Hey, if you thing copyrighting "nothing" is disturbing, google for the terms "/bin/true copyright" and read a few of the matches.
Yes, AT&T really did claim copyright ownership of either "nothing" or a blank line, depending on how you read their copyright notice.
Back in the 80's, when this appeared, I and several others posted the code for the Sys/V/bin/true script on several newsgroups, and openly challenged AT&T to sue us for copyright infringement. For some reason, we never heard from them.
Today, there's a good chance that we would get sued. And if you have any programs that contain blank lines, note that the current heir to AT&T's unix code (SCO) could well charge you with infringement.
The discussions here and elsewhere have included the conclusion that, when SCO made the claim that there were millions of lines of infringing code in linux, the claim could be substantiated if you counted the lines that consist solely of things like "}", "/*" and "*/". So it appears that SCO is seriously making a claim that they own the copyright on such code fragments, and is right now prosecuting infringers.
It doesn't get much more absurd than this.
(Wait, maybe I shouldn't make such a claim. Someone is bound to prove me wrong.;-)
As others have pointed out, google already knows this. It even knows that, in THHGttU, the question turns out to be "What is six times nine?"
But even better, if you ask google about "six times nine" and "42", you'll find pointers to the many discussions of why it would be that, when you don't specify the base, Deep Thought II quite naturally assumed that you wanted to use base 13.
One of my favorites is in the page that google just now turned up as the first match. The author (Michael Thorpe) observed that "By using 42 as the Hubble constant, you can calculate the age of the universe to be approximately 23 billion years, which is consistent with experimental data." This isn't the conventional age of the universe right now, of course. But if you set the Hubble Constant to 42 in base 13, you do get a result closer to the currently-accepted age of 12 to 15 billion years (depending on which cosmologists you believe in). Of course, this may have changed by the time you read this.
Anyway, it seems that google is already well ahead of the schedule posted here.
While libraries have been one way of preserving such information, they do tend to have serious limitations on what they can preserve. They are easy targets for the book burners and people chanting "national security".
Here's a suggestion: When you find interesting things like this, file a copy away in your own computer. When your disk gets full, archive them to CDs or DVDs. Then, when they disappear from the Net, you can restore them yourself. Make sure you keep some notes on when and where you got each document; that'll be useful to people in the future.
If we all start doing this, we can make it very difficult for those in power to suppress the information in the future.
The main remaining problem is the secrecy inside organizations, both government and corporate. The world would be a safer place for all of us if we could open the books on all such organizations. So we should be pushing for laws that require that the inner workings of all organizations be public and available on the internet in public formats. If we teach people to watch and copy interesting docs to their own archives, we will have a powerful tool to keep them honest.
Of course, there is also a powerful move afoot to use the copyright laws as a pretext to cover up organizational malfeasance by suppressing evidence. We need clauses in the laws that make it clear that copyright does not protect you in such situations.
1. Someone posts a blank [ insert fav music editor of choice ] file
Better be careful here. Remember last year, when Mike Batt included a short blank track on a CD, and John Cage's estate sued him for infringing on their copyright of his famous 4'33" work...
If you track down all the parts of the story, you'll find that in the end, Batt had to pay.
Now, you might say that this was because he mentioned Cage in the liner notes. But in fact this is irrelevant in copyright cases. If I "steal" something of yours and publish it without your permission as my own performance, it's an infringement whether or not I acknowledge it as your work.
(Just doing my part to add to the level of surrealism surrounding this whole "IP" issue.;-)
[T]he ideas behind free software are incomprehensible to non-programmers,...
Oh, nonsense. They're comprehensible to the average five-year-old.
When you were that age, how often did you hear adults say "Nice people share their toys with each other"?
Open and/or free software is exactly the same idea.
And software (or any kind of information) has one nice feature that physical toys don't have: With most physical toys, if I'm playing with it, you can't. But with information, you can share it with as many others as want it, and you still have it to play with yourself.
Most five-year-olds can understand this, too. Talk to a few of them, and see how they enjoy word play. This is a sort of play that is only fun if you're sharing it with others.
This isn't just a joke. My wife, who works in a Windows-only medical org (yeah; I know;-) came home a few days ago asking if I'd heard that RedHat was going out of business. It seems that the people she works with had spent a lot of time discussing this, and were glad that they'd stuck with microsoft so they wouldn't have such a disaster hitting them.
She uses a Mac at home. Lately, she's been taking it to work a lot. She seems to enjoy the looks of greed and envy on the faces of her coworkers. She put together a slide show of photos from our recent vacation, using iPhoto, and her other digital-camera toting friends are all suitably impressed by how slick it looks.
But they now feel sorry for those poor folks who were suckered into buying RedHat. They'll probably be puzzled a year or two from now when they hear that RH is still in business. But they won't worry about it too much, since it'll be below their radar.
OTOH, maybe their management will see through this particular bit of FUD and will suddenly announce a conversion program...
Well, I have my doubts about the accuracy of this claim.
