowed to triumph.
Shut up, Darl.
Yes, we should punish those which fuck around - that's what keeps others from fucking around.
You can't be serious.
on
Practical C++
·
· Score: 4, Informative
The Que book is littered with omissions and errors. If you want to learn C++, start with the free Bruce Eckel e-book Thinking in C++, then move on to the Meyers trio, the Sutter pair, Gang of Four, Dewhurst, Alexandrescu, then Agile Software Development, in that order.
Is the screen frame buffer using main memory and thus within the main memory addressable range? Perhaps this is merely hyped up marketing speak for "1GB addressable memory"?
Especially given that elsewhere in the article they quote the addressable memory range at one quarter that, I'm inclined to suggest that unlikely. All palm did was say "let's give them a huge potential screen aperture so they can do weird stuff. 15 bits sounds good, hurf hurf, let's go."
It's no more complex than that. Stop speculating on how absurd the maximum resultant screen could maybe someday be.
Besides, BillG once thought that having over a half million positions in RAM for a home computer would be useless. Palm's setting the same sort of "we'll fix it ten years down the road" limit here. Gigapixel screens won't seem absurd at all in ten years, when our reaction will be "of course it's a gigapixel, it's got to look at least as good as print."
Mmm. You're right to say that. I actually tend to agree that making a new unit prefix system, which was done 35 years ago, is the right thing to do. However, nobody knows what a kibibyte is.
I tend to disagree with this. If they only gave the gee-whiz, then yes, like you, I'd be inlined to think them a charlatan. But if what he does is to first give all of the germane numbers, and then say "because 31,288,536,429 isn't a particularly natural number, here, watch me download the librar of congress. Again. Again. Again." in order to drive it home, that seems perfectly reasonable.
It's one thing for you to give marketers numbers, and another entirely to give them a demonstration. When you sell digging equipment to farmers, after you tell them how many PSI it can apply upwards from ground and how much gas it eats, you show it ripping up a tree stump with no effort. If you're selling a sports car after you tell them how fast it goes and how much bigger it'll make their dicks look to models, you let them take it for a test drive. If you're selling a whole new kind of videogame that shows up in three "episodes," after telling them it supports the newest phong quincunx mipmap shaded gouraud balanced polytrimesh b-spline depth-ordered vertex renderers, you let them take episode one for free. If you're selling heavy weaponry to the villian, after describing its ass kickenation potential, you demonstrate on innocent villagers or the hero's friends. If you're selling bodybuilding supplements or other addictive drugs, here kid, the first hit's free.
A demonstration is a critical part of the selling process. It's equally easy to suggest that someone is hiding behind the numbers as it is to suggest that they're hiding behind the demonstrations. That's called a straw man. If you don't trust your vendor enough to demo a line they just installed, you shouldn't be buying 2.5g lines from them.
Oh yeah, because clearly if you sell a line that you know what to do with for 5,000 people, then you'd better damned well know how to spend it at one box.
Y'know, instead of wasting it, there is *one* possible solution. I'm no marketer, so I don't know if this is feasable, and I also don't know if you can put the kind of computational horsepower at that point that would make this realistic.
I'm assuming that if they need that kind of line, they have a bunch of machines behind it, maybe a lot of which are available at the moment. Universities, corporations deploying a new building, and virtual desktop office spaces which don't get used during certain hours might provide a reasonable arena for this, and for all I know, your test rig might, too.
Let's say your install is scheduled to be done Monday, and the demo is on Wednesday. After your Monday tests are over, point all of your spare cycles at Seti@Home. Create a new team called "MondayWednesdayCorporateSanta" and get it a significant rank by the time the suits come in. If you turn to them on wednesday, show them you pulling down 200 DVDs at once, then show them an estimate of how many 56k modems and how many cable modems they're looking at, that's a start, great. The next thing to do is to house maybe two dozen DVDs on such a line at the home office, and let the suits download an entire movie in seconds (do that to a laptop, and immediately remove the laptop from the network; next, make a big deal about how the laptop wasn't about to access even one thirtieth of a percent of the total pipeline, and how if there were three hundred workers in here doing that at once it wouldn't slow down.
