Even proprietary software has always had a service-oriented business model. Typically you're selling the future upgrades to the user as much as the current "completed" version. Open source tends to smooth out the revision process, but it's the same idea.
Even one-off programming like games is really just a service; it's just done on spec ahead of time, in the hope that people will pay you to instruct their computer how to do stuff (animate aliens decapitating people or whatever) later on. Software was never a "product", and the attempt to sell it as one oughta be pretty much over in five years. And when you think about it it was a pretty short burp - people have been selling the service of computer programming for about fifty years (Ada Lovelace was a volunteer, but the women who figured out how to program the ENIAC were paid, albeit poorly as human adding machines), and we hopefully have thousands of years left to go where people will no longer be able to try and sell it in shrinkwraps.;)
This really brings home some of the commentary in the recent article thread concerning the consequences of poor interface design. Just the site of that nine-tabbed-panel-hell window makes my brain want to turn off, and I'm only reading a BBC article, not trying to decide issues of law and order in the heat of the situation. It's hard enough for judges to interpret and apply the law on sober courtroom consideration; asking them to fight with a typical VB interface in order to do their (difficult) job is just plain cruel. And not just to them, but to the people whose fates they're deciding. I know that poor Windows and X interfaces impede my judgement, and I'm usually judging the computer itself. It seems to me that this sort of application really begs a custom Palm-style or Mac-ish simplified interface, and Brazilians would probably be better off if they'd waited to develop one.
The stability issues are so obvious they don't really bear further mentioning (beyond the rather redundant two times in two sentences for "GPF" in the article capsule).
You're right, of course - that's what I get for posting during daylight hours. Sorry about the misinfo, but then the vagaries of shell quoting have bitten us all.
I think what I actually had in mind when I posted that was something still different:
foo='`/bin/ls`' eval echo "stuff $foo stuff"
Which doesn't actually even relate directly to the example. Oops.
I certainly wasn't arguing that computer science and IT work weren't valuable, nor that Unix work was useless or unimportant. If I truly felt that way then I'm sure I'd either be contemplating a new career, or suicide; I continue with it (and will probably move to California soon to take one or another of the jobs I see advertised) precisely because I feel I can do good with it. My argument is simply that the market isn't smart enough to value those things that we need as a basis - people just forget. But I am inherently skeptical of what new technology can actually provide for (the hypothetical strain of corn for instance), perhaps precisely because I *do* work with computers, and know a lot about them (you have to admit, nearly everything about computing pretty much sucks in its current infant state, short of the more mature mainframe technologies at least).
And it's the newish technolibertarianism of ESR et al that I find especially naive.
But the AC is right, I am too self-absorbed. Why else would I have time to spend reading/.?;)
There's enough good commentary now establishing that the drivers are a binary release (with a source stub for kernel version compatibility), and a fair amount of annoyance at that fact. One thing I haven't seen addressed is the question of *why* hardware companies like nVidia choose not to release source. I can see that they want to protect their R&D investment in the board's hardware and firmware capabilities, but would simply disclosing the API in the form of driver source really give away that much? I'm curious whether the decision not to do so represents fear and bad habits of closed-sourceness, or whether there a genuine justification (from their viewpoint). And if there is one, whether a method of release might be established that's better than this one (which might as well be binary-only as far as the non-x86 or non-Linux crowd is concerned).
This is a genuine question, not rhetorical; I'm not a video driver programmer so I don't know how much the source gives away about the underlying hardware, but my gut says that it can't be very much. An OpenGL driver is an OpenGL driver, surely.
I've always been amused and slightly baffled by the tendency of some geek types to endorse wacky libertarian viewpoints (a la ESR). Living as I do the ridiculously overprivileged lifestyle of a mostly-white North American middle-class male, it's long been pretty damn clear to me that the only reason I can play with all this high-tech stuff and these high-falutin' ideas is that someone else laid down the groundwork of the society to support me. I do Unix and network consulting; I don't grow food, provide childcare, clean up the environment (well, not directly), or do much else besides shuffle bits, and yet I have one of the most highly-valued skillsets around. There's no way I could exist at all without that enforced social structure, let alone with such highly specialised (and frankly useless in the real physical world) knowledge and skills.
In psychology there's a concept called Maslow's Hierarchy of Needs; basically you have to have the basics in life accounted for before you can get into the more refined and esoteric stuff. We're able to do open-source software not just because of government-mandated intellectual-property regulation (though that's a very real effect that Lessig argues for well), but also because someone else made physical life easy for us, and in a lot of cases (not all, but a lot) that someone was everyone, in the form of government trying to establish an equal basis. Technocorporate America isn't going to create the kind of society where you can work on cool code for free and still have food, clothing and shelter; they don't have the agenda, and they frankly don't have the social clue it would take.