In a recent election hereabouts, some time after the election, there was a news report that 30,000 uncounted votes had been discovered in one precinct. They counted them, of course, and claimed that this didn't effect the election results.
In the 2000 election, there was a similar report from Florida, but the "misplaced" boxes of votes contained around 100,000 ballots. After they were counted, there was a similar claim that it hadn't changed the outcome of the election. Right.
My reaction to all the stories like this is to wonder how many other votes were similarly "misplaced" and not "discovered". And how do I know that my vote wasn't similarly "misplaced". (I live in a university neighborhood that often votes differently from the city as a whole, so there is good reason to be suspicious.)
I've long wondered if, when we hear that only 50% of the voters actually voted, what really happened was that 90% voted, but the votes of 40% were "misplaced" and not counted.
With proprietary electronic voting systems, not auditable by the public, it seems that the voting turnout is highly likely to decrease similarly...
[H]ow can you verify that your vote was counted correctly?
While that's obviously desirable, it is nowhere near sufficient. Note the recent discussion of the Diebold-memo fiasco, in which we read the explanation for the funny declaration by the media that Bush had won Florida by a good margin, and then this was retracted the next day. It seems that in one precinct, the Diebold voting equipment did report a correct vote total, but it also sent in a second report that Gore had received -16022 votes. The tallying code didn't do the obvious sanity check here, and it took a while for someone to notice the "interesting" total.
In this case, it seems that the correct total was sent in, so auditing to verify that the votes were counted correctly wouldn't have spotted the problem. They were counted correctly.
There are many ways to tweak an election, other than not counting someone's vote correctly. You have to determine that dead people haven't voted. You should also make sure that the equipment isn't just making up voters. (There have been many cases of more votes being cast than there were registered voters in the district.) You need to flag negative votes. You should verify that the numbers in the total boxes are actually the correct totals of what was supposed to be added. You need to understand the concept of an "integer overflow". And so on.
There was also the quote from Diebold's CEO to a group of Ohio Republicans, to the effect that he promised them that in the next election, he would deliver Ohio to Bush. Coming from an officer of the company that makes the voting equipment, this is not an idle boast. But the right answer isn't to go after him; it's to make sure that nobody can make good on such promises. And the Republicans are far from the only political group that is looking for this sort of support.
Slashdot posting gives some similar stimuli. By posting an excellent message, the author receives moderation and more people start discussing the idea.
Actually, if you look at/. "discussions", they are mostly quite shallow trees. Lots of comments on the article; a few of those comments get a long list of (often not very relevant) replies; a very few get replies to those replies. Not much depth there. Very few cases where the discussion really leads to any conclusions.
OTOH, often the first two levels contain actual information or links to information. And this is often more informative than the original article.
(I wouldn't expect that this comment will lead to many, if any, actual conclusions of any sort...;-)
... system in that future. It is almost a done deal - when major corporations imagine Linux as central to the future, Linux becomes central to the future.
Well, yes and no. Linux (i.e., the kernel) is really just an instance of POSIX, and code that runs on linux will usually run on other POSIX systems with little more than a recompile. This now includes pretty much all the available unix-like systems, even those that call themselves "BSD".
It's true that there are inevitably problems with drivers, but that's more of a hardware problems than an OS problem. A driver compiled for a different processor may need tweaking to make it work right. But the driver's interface to a new unix-like system probably needs little tweaking.
Building on linux really just makes your code highly portable, so you can easily treat the OS as a commodity. Put out bids, and use the hardware/OS combination that costs least (among POSIX systems that satisfy your minimum speed requirements).
Linux does tend to win such cost/benefit calculations, since the OS itself has a rather low cost. But this may not be true 5 or 10 years from now.
Part of the cost calculation is something often missed in simplistic analyses: Porting to a new platform does always take a fair amount of work, even when everything goes well. You need a very thorough test suite, and you have to hunt down and fix a flock of subtle bugs due to different compilers, runtime libraries, etc. This is even true when going to the next release of your current platform. Fixing the "small" bugs is especially tricky when the trail leads down into the OS. Vendors are often not very cooperative when you start asking questions about the internals, which they consider proprietary. With linux, all the source is freely available, so it's easy to trace bugs down into the OS. Well, it's easy for expert C hackers.
So porting to linux is usually faster and easier than to commercial systems. But linux has no advantage here over the *BSD systems. It also looks like OSX is becoming (nearly) as open as linux and *BSD.
In any case, I wouldn't bet on IBM being loyal to linux over the long term. More likely, they favor it because of the advantages of a system with freely-available, license-free source. Any other system with these properties (and which has a POSIX library) will be a very real possibility in the future.
To be a PBX replacement, you should be able to interface to telephones. E.g. 100 or 200 of them.
How about just using bluetooth-enabled phones?
This sort of thing isn't all that new a concept. Back in the mid 1980's, I saw and used internal versions of phones via wireless modems at some Motorola sites where I was working. They had both voice and IP traffic going over the wireless system, and various parts of the "PBX" used TCP to transfer voice traffic. It was all very experimental, but much of it worked ok. And this was before RTP was developed to fix the problems with voice over TCP.