Ah, but then the killer. "In fact, gentlemen, this line is so large that I couldn't think of anything one computer could do to come anywhere near taxing it, and that makes presenting it somewhat difficult. I took the liberty of running the SETI@Home client, a very popular distributed computing client which has now been receiving donated cycles from networks, clusters, vector computers and supercomputers for six years. They rank their participants, some of whom have been donating supercomputer time since day one. We've been at it since (check watch) Monday night, about 52 hours ago. Out of 8,000 participants, we're already ranked #23. This, my friends, is too large of a line to describe."
Honestly, I'd rather see it go towards the protien folding effort rather than seti@home. However, if one of the suits gets impressed upon them the computational resources they could be bending towards the public and smells the PR (I mean, all they'd be spending was electric, which is a hell of a lot cheaper than a PBS spot,) they might be convinced to donate a corporate superteam. Get a couple of corporations in competition, and Wham! Massive data crunching power out of nowhere, make the suits PR happy. Good karma.
So, yeah, treat it as that last gem to stack on top of your demo presentations. Note the plural - just one won't be impressive. But in the effort to build things to wow the Management, do some good.
Man, this would be insightful if what you were claiming was a boundary power of two. Unfortunately for you, that's 2^15, not 2^16, and the whole-screen resolution is 2^30, not 2^32.
I think the real issue here is that someone at Palm said "hey, let's not impose stupid limits on the platform for five generations from now" and everyone went "What? i can't count that high. That am be dumm. Let's guess why! It's got to be... uh... about... uh... the *machine*. Yeah."
Of course, the astute notice that 32,000 isn't 32k, that a word isn't nessecarily (soon even frequently) 32 bits, and that colors don't always come nearly packaged in machine-sized powers of two.
Slashdot Mirror
Of course, that's exactly what the article said, so the moderator that marked this interesting should be shot.
All true such comparisons must be made between NetHack and Angband. Thank you for playing; you may apply for the Nerd Badge rank six again next month.
I don't know who this "they" is, but RSS is just an XML dialect over HTTP. HTTP has push. That probably just means someone wrote a server that sould keep streams open and a client which could, too. RSS shouldn't be at all difficult to adapt for something like that; write an RFC, throw it on usenet, wait two weeks, and implement it in Mozilla. The rest will follow under the weight of whether people actually want it.
A real-time form of HTML would be a completely new concept altogether.
Oh, sure. If it's 1997. Try checking. Section 8.1.1, page 43.
Moderators: don't moderate something insightful that claims X Topic would be revolutionary unless you know about X Topic, please. Push HTTP has been there since day one. It started push, not pull.
Why not set up a real-time form of html?
There is push HTML. It sucks. The reason RSS can't get spam is that you request things; they're not sent to you. You don't want to introduce a way for a server to reach a specific user.
why not make it work right and handle refreshing server-side?
Because, er, you hold open a few sockets that way. Sockets have overhead. You seem not to realize the kind of resources IRC and Telnet take up per user compared to HTTP.
How about XP 1.1.0? The first digit would stand for a really important revision. The second digit ("1") would indicate that this was a minor functional upgrade from 1.0.0, and the third digit would indicate the number of very minor changes or patches applied . . .
So we're on XP 1.3.28514 ?
Ahahhaa. Since when is marathon deleting spam on monday morning anything but the norm?
I presume your company doesn't put your email address up on the site, then.
E911 site seems to disagree with you. The Wireless Communications and Safety Act of 1999 (frequently referred to as the 911 Act) is the bill that required national 911 compliance by September '02. It has been required for about a year and a half now.
(please don't reply telling me how)
Kinda makes you wonder how they could get from San Diego to Los Angeles in under three hours. Please don't reply telling me how.
(Extra joke for southern californians: wait for the first person to say "the highway!" Laugh, rinse, repeat.)
Pretty much how you'd expect: by making fewer mistakes.