It may seem non-germane to the IP-law argument to talk about broader social structures, but they're all part of the same viewpoint (basically, that far libertarianism is for blind kooks).
As someone else pointed out on/. a while back, you can be governed by elected officials, by corporations, or by roving street gangs (Chaos Overlords anyone?), but you're still gonna be governed.
Oh well, that's my rant. I'm sure most of the/. crowd is aching to moderate me down now.;)
(note that I'm not slagging ESR personally, just his viewpoint - Even though I don't think as much of his OSS papers as some people, I've actually long had respect for him, pretty much ever since he took over editing the Jargon file / NHD)
I guess this was sort of a meta-story, but it barely deals with PCBs past the headline (do the/. story posters bother to read these links?). The more substantive issues it mentions seem to be lead (well-documented, though no one seems to be doing anything about it) and the inability of agribusiness "Green Revolution" crops to actually nourish anyone. Is this really any surprise? You burn away all the native foods to establish an export processing zone for some monoculture cash crop, more-or-less enslave the local population to produce it (the name "Green Revolution" is more than a little Orwellian), and then note that they have nothing nutritious to eat. One might argue that for many employer multinationals this is an unexpected bonus, since a stupider population is less likely to rise up against you. Naming PCBs as the culprit just gives the false sense that there's one manageable problem we've created, with one possible solution out there waiting somewhere. There isn't.
I'd encourage anyone who's interested to read a book like Bananas, Beaches and Bases (I especially like "Carmen Miranda on my Mind") to see where all this came from. The BBC article gives you an inkling where it's going.
The saddest thing about this run-on navel-gazing review is the way the authors, in taking themselves and their "ethic" so seriously, seem to have missed the twisted, smirking humour of Gilliam's movie. Get this: De Niro's Tuttle isn't your hacker hero, he's a caricature. He repairs ducts ferchrissakes.
Yes, I understand what being a hacker is all about (my first post to alt.hackers was around 1990 under this same nick; hell, it was probably the same topic;). It's about enjoying what you do. Sometimes that involves working with technology, often it involves avoiding it instead because it usually sucks in some nasty way (think about it).
The real trick is to never, ever subject user-supplied input to a shell eval. This is one of the reasons I never, ever use the csh (among other things, it implicitly does an eval on variables like HOME, TERM, and USER when the shell is invoked).
It seems likely that the AP headline write just missed the point (headlines frequently have little to do with the real discovery in a science piece). I seem to recall that it's been generally recognised for years (decades, actually) that dinos were probably warm-blooded (like birds), but finding a four-chambered heart with one large aorta sounds fairly new to me. A three chambered heart (like modern reptiles have) makes it difficult to exert large amounts of energy in a short time - a sprint, for instance - because returning venous blood gets mixed with the fresh oxygenated arterial blood from the lungs. A heart like the one they've found (more resembling that in birds or mammals) would be better for jumping around predator style, as my cat is having fun doing at the moment. Since dinos are supposed to be more the ancestors of birds than modern reptiles, this would seem to make sense.
Of course I could be off; it's been a while since evolutionary biology. Anyone remember the author and title of a Scifi mystery story where a creature's three-chambered heart was the deciding whodunit factor?
Another benefit of the glory motivation for writing open source is that it creates concrete examples of just why you're good at what you do. It's very difficult to find good people for technical deeds that need doing (for pay), and a well-used, well-examined piece of open-source software is a major stomach-settler for someone doing some hiring. It's better than a resume because it's vetted by lots of people who would have and will have commented on it, and it's better than past private work for the obvious reason that it's available to anyone with a job that needs doing. The employment "system" we have has long been a complete joke, and anything that helps put good people in touch with good positions helps everyone.
On the other hand it can make your learning process pretty public as well. I have a copylefted IRC client here that I wrote in 1992, and I sure hope no one looking to hire me as a network C programmer has seen it.;)
For the purpose of this discussion, On and Off are merely alternate names for 1 and 0. "Digital" doesn't have to mean "bistate". If all the states of the system can be expressed as discrete values (essentially integers), the system (as a whole at least) is digital. There could be many more possible states available than two; a rotating cylinder arrangement might have four, a dice toss could have six. A buckyball toss could have thousands.;)
Since real-world stuff is continuously variable (leaving out quantum effects), digital systems generally rest on some sort of quantisation, where you draw lines between real states and create virtual "steps". Again, there don't have to be two (and typically there are many more), though some power of two is usually convenient when a computer is going to deal with the result.