One of the fun parts was the electronic id tags you could get, similar in function to RFID, which would let the system locate you. If you carried the tag, phone calls would be routed to the phone that was physically closest to you. So if a nearby phone rang, you would pick it up, and it would be for you or someone else very close. Of course, this did lead to all the obvious discussions about the phone system tracking your every move. Many people just left their id sitting under their desk phone.
The curious thing is that VoIP has taken so long to catch on with "The Market". Here in North America, the phone companies have been reporting that traffic over long lines has been pretty much all converted to IP by now, and the idea is slowly making its way upward through the entire phone system. But they've resisted offering it at the retail level for a very obvious reason: It would kill all that nice income from long-distance charges. With IP, "long distance" isn't a very meaningful concept.
Google has around US$700-million in annual revenues, and it makes about US$100-million a year in profits.
An interesting question is: If google can be used by anyone for free, where does their income come from?
As far as I've read, one of the major sources is that they license their software to corporations, governments, and other organizations that want to get access to their own internal computer systems. This makes sense. Why develop it yourself, when for a much lower price you can get a few google people to come in and get it all running in a few days?
But this in turn brings up a different question: If Microsoft wants its own search engine, why would it want to buy google? Why not just lease the software, as others do? Use it as a starting point, and start building your own plugins to add new capabilities. They'd have an "embraced, extended" google, with all of google's capabilities plus a bunch of new goodies.
The obvious answer: Microsoft isn't just interested in building a new, more powerful search site. They also want to shut down its biggest competitor.
Presumably, their real plan now is to wait for google to go public. Then they can buy up the stock and take over that way. It wouldn't be the first time that a firm has used this approach to eliminate the competition.
Maybe we should be telling google to not go public. Really; what's in it for them, other than the grief of seeing a hostile takeover from the inside?
Yeah, but he's still wrong. Users shouldn't use Windows at home; they should use OSX.
A couple of months back, after noticing the growing rate of obscenities coming from my wife while she was using her Windows box (for reasons of compatibility with machines at work, of course), I talked her into trying a Mac Powerbook. Despite her years of experience with Windows, she adapted quickly to the Mac, with no obscenities at all. In only an hour or so, she was saying how much she liked it.
Now, a couple of months later, she will tell you of a few things that she things work better in Windows. But she shudders at the thought of going back, and mutters obscenities when discussing the machines that she uses at work.
As a longtime unix/linux geek, I don't think that the Mac is an improvement. The various X window managers are nearly all better than the Mac (and much better than the Windows GUI). But the Mac is pretty good, and is easy to learn.
One thing that the Mac has going for it that linux can't (yet) duplicate to my knowledge: Since the Mac moved into our house, our use of the TV sets and the CD player has dropped to nearly zero. We have a Netflix account, and their DVDs work just fine in the Mac, anywhere in the house (or back yard;-). CDs work fine with iTunes (with earphones or a good speaker plugged in). Many TV shows are also available via the internet, and with a cable modem and airport, we can watch them wherever we are, not just where the TV is. Instead of the weather channel, we use weather.com, from anywhere. And so on.
We're seriously looking into disposing of our TVs and just using the wireless Mac. The only real question here is whether we can get internet service via cable without paying for the bundled TV channels, or can we find DSL that's fast enough?
I've been wondering if there's some way we could introduce the Mac and OSX to Red Hat's CEO. Give him a bit of guidance in using it. Maybe he'd turn into a convert, too. Maybe he'd start a Red Hat project to provide a wireless linux-based laptop with lots of good open-source audio and video stuff preconfigured. They could have a real winner there.
Actually, most artists are giving you their work for free. All that money you pay goes to the distributors; none of it reaches the artists.
Recording industry contracts pretty much guarantee that the artist gets nothing (and in fact usually ends up in debt) until at least 1.5 million CDs are sold. For new groups, the crossover point is often much higher, and is hardly ever reached.
But this is hardly news in this forum. Anyone reading/. for more than a month or so should have seen the data by now.
Well, yeah, but people have been saying for years that most CDs have only one or sometimes two songs that most purchasers want. So in most caseds, a single-tune download has literally replaced a single CD sale.
Actually, of course, it's just normal American corporate management practices that I'm talking about here. I keep getting the feeling that it's not outsourcing to cheaper parts of the world that we should be worried about. If any other part of the world ever invents a rational scheme for organizing companies, they'll wipe out our economy overnight.
Fortunately, there seems little danger of this threat materializing.
The funniest case was a few years back, when the project's management decreed the Netscape server as the standard. We tried several times. But the same thing always killed the effort: This server can be configured only through its web interface. Invariably, we would make some config mistake that turned the server into a zombie. At that point, there was no way to correct the problem because we couldn't change the configuration any more. We'd wipe the server's directories, reinstall -- and it would happen again. Sometimes we'd get it running for a few days, but every config change carried with it the possibility that we'd have to wipe the server and start over.