This seems a little like suggesting that a sewing machine isn't more accurate than a human seamstress, because it doesn't *know* about its misstitches. Comprehension isn't the issue. I have deleted legitimate email as spam by knee-jerk, and the math above (which leads to 1 in 625 deletions as in error) seems perfectly legitimate to me.
Or are you one of those that thinks that mathematical models aren't useful if they don't have a deep love for and brotherly understanding of what they're approximating?
I'm confused. Why is it that requiring a network to carry emergency services equates in the average slashdotter's mind to unwanted regulation? They're not taxing, they're not restricting, and frankly, I think the extra tenth of a cent per month each person has to shell out is responsible, when you're covering for things like fire, burglary and murder.
Consider that every telephone in the nation on the traditional network - even ones shut off for nonpayment! - must respond to 911. So, you're in a horror movie, out in the forest, being chased by a murderer, and the writer thinks it'd be cute to send you into a shack after a phone, only to have it be disconnected, so that your perfectly reasonable civilized response is useless.
In the real world, that doesn't happen. If the phone company shuts off your line, they must still respond to calls to the operator, to 911, and to repair (and they usually also respond to calls to the business office for obvious reasons.) This is a rational behavior and the law requires it as a safety measure.
I think it's quite the appropriate thing to require this of VoIP providers, just as they required it of cell phone providers. Save your battle cries and sabre-rattling for when they do bad things. Go yell at SCO or something.
Close. The premise is that ancient earth civilizations are the societies structured according to the whims each of one, or occasionally two gou'auld. The reason the cultures are mimiced in other planets is that thouse gou'auld hold sway there too; you'll note that the randomness only applies while they're trying unknown addresses which former gou'auld have lost control of. By comparison, there are some planets where the Gou'auld have not had influence, and even a few where alternate aliens have had input.
The only reason we didn't stagnate like those other planets is that we managed to banish our gou'auld controllers (which, over time, is starting to look more and more false; I believe we've had one all along, lying quietly among us.)
They pulled a standard-issue play it up like it was the main character and then kill a secondary character routine.
No, the secondary character isn't that grunt, though; in fact, the secondary character is someone whose name you know.
Especially given that elsewhere in the article they quote the addressable memory range at one quarter that
No they do not...
With support for up to 256MB of RAM and 256MB of ROM
That specifically refers only to RAM and ROM, which adds up to double what you claim.
Ah, yes, I forgot how useful ROM was for storing video memory, especially with its high speed write capabilities. So clearly I was a fool not to include ROM in mention of how much Video RAM was available. Silly me.
All palm did was say "let's give them a huge potential screen aperture so they can do weird stuff. 15 bits sounds good, hurf hurf, let's go."
You speak with such authority!
Tu quoque, I see. Occam's razor. Got a better explanation?
It's no more complex than that.
Back it up then.
Rephrased for demonstration purposes: "There's nothing more to say." "Tell us more."
Stop speculating on how absurd the maximum resultant screen could maybe someday be.
A gigapixel on a 3 inch screen will ALWAYS BE ABSURD unless it is to be projected.
No, it isn't. That's about 10,000 dpi, or roughly eight and a half times the linear resolution of a $30 inkjet printer. Most of the prototype digital ink projects are 1600-2000 dpi already. That sort of resolution has a big impact on the quality of deep color images and on text. That sort of resolution would have a tremendous impact on zoomed documents, which on a very small screen like a PDA is often a critical issue.
What's more, what good is an OS capable of addressing a 2 Gbyte display (as an example, 32,768x32,768x16bit) when it can only address 256Mbytes of RAM!
That's exactly what I just told you, and you tried to tell me I had to count the rom and that I was off by a factor of two. Now, you're informing me? Are you drunk, or do you post a bit at a time over the course of a day?
Garbage. Combining gigapixel display ability with 256MB main memory limit IS NOT FUTURE PROOFING!