Naturally if you look at a digital value as a number to some base (say ten, or sixteen), you can always re-express it as a binary value, just because any integer number can be expressed in any base. But it's just as much a digital value regardless of that math trick. The essence of what made it digital is the quantisation, not the base.
It's not clear just what aspect of these games Hasbro could have acquired rights to, or what component of intellectual property law they feel they need (or merely are able) to defend them. The dreaded "look and feel" perhaps? It seems likely that they're merely exploiting their position as a deeper pocket than most any video game company to win out-of-court through fear-induced settlement. Hopefully one of these conglomerates (with acquired copyrights to old games) will eventually come up against someone (a) large enough to defend the matter in court and (b) with a sufficient financial and moral motivation to actually do it. In this particular case, they probably only own the rights to some games in some contexts (Pac-man is the Midway licence of a Namco game, for instance; AFAIK Atari only had the rights to home computer renditions, via the 2600 licence and later Atarisoft). Tetris was probably just the arcade version licence.
In any event, this sort of chill on new versions of old games might actually be invigorating; I've enjoyed playing retro versions of cool 80's ideas, but I'd always prefer to see a new idea, and those have been few and far between for the last fifteen years or so. Since the game market isn't going away and this sort of precedent also makes any *new* ideas that much more valuable (since you can shake down later derivative versions for, um, protection money), new innovative development becomes that much more appealing and valuable.
Gawd I can't believe I used the word "innovative".
>You have to change your file mappings so QuickTime doesn't open it >How do you do that on Windows? I have the.png extension associated >with an image viewer, but IE sends all of the PNG images to quicktime.
It's not necessary on the Mac side, but on Windows you might try un-checking PNG files in the Quicktime control panel's media selection.
>This wouldn't be all that annoying, except that quicktime "forgets" to >put scrollbars on large PNG images such as screenshots.
Yeah, the Quicktime plugin really expects media items to be standalone. I think the real problem is in the plugin spec, though; it's not really clear whether scrollbars would be the right solution (the overall right thing would be for the plugin to communicate the oversize element's presence back to the browser).
That was one *extremely* subtle troll (for slashdot, anyway).
In point of fact, I did recently receive a mail attachment (happened to be a JPEG, not a GIF) which was actually a Windoze executable (by magic number) with the JPEG data appended. It took me a while to actually get around to playing with it because I alternately use LinuxPPC and the Mac OS (but this did make it clear to me that it wasn't an easily viewable JPEG). Eventually I got around to dragging it over to Virtual PC and double-clicking it (after backing up the VPC drive image, natch). The resulting Windoze behaviour was typically inconsistent - once it failed to open, another time (fresh image) it simply displayed, but on most occasions it actually ran the executable code despite the.jpg extension. This code apparently spawned a background process that made an SMTP connection (to a different IP each time I tried it, until I got bored and trashed it completely). It also extracted the JPEG data into a temp file and ran MSIE to view it, thus hiding the shenanigans from a less technical user (or just one with better things to do). I never did pin down just what the hell it was trying to do, mostly because the payload seemed pretty buggy; if not for the emulation layer and the Open Transport monitoring crap I had running on the Mac OS side, I probably wouldn't have noticed the socket connection at all. It may have been reporting on the machine (not knowing it was a Virtual PC playground), or just passing on some chain-letter style spam. It did also install a virus version of itself so that the spawned process would run on subsequent boots, but sadly that drive image file wasn't destined to live very long. B+ for effort, in case the author is reading this (grin).
All very off-topic as far as GIF patents, but amusingly on-topic for the parent article troll.;)
My "favourite" of these (meaning that which appalls me the most) is the antifeminist-masquerading-as-feminist icleo.com. Initially billed as "everything women could be interested in" and advertising such diverse subjects as fashion, astrology, and beauty tips, the current "hot topic" is whether girls [sic] who dress skimpily are asking for trouble.
Not every site needs to be a world-shaker, but a lot of the Indian ones seem convinced that only inoffensive pap ought to be electronically available.