You'd think that people would understand why you can't trust a web server to handle changing its own config files. But the managers couldn't be convinced that there was a fundamental problem here. And we never found a way to get at those files with a plain editor. They just didn't make sense, and weren't documented anywhere that we could find.
I've long argued that one of apache's real strengths is its plain-text config file (with lots of good comments in the text). The commercial guys don't seem to be able to figure out why this is a good idea.
How can this be true when many people run Apache on Windows?
Funny thing about this: On many occasions, I've found myself looking at a group trying to install IIS (or the Netscape server or some other commercial server), and getting more and more frustrated over the problems getting it to work.
So, while they're fighting with it, I sit down at an idle machine, point the browser at apache.org, download the latest apache for that platform, and ask them questions while I twiddle the configuration. Within 10 to 20 minutes, depending on how much configging is needed, I fire up the server, and it runs the first try. I invite them to check it out from the other test machines, and they find that it's working. We copy a few web pages to that machine, and they work
The result in almost all cases is that they decide to go with apache "for a while". It's just an interim measure, you understand, until they can get the real web server running. But meanwhile, they have a web server that they can put online. The web developers aren't sitting around idle; they're building the web site.
In the ensuing months or years, I occasionally prod them with "You know, we really should try to get the officially-mandated web server running." The response is usually to put it off until they can get through the huge pile of stuff that they need to put online.
In a few cases, management has gotten upset, and created a team to get the officially-mandated server running. This often succeeds after a few weeks. Then they put that server online, and it's a real disaster. It crashes repeatedly, produces a flood of complaints from baffled customers along the lines of "How the @#&$^%*& do I order things from you now? Your online ordering pages are broken."
After management notices the loss of income from IIS or whatever, they grudgingly agree to go back to apache "until the problems can be worked out."
Not materially different at all; it's using HTTP to do what thousands of us have been doing with NFL and FTP for several decades.
For a good parallel, imagine that a new sort of road surface material were developed and started to be used on some highways. If this patent holds up, then we could expect to see patents for methods of driving on this new material. Yes, people have been driving on roads for a long time. But they haven't been driving on this new material, so this would be patentable.
The USPTO is getting to be a real frustration for satirists. You can't think of a bizarre IP-related idea without reading that they just outdid your idea with something that's even more bizarre.
I do note that the anti-caffeine item says "avoid", which isn't quite as absolute as the way the other items are phrased. And it comes with the explanation "Excessive caffeine has been shown to cause low blood sugar, which can make you crave sugar." Note that word "excessive".
Also, the very next item suggests 8 glasses of water per day.
A reasonable interpretation is that a low level of caffeine, say in the form of unsweetened coffee or tea, for one or two of those glasses of water would not be a material violation of the rules. But you do have to be the sort that can resist any sugar cravings that result. And this might not even happen with a small amount of caffeine, if it is consumed along with the meats, eggs and cheese, which do contain small quantities of carbs.
OTOH, if you're the sort who can't resist your 8 cans of Jolt each day, maybe you'd be better off just going cold turkey on the caffeine.
(I wonder if cocaine would be a good substitute? That's supposed to be a diet drug itself, right?;-)
The day a email tax is born will be the very same day when email will die!
Nah; it'll just morph into something not legally recognized as email.
In fact, just this sort of thing has already happened many times, for several reasons.
Thus, my wife and I have a joint account with a local ISP that also supplies things like email accounts, unix shell accounts, etc. We had been using our mail readers to access this account from home. But one day it stopped working. A few hours of testing verified that our (cable) ISP had started blocking outgoing connections to port 199 except on its own machine. They want all our email stored on their machines, not on some competitor's machines.
It didn't take long before the other ISP announced a solution: They simply set up a web interface to their email accounts. Now our home ISP can't block the access, because they only see a TCP connection to port 80. The url starts with "https://", so the email is encrypted, and they can't tell that it's email. Even if they suspected this (which I'm sure they do), there's not much they can do. They can't block all https pages, because online commerce depends on them.
It's much like the shutdown of Napster. The only real result was to replace it with a whole flock of other P2P services, no two alike. Shutting them down will just annoy your customers and encourage replacements that your blocking/taxing code can't recognize.
Way back when the US government passed the 55 mph national speed limit, there were a number of people who pointed out the same sort of problem. The usual reason was for saving lives. But a straightforward cost-benefit analysis, using pulicly-available actuarial and highway data, showed that when you added up the extra travel time from the lower speed limit, it would take roughly three lifetimes of travel to match a single traffic death. Since the extra travel time is "dead" time that can be considered stolen from your life, the same reasoning could be applied.
Not that this convinced much of anyone. But it was a fun bit of reasoning to read about.
We do have a problem that not just spammers, but pretty much anyone in a position of authority generally considers wasting your and my time to be not a problem. The only way I've ever found to fight this is when I'm working on an hourly basis, and I can tell someone "This discussion we have just had, which has accomplished little other than prevent me from doing the job I'm being paid to do, has cost your company $N. Should we continue the discussion, or should I get back to work?"