Apparently nobody's informed you that not all screens are square, and that especially in industry dramatically lopsided aspect ratio screens are moderately common. Or maybe you're just too dense to realize that there's no good reason to make it any smaller. I do find it amusing, though, that my suggestion that this was a failure to handle future proofing in an appropriate fashion was met with "this is not future proofing." I wonder if you're also the type which believes that if you just set the bar high by today's manufacturing cost standards that nobody six years down the road will find a way and then a reason.
I did not say they would. I said, gigapixel 3 inch displays (implying those to be viewed with the naked eye without projection) are absurd.
It's quite annoying when someone takes a statement you make and tries to suggest that it's worthless because it's tangent to something they earlier said and therefore isn't germane. I wasn't rebutting you. You carried little content. I was bringing up a new issue. I don't give a damn what you said. Gigapixel screens won't seem absurd at all in ten years, when our reaction will be "of course it's a gigapixel, it's got to look at least as good as print."
It's funny how now you're suggesting that I should have taken into account what you didn't say about projection. I don't need to; I was simply not presuming to know what sorts of screens PDAs will use in ten years. I don't believe in projection; it needs a flat white surface (those fog screens may prove me wrong, but I wouldn't want my screen reservoir to run dry.) Maybe it'll be a foldable OLED screen. Maybe it'll be holographic. Maybe it'll be in our eyewear. Maybe it'll be traditional.
The point is, you're making the hasty assumption that in ten years, Palm devices will be limited to a three inch screen. Don't do that. They're only limited to that right now b
"I was basically unlucky. I got two jobs on the cusp of the companies shutting down, one of which moved me out of state. Neither had anything to do with my job performance; note that the companies disappear on the same date. Instead, please look at my resume from before the dot com bubble made hiring practices frequently unethical."
Despite all that Americans believe that the guilty should be punished. That people should be held responsible for for their actions. That evil should not be allowed to triumph.
Shut up, Darl.
Yes, we should punish those which fuck around - that's what keeps others from fucking around.
The Que book is littered with omissions and errors. If you want to learn C++, start with the free Bruce Eckel e-book Thinking in C++, then move on to the Meyers trio, the Sutter pair, Gang of Four, Dewhurst, Alexandrescu, then Agile Software Development, in that order.
See Accu's booklist, EfNet #c++'s book list, or Yechiel Kimchi's list of bad books for opposing opinions.
Nope, I didn't. I realize it's a long sentence, but do try to read it until you grasp it. :)
Is the screen frame buffer using main memory and thus within the main memory addressable range? Perhaps this is merely hyped up marketing speak for "1GB addressable memory"?
Especially given that elsewhere in the article they quote the addressable memory range at one quarter that, I'm inclined to suggest that unlikely. All palm did was say "let's give them a huge potential screen aperture so they can do weird stuff. 15 bits sounds good, hurf hurf, let's go."
It's no more complex than that. Stop speculating on how absurd the maximum resultant screen could maybe someday be.
Besides, BillG once thought that having over a half million positions in RAM for a home computer would be useless. Palm's setting the same sort of "we'll fix it ten years down the road" limit here. Gigapixel screens won't seem absurd at all in ten years, when our reaction will be "of course it's a gigapixel, it's got to look at least as good as print."
Mmm. You're right to say that. I actually tend to agree that making a new unit prefix system, which was done 35 years ago, is the right thing to do. However, nobody knows what a kibibyte is.
I tend to disagree with this. If they only gave the gee-whiz, then yes, like you, I'd be inlined to think them a charlatan. But if what he does is to first give all of the germane numbers, and then say "because 31,288,536,429 isn't a particularly natural number, here, watch me download the librar of congress. Again. Again. Again." in order to drive it home, that seems perfectly reasonable.