(I'm indo-european, and some of my family probably read these same sites)
>Is that really true that the "Close Door" button on elevators was not >always there but added on? When was this added to our everyday >elevators? This really does say something about us if we feel that >those 2 seconds we just saved by closing the elevator door a hair >faster really affected our time management. LOL!:)
My condo elevators have "close" buttons, and they do work. I've also seen the kind that are mere placebos (or perhaps trigger a timer). I tend to slyly observe people in situations like elevator rides, and I think I can say that for a lot of people it's not about impatience, but about control. An elevator ride is already a somewhat uncomfortable situation for many people, and pressing the button (however impotently) feels better than waiting for the doors to close at their own machine whim. When I'm not in my little tower condo I work as a sysadmin, and I'd venture to say the observation extends to a lot of other places where people have to wait on technology, especially technology they don't (and increasingly, can't hope to) understand. For my part, I rarely ride the elevators anyway, and I feel that time passes slower and qualitatively better when I take the stairs (even though the time invested averages about the same).
Dude, it is not cool to publicly post real people's phone numbers and addresses. If someone does want to swing by and talk to him, they can email him first, allowing him to tell them his address and invite them over for tea if he wants to (presumably depending on whether they're cubic enough).
And yes, I know it wasn't hard to find in this case (or in any case where someone has an NSI domain name, unfortunately) - but that's not the point.
>As for canceling the Service or getting a free service like NetZero, FreeI, TheSimpsons, Lycos, Altavista, or >BlueLight you are forced into this horrid agreement. Not to mention what if you own a cable modem and want to >just network the box into the home lan. Can't do that.
Not that any of those free ISPs would be useful to a iOpener running BSD or Linux; they're all Windows or Mac only. Three of them are the same ISP, in fact (a rebranded 1stup.com - the ridiculous "health meter" is a dead giveaway). Something like freewwweb or worldshare might be better, since you can use them from any OS you like and they just basically suck less.
Do the new Netpliance terms make it impossible to cancel the service after one month? $99 US plus one month's service is still cheap.
It works well in every browser I've tried except for Windoze IE (it's also somewhat trickier for Mac IE, but it does work there with some minor Resedit hacking). Since I have no need to use Windows and less to use IE there, that hasn't been a problem.
Just unlock the file (or leave it user-writable for a Unix version of Netscape) and visit the sites you want persistent cookies from, then edit the file and make it read-only. After that you can "accept all" cookies and know that any new persistent ones will simply be lost when the browser silently fails to update the cookie file. Sites tracking you by cookieology will still be able to get short looks at where you go, but not an extended user profile.
One idea that I haven't yet pursued is manually devising a cookie for sites like Doubleclick that would either be dropped by their tracking mechanism, mislead them for those short windows by taking the place of the cookie they'd like to send, or perhaps even poison their database (depending on how lame their software is). That might be an interesting and useful thing to leave in one's read-only persistent cookie store.;)
Sure, anywhere at w3c.org, or occasionally at a couple of other web design oriented sites. Plus icab.de itself, of course.;)
iCab's cookie management features are great, although I sidestep the issue by using it to simply filter images from doubleclick entirely (no HTTP request is ever sent). I actually proposed the "accept only for this session" mode to the author (based on my own habit of locking the cookie file (or resource) once it has the few persistent cookies I want in it) and was pleased to see it incorporated into the very next version. In fact the whole browser is more customisable and tightly-coded than anything else I've seen. The page rendering speed could be better, though (MSIE is still faster at the moment).
You're correct that a lot of upload caps are lower than they need to be, but it's not so much that the company wants to limit you from serving files. Cable modem data networks are simply asymmetrical by design. AFAIK (and that's not much:) a large part of the problem is that all of those other passively-connected nodes in your neighbourhood spread out to form a tree that effectively concentrates noise in the upstream direction. Data coming downstream only picks up the noise on the way to your house, not everyone else's. The result is that 1Mb/s is about the maximum upstream speed even on a sparsely-populated loop where the downstream could be four or five times as high.
The cable company hereabouts actually does allow "servers", both in the TOS and in that they don't filter the ports for well-known services, but the upload cap is still there, because it's apparently fundamental to cable topologies. Companies that enforce much lower caps are probably doing so mostly much out of paranoia, because they don't know and haven't tested how much data can actually go upstream, and they don't want to find out by seeing service disrupted one fine day. You may be able to raise the cap from your end, but I would have thought you'd have to configure the cable head for that; your end is usually a slave component of the bridge. Even if you do, it may not raise the effective throughput as much as you'd think, kind of like what happens when you force a 56k modem to stay connected faster than it wants to (with commensurately higher error rates).
ADSL is similarly asymmetrical for other technical (cost-reducing) reasons as much as administrative ones (check the price against traditional [S]DSL technologies from your telco).