Unfortunately, even this isn't always effective. But sometimes it does result in me not getting invited to meetings.
Slashdot Mirror
Hey, if you thing copyrighting "nothing" is disturbing, google for the terms "/bin/true copyright" and read a few of the matches.
/bin/true script on several newsgroups, and openly challenged AT&T to sue us for copyright infringement. For some reason, we never heard from them.
;-)
Yes, AT&T really did claim copyright ownership of either "nothing" or a blank line, depending on how you read their copyright notice.
Back in the 80's, when this appeared, I and several others posted the code for the Sys/V
Today, there's a good chance that we would get sued. And if you have any programs that contain blank lines, note that the current heir to AT&T's unix code (SCO) could well charge you with infringement.
The discussions here and elsewhere have included the conclusion that, when SCO made the claim that there were millions of lines of infringing code in linux, the claim could be substantiated if you counted the lines that consist solely of things like "}", "/*" and "*/". So it appears that SCO is seriously making a claim that they own the copyright on such code fragments, and is right now prosecuting infringers.
It doesn't get much more absurd than this.
(Wait, maybe I shouldn't make such a claim. Someone is bound to prove me wrong.
As others have pointed out, google already knows this. It even knows that, in THHGttU, the question turns out to be "What is six times nine?"
But even better, if you ask google about "six times nine" and "42", you'll find pointers to the many discussions of why it would be that, when you don't specify the base, Deep Thought II quite naturally assumed that you wanted to use base 13.
One of my favorites is in the page that google just now turned up as the first match. The author (Michael Thorpe) observed that "By using 42 as the Hubble constant, you can calculate the age of the universe to be approximately 23 billion years, which is consistent with experimental data." This isn't the conventional age of the universe right now, of course. But if you set the Hubble Constant to 42 in base 13, you do get a result closer to the currently-accepted age of 12 to 15 billion years (depending on which cosmologists you believe in). Of course, this may have changed by the time you read this.
Anyway, it seems that google is already well ahead of the schedule posted here.
While libraries have been one way of preserving such information, they do tend to have serious limitations on what they can preserve. They are easy targets for the book burners and people chanting "national security".
Here's a suggestion: When you find interesting things like this, file a copy away in your own computer. When your disk gets full, archive them to CDs or DVDs. Then, when they disappear from the Net, you can restore them yourself. Make sure you keep some notes on when and where you got each document; that'll be useful to people in the future.
If we all start doing this, we can make it very difficult for those in power to suppress the information in the future.
The main remaining problem is the secrecy inside organizations, both government and corporate. The world would be a safer place for all of us if we could open the books on all such organizations. So we should be pushing for laws that require that the inner workings of all organizations be public and available on the internet in public formats. If we teach people to watch and copy interesting docs to their own archives, we will have a powerful tool to keep them honest.
Of course, there is also a powerful move afoot to use the copyright laws as a pretext to cover up organizational malfeasance by suppressing evidence. We need clauses in the laws that make it clear that copyright does not protect you in such situations.
1. Someone posts a blank [ insert fav music editor of choice ] file
...
;-)
Better be careful here. Remember last year, when Mike Batt included a short blank track on a CD, and John Cage's estate sued him for infringing on their copyright of his famous 4'33" work
If you track down all the parts of the story, you'll find that in the end, Batt had to pay.
Now, you might say that this was because he mentioned Cage in the liner notes. But in fact this is irrelevant in copyright cases. If I "steal" something of yours and publish it without your permission as my own performance, it's an infringement whether or not I acknowledge it as your work.
(Just doing my part to add to the level of surrealism surrounding this whole "IP" issue.
[T]he ideas behind free software are incomprehensible to non-programmers, ...
Oh, nonsense. They're comprehensible to the average five-year-old.
When you were that age, how often did you hear adults say "Nice people share their toys with each other"?
Open and/or free software is exactly the same idea.
And software (or any kind of information) has one nice feature that physical toys don't have: With most physical toys, if I'm playing with it, you can't. But with information, you can share it with as many others as want it, and you still have it to play with yourself.
Most five-year-olds can understand this, too. Talk to a few of them, and see how they enjoy word play. This is a sort of play that is only fun if you're sharing it with others.
People are making it seem like Redhat is dead ...
;-) came home a few days ago asking if I'd heard that RedHat was going out of business. It seems that the people she works with had spent a lot of time discussing this, and were glad that they'd stuck with microsoft so they wouldn't have such a disaster hitting them.
...
This isn't just a joke. My wife, who works in a Windows-only medical org (yeah; I know
She uses a Mac at home. Lately, she's been taking it to work a lot. She seems to enjoy the looks of greed and envy on the faces of her coworkers. She put together a slide show of photos from our recent vacation, using iPhoto, and her other digital-camera toting friends are all suitably impressed by how slick it looks.