It's one thing for you to give marketers numbers, and another entirely to give them a demonstration. When you sell digging equipment to farmers, after you tell them how many PSI it can apply upwards from ground and how much gas it eats, you show it ripping up a tree stump with no effort. If you're selling a sports car after you tell them how fast it goes and how much bigger it'll make their dicks look to models, you let them take it for a test drive. If you're selling a whole new kind of videogame that shows up in three "episodes," after telling them it supports the newest phong quincunx mipmap shaded gouraud balanced polytrimesh b-spline depth-ordered vertex renderers, you let them take episode one for free. If you're selling heavy weaponry to the villian, after describing its ass kickenation potential, you demonstrate on innocent villagers or the hero's friends. If you're selling bodybuilding supplements or other addictive drugs, here kid, the first hit's free.
A demonstration is a critical part of the selling process. It's equally easy to suggest that someone is hiding behind the numbers as it is to suggest that they're hiding behind the demonstrations. That's called a straw man. If you don't trust your vendor enough to demo a line they just installed, you shouldn't be buying 2.5g lines from them.
If they're putting in 2.5g links, they can afford a demo rig with two maxed FCA setups.
... to Brewster's Gigabits?
You can be Richard Pryor. I'll be John Candy. Hemos can be Cecil B. de Mille.
Oh yeah, because clearly if you sell a line that you know what to do with for 5,000 people, then you'd better damned well know how to spend it at one box.
Y'know, instead of wasting it, there is *one* possible solution. I'm no marketer, so I don't know if this is feasable, and I also don't know if you can put the kind of computational horsepower at that point that would make this realistic.
I'm assuming that if they need that kind of line, they have a bunch of machines behind it, maybe a lot of which are available at the moment. Universities, corporations deploying a new building, and virtual desktop office spaces which don't get used during certain hours might provide a reasonable arena for this, and for all I know, your test rig might, too.
Let's say your install is scheduled to be done Monday, and the demo is on Wednesday. After your Monday tests are over, point all of your spare cycles at Seti@Home. Create a new team called "MondayWednesdayCorporateSanta" and get it a significant rank by the time the suits come in. If you turn to them on wednesday, show them you pulling down 200 DVDs at once, then show them an estimate of how many 56k modems and how many cable modems they're looking at, that's a start, great. The next thing to do is to house maybe two dozen DVDs on such a line at the home office, and let the suits download an entire movie in seconds (do that to a laptop, and immediately remove the laptop from the network; next, make a big deal about how the laptop wasn't about to access even one thirtieth of a percent of the total pipeline, and how if there were three hundred workers in here doing that at once it wouldn't slow down.
Ah, but then the killer. "In fact, gentlemen, this line is so large that I couldn't think of anything one computer could do to come anywhere near taxing it, and that makes presenting it somewhat difficult. I took the liberty of running the SETI@Home client, a very popular distributed computing client which has now been receiving donated cycles from networks, clusters, vector computers and supercomputers for six years. They rank their participants, some of whom have been donating supercomputer time since day one. We've been at it since (check watch) Monday night, about 52 hours ago. Out of 8,000 participants, we're already ranked #23. This, my friends, is too large of a line to describe."
Honestly, I'd rather see it go towards the protien folding effort rather than seti@home. However, if one of the suits gets impressed upon them the computational resources they could be bending towards the public and smells the PR (I mean, all they'd be spending was electric, which is a hell of a lot cheaper than a PBS spot,) they might be convinced to donate a corporate superteam. Get a couple of corporations in competition, and Wham! Massive data crunching power out of nowhere, make the suits PR happy. Good karma.
So, yeah, treat it as that last gem to stack on top of your demo presentations. Note the plural - just one won't be impressive. But in the effort to build things to wow the Management, do some good.
Man, this would be insightful if what you were claiming was a boundary power of two. Unfortunately for you, that's 2^15, not 2^16, and the whole-screen resolution is 2^30, not 2^32.
I think the real issue here is that someone at Palm said "hey, let's not impose stupid limits on the platform for five generations from now" and everyone went "What? i can't count that high. That am be dumm. Let's guess why! It's got to be... uh... about... uh... the *machine*. Yeah."
Of course, the astute notice that 32,000 isn't 32k, that a word isn't nessecarily (soon even frequently) 32 bits, and that colors don't always come nearly packaged in machine-sized powers of two.