I think you've missed the main point of this particular product; it integrates the cable modem in the same box (and possibly administrative package) as the router and/or NAT gateway. Lots of companies make little NAT boxes with DHCP, etc. and I can set up a lovely free equivalent on an old 386 with FreeBSD in ten minutes, but it still has to connect to another box (the cable modem or DSL DTU) via an ethernet (preferably a physically separate one). This saves that step (and power connection and fan and cords) but it potentially gives cable companies a way to market NAT, which may not be good news to/. types.
Slashdot Mirror
Even proprietary software has always had a service-oriented business model. Typically you're selling the future upgrades to the user as much as the current "completed" version. Open source tends to smooth out the revision process, but it's the same idea.
;)
Even one-off programming like games is really just a service; it's just done on spec ahead of time, in the hope that people will pay you to instruct their computer how to do stuff (animate aliens decapitating people or whatever) later on. Software was never a "product", and the attempt to sell it as one oughta be pretty much over in five years. And when you think about it it was a pretty short burp - people have been selling the service of computer programming for about fifty years (Ada Lovelace was a volunteer, but the women who figured out how to program the ENIAC were paid, albeit poorly as human adding machines), and we hopefully have thousands of years left to go where people will no longer be able to try and sell it in shrinkwraps.
This really brings home some of the commentary in the recent article thread concerning the consequences of poor interface design. Just the site of that nine-tabbed-panel-hell window makes my brain want to turn off, and I'm only reading a BBC article, not trying to decide issues of law and order in the heat of the situation. It's hard enough for judges to interpret and apply the law on sober courtroom consideration; asking them to fight with a typical VB interface in order to do their (difficult) job is just plain cruel. And not just to them, but to the people whose fates they're deciding. I know that poor Windows and X interfaces impede my judgement, and I'm usually judging the computer itself. It seems to me that this sort of application really begs a custom Palm-style or Mac-ish simplified interface, and Brazilians would probably be better off if they'd waited to develop one.
The stability issues are so obvious they don't really bear further mentioning (beyond the rather redundant two times in two sentences for "GPF" in the article capsule).
You're right, of course - that's what I get for posting during daylight hours. Sorry about the misinfo, but then the vagaries of shell quoting have bitten us all.
I think what I actually had in mind when I posted that was something still different:
foo='`/bin/ls`'
eval echo "stuff $foo stuff"
Which doesn't actually even relate directly to the example. Oops.
You should really be writing the Coles Notes for the collected works of Ayn Rand. Besides the fact that you just did, I mean. ;)
I certainly wasn't arguing that computer science and IT work weren't valuable, nor that Unix work was useless or unimportant. If I truly felt that way then I'm sure I'd either be contemplating a new career, or suicide; I continue with it (and will probably move to California soon to take one or another of the jobs I see advertised) precisely because I feel I can do good with it. My argument is simply that the market isn't smart enough to value those things that we need as a basis - people just forget. But I am inherently skeptical of what new technology can actually provide for (the hypothetical strain of corn for instance), perhaps precisely because I *do* work with computers, and know a lot about them (you have to admit, nearly everything about computing pretty much sucks in its current infant state, short of the more mature mainframe technologies at least).
/.? ;)
And it's the newish technolibertarianism of ESR et al that I find especially naive.
But the AC is right, I am too self-absorbed. Why else would I have time to spend reading
There's enough good commentary now establishing that the drivers are a binary release (with a source stub for kernel version compatibility), and a fair amount of annoyance at that fact. One thing I haven't seen addressed is the question of *why* hardware companies like nVidia choose not to release source. I can see that they want to protect their R&D investment in the board's hardware and firmware capabilities, but would simply disclosing the API in the form of driver source really give away that much? I'm curious whether the decision not to do so represents fear and bad habits of closed-sourceness, or whether there a genuine justification (from their viewpoint). And if there is one, whether a method of release might be established that's better than this one (which might as well be binary-only as far as the non-x86 or non-Linux crowd is concerned).
This is a genuine question, not rhetorical; I'm not a video driver programmer so I don't know how much the source gives away about the underlying hardware, but my gut says that it can't be very much. An OpenGL driver is an OpenGL driver, surely.
I've always been amused and slightly baffled by the tendency of some geek types to endorse wacky libertarian viewpoints (a la ESR). Living as I do the ridiculously overprivileged lifestyle of a mostly-white North American middle-class male, it's long been pretty damn clear to me that the only reason I can play with all this high-tech stuff and these high-falutin' ideas is that someone else laid down the groundwork of the society to support me. I do Unix and network consulting; I don't grow food, provide childcare, clean up the environment (well, not directly), or do much else besides shuffle bits, and yet I have one of the most highly-valued skillsets around. There's no way I could exist at all without that enforced social structure, let alone with such highly specialised (and frankly useless in the real physical world) knowledge and skills.