But they now feel sorry for those poor folks who were suckered into buying RedHat. They'll probably be puzzled a year or two from now when they hear that RH is still in business. But they won't worry about it too much, since it'll be below their radar.
OTOH, maybe their management will see through this particular bit of FUD and will suddenly announce a conversion program
US voter turnout is also right around 50%.
...
Well, I have my doubts about the accuracy of this claim.
In a recent election hereabouts, some time after the election, there was a news report that 30,000 uncounted votes had been discovered in one precinct. They counted them, of course, and claimed that this didn't effect the election results.
In the 2000 election, there was a similar report from Florida, but the "misplaced" boxes of votes contained around 100,000 ballots. After they were counted, there was a similar claim that it hadn't changed the outcome of the election. Right.
My reaction to all the stories like this is to wonder how many other votes were similarly "misplaced" and not "discovered". And how do I know that my vote wasn't similarly "misplaced". (I live in a university neighborhood that often votes differently from the city as a whole, so there is good reason to be suspicious.)
I've long wondered if, when we hear that only 50% of the voters actually voted, what really happened was that 90% voted, but the votes of 40% were "misplaced" and not counted.
With proprietary electronic voting systems, not auditable by the public, it seems that the voting turnout is highly likely to decrease similarly
[H]ow can you verify that your vote was counted correctly?
While that's obviously desirable, it is nowhere near sufficient. Note the recent discussion of the Diebold-memo fiasco, in which we read the explanation for the funny declaration by the media that Bush had won Florida by a good margin, and then this was retracted the next day. It seems that in one precinct, the Diebold voting equipment did report a correct vote total, but it also sent in a second report that Gore had received -16022 votes. The tallying code didn't do the obvious sanity check here, and it took a while for someone to notice the "interesting" total.
In this case, it seems that the correct total was sent in, so auditing to verify that the votes were counted correctly wouldn't have spotted the problem. They were counted correctly.
There are many ways to tweak an election, other than not counting someone's vote correctly. You have to determine that dead people haven't voted. You should also make sure that the equipment isn't just making up voters. (There have been many cases of more votes being cast than there were registered voters in the district.) You need to flag negative votes. You should verify that the numbers in the total boxes are actually the correct totals of what was supposed to be added. You need to understand the concept of an "integer overflow". And so on.
There was also the quote from Diebold's CEO to a group of Ohio Republicans, to the effect that he promised them that in the next election, he would deliver Ohio to Bush. Coming from an officer of the company that makes the voting equipment, this is not an idle boast. But the right answer isn't to go after him; it's to make sure that nobody can make good on such promises. And the Republicans are far from the only political group that is looking for this sort of support.
Slashdot posting gives some similar stimuli. By posting an excellent message, the author receives moderation and more people start discussing the idea.
/. "discussions", they are mostly quite shallow trees. Lots of comments on the article; a few of those comments get a long list of (often not very relevant) replies; a very few get replies to those replies. Not much depth there. Very few cases where the discussion really leads to any conclusions.
... ;-)
Actually, if you look at
OTOH, often the first two levels contain actual information or links to information. And this is often more informative than the original article.
(I wouldn't expect that this comment will lead to many, if any, actual conclusions of any sort
... system in that future. It is almost a done deal - when major corporations imagine Linux as central to the future, Linux becomes central to the future.
Well, yes and no. Linux (i.e., the kernel) is really just an instance of POSIX, and code that runs on linux will usually run on other POSIX systems with little more than a recompile. This now includes pretty much all the available unix-like systems, even those that call themselves "BSD".
It's true that there are inevitably problems with drivers, but that's more of a hardware problems than an OS problem. A driver compiled for a different processor may need tweaking to make it work right. But the driver's interface to a new unix-like system probably needs little
tweaking.
Building on linux really just makes your code highly portable, so you can easily treat the OS as a commodity. Put out bids, and use the hardware/OS combination that costs least (among POSIX systems that satisfy your minimum speed requirements).
Linux does tend to win such cost/benefit calculations, since the OS itself has a rather low cost. But this may not be true 5 or 10 years from now.
Part of the cost calculation is something often missed in simplistic analyses: Porting to a new platform does always take a fair amount of work, even when everything goes well. You need a very thorough test suite, and you have to hunt down and fix a flock of subtle bugs due to different compilers, runtime libraries, etc. This is even true when going to the next release of your current platform. Fixing the "small" bugs is especially tricky when the trail leads down into the OS. Vendors are often not very cooperative when you start asking questions about the internals, which they consider proprietary. With linux, all the source is freely available, so it's easy to trace bugs down into the OS. Well, it's easy for expert C hackers.
So porting to linux is usually faster and easier than to commercial systems. But linux has no advantage here over the *BSD systems. It also looks like OSX is becoming (nearly) as open as linux and *BSD.
In any case, I wouldn't bet on IBM being loyal to linux over the long term. More likely, they favor it because of the advantages of a system with freely-available, license-free source. Any other system with these properties (and which has a POSIX library) will be a very real possibility in the future.