/. a while back, you can be governed by elected officials, by corporations, or by roving street gangs (Chaos Overlords anyone?), but you're still gonna be governed.
/. crowd is aching to moderate me down now. ;)
In psychology there's a concept called Maslow's Hierarchy of Needs; basically you have to have the basics in life accounted for before you can get into the more refined and esoteric stuff. We're able to do open-source software not just because of government-mandated intellectual-property regulation (though that's a very real effect that Lessig argues for well), but also because someone else made physical life easy for us, and in a lot of cases (not all, but a lot) that someone was everyone, in the form of government trying to establish an equal basis. Technocorporate America isn't going to create the kind of society where you can work on cool code for free and still have food, clothing and shelter; they don't have the agenda, and they frankly don't have the social clue it would take.
It may seem non-germane to the IP-law argument to talk about broader social structures, but they're all part of the same viewpoint (basically, that far libertarianism is for blind kooks).
As someone else pointed out on
Oh well, that's my rant. I'm sure most of the
(note that I'm not slagging ESR personally, just his viewpoint - Even though I don't think as much of his OSS papers as some people, I've actually long had respect for him, pretty much ever since he took over editing the Jargon file / NHD)
You obviously haven't lived in many glass houses. After a while you don't even notice all those reflections from binoculars.
;)
(Arguably even better is the alternative sex life of the voyeur living next door
I'd encourage anyone who's interested to read a book like Bananas, Beaches and Bases (I especially like "Carmen Miranda on my Mind") to see where all this came from. The BBC article gives you an inkling where it's going.
The saddest thing about this run-on navel-gazing review is the way the authors, in taking themselves and their "ethic" so seriously, seem to have missed the twisted, smirking humour of Gilliam's movie. Get this: De Niro's Tuttle isn't your hacker hero, he's a caricature. He repairs ducts ferchrissakes.
;). It's about enjoying what you do. Sometimes that involves working with technology, often it involves avoiding it instead because it usually sucks in some nasty way (think about it).
Yes, I understand what being a hacker is all about (my first post to alt.hackers was around 1990 under this same nick; hell, it was probably the same topic
Actually, the AC's example will work. Try
foo="`/bin/ls`"
echo "stuff $foo stuff"
The real trick is to never, ever subject user-supplied input to a shell eval. This is one of the reasons I never, ever use the csh (among other things, it implicitly does an eval on variables like HOME, TERM, and USER when the shell is invoked).
It seems likely that the AP headline write just missed the point (headlines frequently have little to do with the real discovery in a science piece). I seem to recall that it's been generally recognised for years (decades, actually) that dinos were probably warm-blooded (like birds), but finding a four-chambered heart with one large aorta sounds fairly new to me. A three chambered heart (like modern reptiles have) makes it difficult to exert large amounts of energy in a short time - a sprint, for instance - because returning venous blood gets mixed with the fresh oxygenated arterial blood from the lungs. A heart like the one they've found (more resembling that in birds or mammals) would be better for jumping around predator style, as my cat is having fun doing at the moment. Since dinos are supposed to be more the ancestors of birds than modern reptiles, this would seem to make sense.
Of course I could be off; it's been a while since evolutionary biology. Anyone remember the author and title of a Scifi mystery story where a creature's three-chambered heart was the deciding whodunit factor?
Another benefit of the glory motivation for writing open source is that it creates concrete examples of just why you're good at what you do. It's very difficult to find good people for technical deeds that need doing (for pay), and a well-used, well-examined piece of open-source software is a major stomach-settler for someone doing some hiring. It's better than a resume because it's vetted by lots of people who would have and will have commented on it, and it's better than past private work for the obvious reason that it's available to anyone with a job that needs doing. The employment "system" we have has long been a complete joke, and anything that helps put good people in touch with good positions helps everyone.
;)
On the other hand it can make your learning process pretty public as well. I have a copylefted IRC client here that I wrote in 1992, and I sure hope no one looking to hire me as a network C programmer has seen it.
For the purpose of this discussion, On and Off are merely alternate names for 1 and 0. "Digital" doesn't have to mean "bistate". If all the states of the system can be expressed as discrete values (essentially integers), the system (as a whole at least) is digital. There could be many more possible states available than two; a rotating cylinder arrangement might have four, a dice toss could have six. A buckyball toss could have thousands. ;)
Since real-world stuff is continuously variable (leaving out quantum effects), digital systems generally rest on some sort of quantisation, where you draw lines between real states and create virtual "steps". Again, there don't have to be two (and typically there are many more), though some power of two is usually convenient when a computer is going to deal with the result.