To be a PBX replacement, you should be able to interface to telephones. E.g. 100 or 200 of them.
How about just using bluetooth-enabled phones?
This sort of thing isn't all that new a concept. Back in the mid 1980's, I saw and used internal versions of phones via wireless modems at some Motorola sites where I was working. They had both voice and IP traffic going over the wireless system, and various parts of the "PBX" used TCP to transfer voice traffic. It was all very experimental, but much of it worked ok. And this was before RTP was developed to fix the problems with voice over TCP.
One of the fun parts was the electronic id tags you could get, similar in function to RFID, which would let the system locate you. If you carried the tag, phone calls would be routed to the phone that was physically closest to you. So if a nearby phone rang, you would pick it up, and it would be for you or someone else very close. Of course, this did lead to all the obvious discussions about the phone system tracking your every move. Many people just left their id sitting under their desk phone.
The curious thing is that VoIP has taken so long to catch on with "The Market". Here in North America, the phone companies have been reporting that traffic over long lines has been pretty much all converted to IP by now, and the idea is slowly making its way upward through the entire phone system. But they've resisted offering it at the retail level for a very obvious reason: It would kill all that nice income from long-distance charges. With IP, "long distance" isn't a very meaningful concept.
Google has around US$700-million in annual revenues, and it makes about US$100-million a year in profits.
An interesting question is: If google can be used by anyone for free, where does their income come from?
As far as I've read, one of the major sources is that they license their software to corporations, governments, and other organizations that want to get access to their own internal computer systems. This makes sense. Why develop it yourself, when for a much lower price you can get a few google people to come in and get it all running in a few days?
But this in turn brings up a different question: If Microsoft wants its own search engine, why would it want to buy google? Why not just lease the software, as others do? Use it as a starting point, and start building your own plugins to add new capabilities. They'd have an "embraced, extended" google, with all of google's capabilities plus a bunch of new goodies.
The obvious answer: Microsoft isn't just interested in building a new, more powerful search site. They also want to shut down its biggest competitor.
Presumably, their real plan now is to wait for google to go public. Then they can buy up the stock and take over that way. It wouldn't be the first time that a firm has used this approach to eliminate the competition.
Maybe we should be telling google to not go public. Really; what's in it for them, other than the grief of seeing a hostile takeover from the inside?
Yeah, but he's still wrong. Users shouldn't use Windows at home; they should use OSX.
;-). CDs work fine with iTunes (with earphones or a good speaker plugged in). Many TV shows are also available via the internet, and with a cable modem and airport, we can watch them wherever we are, not just where the TV is. Instead of the weather channel, we use weather.com, from anywhere. And so on.
...
A couple of months back, after noticing the growing rate of obscenities coming from my wife while she was using her Windows box (for reasons of compatibility with machines at work, of course), I talked her into trying a Mac Powerbook. Despite her years of experience with Windows, she adapted quickly to the Mac, with no obscenities at all. In only an hour or so, she was saying how much she liked it.
Now, a couple of months later, she will tell you of a few things that she things work better in Windows. But she shudders at the thought of going back, and mutters obscenities when discussing the machines that she uses at work.
As a longtime unix/linux geek, I don't think that the Mac is an improvement. The various X window managers are nearly all better than the Mac (and much better than the Windows GUI). But the Mac is pretty good, and is easy to learn.
One thing that the Mac has going for it that linux can't (yet) duplicate to my knowledge: Since the Mac moved into our house, our use of the TV sets and the CD player has dropped to nearly zero. We have a Netflix account, and their DVDs work just fine in the Mac, anywhere in the house (or back yard
We're seriously looking into disposing of our TVs and just using the wireless Mac. The only real question here is whether we can get internet service via cable without paying for the bundled TV channels, or can we find DSL that's fast enough?
I've been wondering if there's some way we could introduce the Mac and OSX to Red Hat's CEO. Give him a bit of guidance in using it. Maybe he'd turn into a convert, too. Maybe he'd start a Red Hat project to provide a wireless linux-based laptop with lots of good open-source audio and video stuff preconfigured. They could have a real winner there.
Ah, such a dreamer
it is called the biography section. ...
... I poked around in the iTunes store, but I couldn't find it. Where is it hidden?
Hmmm
Actually, most artists are giving you their work for free. All that money you pay goes to the distributors; none of it reaches the artists.
/. for more than a month or so should have seen the data by now.
Recording industry contracts pretty much guarantee that the artist gets nothing (and in fact usually ends up in debt) until at least 1.5 million CDs are sold. For new groups, the crossover point is often much higher, and is hardly ever reached.
But this is hardly news in this forum. Anyone reading
CD singles? Where do you buy those? I don't think I've seen them for sale for years.
... could it be that that's why people aren't buying them?
Hmmm
Well, yeah, but people have been saying for years that most CDs have only one or sometimes two songs that most purchasers want. So in most caseds, a single-tune download has literally replaced a single CD sale.