Naturally if you look at a digital value as a number to some base (say ten, or sixteen), you can always re-express it as a binary value, just because any integer number can be expressed in any base. But it's just as much a digital value regardless of that math trick. The essence of what made it digital is the quantisation, not the base.
It seems likely that they're merely exploiting their position as a deeper pocket than most any video game company to win out-of-court through fear-induced settlement. Hopefully one of these conglomerates (with acquired copyrights to old games) will eventually come up against someone (a) large enough to defend the matter in court and (b) with a sufficient financial and moral motivation to actually do it.
In this particular case, they probably only own the rights to some games in some contexts (Pac-man is the Midway licence of a Namco game, for instance; AFAIK Atari only had the rights to home computer renditions, via the 2600 licence and later Atarisoft). Tetris was probably just the arcade version licence.
In any event, this sort of chill on new versions of old games might actually be invigorating; I've enjoyed playing retro versions of cool 80's ideas, but I'd always prefer to see a new idea, and those have been few and far between for the last fifteen years or so. Since the game market isn't going away and this sort of precedent also makes any *new* ideas that much more valuable (since you can shake down later derivative versions for, um, protection money), new innovative development becomes that much more appealing and valuable.
Gawd I can't believe I used the word "innovative".
>You have to change your file mappings so QuickTime doesn't open it .png extension associated
>How do you do that on Windows? I have the
>with an image viewer, but IE sends all of the PNG images to quicktime.
It's not necessary on the Mac side, but on Windows you might try un-checking PNG files in the Quicktime control panel's media selection.
>This wouldn't be all that annoying, except that quicktime "forgets" to
>put scrollbars on large PNG images such as screenshots.
Yeah, the Quicktime plugin really expects media items to be standalone. I think the real problem is in the plugin spec, though; it's not really clear whether scrollbars would be the right solution (the overall right thing would be for the plugin to communicate the oversize element's presence back to the browser).
That was one *extremely* subtle troll (for slashdot, anyway).
.jpg extension. This code apparently spawned a background process that made an SMTP connection (to a different IP each time I tried it, until I got bored and trashed it completely). It also extracted the JPEG data into a temp file and ran MSIE to view it, thus hiding the shenanigans from a less technical user (or just one with better things to do).
;)
In point of fact, I did recently receive a mail attachment (happened to be a JPEG, not a GIF) which was actually a Windoze executable (by magic number) with the JPEG data appended. It took me a while to actually get around to playing with it because I alternately use LinuxPPC and the Mac OS (but this did make it clear to me that it wasn't an easily viewable JPEG).
Eventually I got around to dragging it over to Virtual PC and double-clicking it (after backing up the VPC drive image, natch). The resulting Windoze behaviour was typically inconsistent - once it failed to open, another time (fresh image) it simply displayed, but on most occasions it actually ran the executable code despite the
I never did pin down just what the hell it was trying to do, mostly because the payload seemed pretty buggy; if not for the emulation layer and the Open Transport monitoring crap I had running on the Mac OS side, I probably wouldn't have noticed the socket connection at all. It may have been reporting on the machine (not knowing it was a Virtual PC playground), or just passing on some chain-letter style spam. It did also install a virus version of itself so that the spawned process would run on subsequent boots, but sadly that drive image file wasn't destined to live very long. B+ for effort, in case the author is reading this (grin).
All very off-topic as far as GIF patents, but amusingly on-topic for the parent article troll.
My "favourite" of these (meaning that which appalls me the most) is the antifeminist-masquerading-as-feminist icleo.com. Initially billed as "everything women could be interested in" and advertising such diverse subjects as fashion, astrology, and beauty tips, the current "hot topic" is whether girls [sic] who dress skimpily are asking for trouble.
Not every site needs to be a world-shaker, but a lot of the Indian ones seem convinced that only inoffensive pap ought to be electronically available.
(I'm indo-european, and some of my family probably read these same sites)
>Is that really true that the "Close Door" button on elevators was not :)
>always there but added on? When was this added to our everyday
>elevators? This really does say something about us if we feel that
>those 2 seconds we just saved by closing the elevator door a hair
>faster really affected our time management. LOL!
My condo elevators have "close" buttons, and they do work. I've also seen the kind that are mere placebos (or perhaps trigger a timer).
I tend to slyly observe people in situations like elevator rides, and I think I can say that for a lot of people it's not about
impatience, but about control. An elevator ride is already a somewhat uncomfortable situation for many people, and pressing the button (however impotently) feels better than waiting for
the doors to close at their own machine whim. When I'm not in my little tower condo I work as a sysadmin, and I'd venture to say the observation extends to a lot of other places where people have to wait on technology, especially technology they don't (and increasingly, can't hope to) understand. For my part, I rarely ride the elevators anyway, and I feel that time passes slower and qualitatively better when I take the stairs (even though the time invested averages about the same).
Dude, it is not cool to publicly post real people's phone numbers and addresses. If someone does want to swing by and talk to him, they can email him first, allowing him to tell them his address and invite them over for tea if he wants to (presumably depending on whether they're cubic enough).
And yes, I know it wasn't hard to find in this case (or in any case where someone has an NSI domain name, unfortunately) - but that's not the point.
>BlueLight you are forced into this horrid agreement. Not to mention what if you own a cable modem and want to
>just network the box into the home lan. Can't do that.
Not that any of those free ISPs would be useful to a iOpener running BSD or Linux; they're all Windows or Mac only. Three of them are the same ISP, in fact (a rebranded 1stup.com - the ridiculous "health meter" is a dead giveaway). Something like freewwweb or worldshare might be better, since you can use them from any OS you like and they just basically suck less.
Do the new Netpliance terms make it impossible to cancel the service after one month? $99 US plus one month's service is still cheap.
It works well in every browser I've tried except for Windoze IE (it's also somewhat trickier for Mac IE, but it does work there with some minor Resedit hacking). Since I have no need to use Windows and less to use IE there, that hasn't been a problem.
;)
Just unlock the file (or leave it user-writable for a Unix version of Netscape) and visit the sites you want persistent cookies from, then edit the file and make it read-only. After that you can "accept all" cookies and know that any new persistent ones will simply be lost when the browser silently fails to update the cookie file. Sites tracking you by cookieology will still be able to get short looks at where you go, but not an extended user profile.
One idea that I haven't yet pursued is manually devising a cookie for sites like Doubleclick that would either be dropped by their tracking mechanism, mislead them for those short windows by taking the place of the cookie they'd like to send, or perhaps even poison their database (depending on how lame their software is). That might be an interesting and useful thing to leave in one's read-only persistent cookie store.
>BTW, have you ever, ever ever seen iCab smile?
;)
Sure, anywhere at w3c.org, or occasionally at a couple of other web design oriented sites. Plus icab.de itself, of course.
iCab's cookie management features are great, although I sidestep the issue by using it to simply filter images from doubleclick entirely (no HTTP request is ever sent). I actually proposed the "accept only for this session" mode to the author (based on my own habit of locking the cookie file (or resource) once it has the few persistent cookies I want in it) and was pleased to see it incorporated into the very next version.
In fact the whole browser is more customisable and tightly-coded than anything else I've seen. The page rendering speed could be better, though (MSIE is still faster at the moment).
You're correct that a lot of upload caps are lower than they need to be, but it's not so much that the company wants to limit you from serving files. Cable modem data networks are simply asymmetrical by design. AFAIK (and that's not much :) a large part of the problem is that all of those other passively-connected nodes in your neighbourhood spread out to form a tree that effectively concentrates noise in the upstream direction. Data coming downstream only picks up the noise on the way to your house, not everyone else's. The result is that 1Mb/s is about the maximum upstream speed even on a sparsely-populated loop where the downstream could be four or five times as high.
The cable company hereabouts actually does allow "servers", both in the TOS and in that they don't filter the ports for well-known services, but the upload cap is still there, because it's apparently fundamental to cable topologies. Companies that enforce much lower caps are probably doing so mostly much out of paranoia, because they don't know and haven't tested how much data can actually go upstream, and they don't want to find out by seeing service disrupted one fine day. You may be able to raise the cap from your end, but I would have thought you'd have to configure the cable head for that; your end is usually a slave component of the bridge. Even if you do, it may not raise the effective throughput as much as you'd think, kind of like what happens when you force a 56k modem to stay connected faster than it wants to (with commensurately higher error rates).
ADSL is similarly asymmetrical for other technical (cost-reducing) reasons as much as administrative ones (check the price against traditional [S]DSL technologies from your telco).
I think you've missed the main point of this particular product; it integrates the cable modem in the same box (and possibly administrative package) as the router and/or NAT gateway. Lots of companies make little NAT boxes with DHCP, etc. and I can set up a lovely free equivalent on an old 386 with FreeBSD in ten minutes, but it still has to connect to another box (the cable modem or DSL DTU) via an ethernet (preferably a physically separate one). This saves that step (and power connection and fan and cords) but it potentially gives cable companies a way to market NAT, which may not be good news to /. types.