Now if they could figure out how to deliver the liner notes along with the music.
;-)
Some of us do like to read that stuff.
(And the info about composer - and copyright holder - is useful too. Duh.
Well, yes; sometimes it feels that way. ;-)
Actually, of course, it's just normal American corporate management practices that I'm talking about here. I keep getting the feeling that it's not outsourcing to cheaper parts of the world that we should be worried about. If any other part of the world ever invents a rational scheme for organizing companies, they'll wipe out our economy overnight.
Fortunately, there seems little danger of this threat materializing.
The funniest case was a few years back, when the project's management decreed the Netscape server as the standard. We tried several times. But the same thing always killed the effort: This server can be configured only through its web interface. Invariably, we would make some config mistake that turned the server into a zombie. At that point, there was no way to correct the problem because we couldn't change the configuration any more. We'd wipe the server's directories, reinstall -- and it would happen again. Sometimes we'd get it running for a few days, but every config change carried with it the possibility that we'd have to wipe the server and start over.
You'd think that people would understand why you can't trust a web server to handle changing its own config files. But the managers couldn't be convinced that there was a fundamental problem here. And we never found a way to get at those files with a plain editor. They just didn't make sense, and weren't documented anywhere that we could find.
I've long argued that one of apache's real strengths is its plain-text config file (with lots of good comments in the text). The commercial guys don't seem to be able to figure out why this is a good idea.
[A]ll the software we're using today will have been replaced by 9999.
Back in '99 there was a story about this.
How can this be true when many people run Apache on Windows?
Funny thing about this: On many occasions, I've found myself looking at a group trying to install IIS (or the Netscape server or some other commercial server), and getting more and more frustrated over the problems getting it to work.
So, while they're fighting with it, I sit down at an idle machine, point the browser at apache.org, download the latest apache for that platform, and ask them questions while I twiddle the configuration. Within 10 to 20 minutes, depending on how much configging is needed, I fire up the server, and it runs the first try. I invite them to check it out from the other test machines, and they find that it's working. We copy a few web pages to that machine, and they work
The result in almost all cases is that they decide to go with apache "for a while". It's just an interim measure, you understand, until they can get the real web server running. But meanwhile, they have a web server that they can put online. The web developers aren't sitting around idle; they're building the web site.
In the ensuing months or years, I occasionally prod them with "You know, we really should try to get the officially-mandated web server running." The response is usually to put it off until they can get through the huge pile of stuff that they need to put online.
In a few cases, management has gotten upset, and created a team to get the officially-mandated server running. This often succeeds after a few weeks. Then they put that server online, and it's a real disaster. It crashes repeatedly, produces a flood of complaints from baffled customers along the lines of "How the @#&$^%*& do I order things from you now? Your online ordering pages are broken."
After management notices the loss of income from IIS or whatever, they grudgingly agree to go back to apache "until the problems can be worked out."
Does this sound familiar to anyone?
How is this different than using NFS of FTP ...?
Not materially different at all; it's using HTTP to do what thousands of us have been doing with NFL and FTP for several decades.
For a good parallel, imagine that a new sort of road surface material were developed and started to be used on some highways. If this patent holds up, then we could expect to see patents for methods of driving on this new material. Yes, people have been driving on roads for a long time. But they haven't been driving on this new material, so this would be patentable.
The USPTO is getting to be a real frustration for satirists. You can't think of a bizarre IP-related idea without reading that they just outdid your idea with something that's even more bizarre.
I do note that the anti-caffeine item says "avoid", which isn't quite as absolute as the way the other items are phrased. And it comes with the explanation "Excessive caffeine has been shown to cause low blood sugar, which can make you crave sugar." Note that word "excessive".
;-)
Also, the very next item suggests 8 glasses of water per day.
A reasonable interpretation is that a low level of caffeine, say in the form of unsweetened coffee or tea, for one or two of those glasses of water would not be a material violation of the rules. But you do have to be the sort that can resist any sugar cravings that result. And this might not even happen with a small amount of caffeine, if it is consumed along with the meats, eggs and cheese, which do contain small quantities of carbs.
OTOH, if you're the sort who can't resist your 8 cans of Jolt each day, maybe you'd be better off just going cold turkey on the caffeine.
(I wonder if cocaine would be a good substitute? That's supposed to be a diet drug itself, right?
The day a email tax is born will be the very same day when email will die!
Nah; it'll just morph into something not legally recognized as email.
In fact, just this sort of thing has already happened many times, for several reasons.
Thus, my wife and I have a joint account with a local ISP that also supplies things like email accounts, unix shell accounts, etc. We had been using our mail readers to access this account from home. But one day it stopped working. A few hours of testing verified that our (cable) ISP had started blocking outgoing connections to port 199 except on its own machine. They want all our email stored on their machines, not on some competitor's machines.
It didn't take long before the other ISP announced a solution: They simply set up a web interface to their email accounts. Now our home ISP can't block the access, because they only see a TCP connection to port 80. The url starts with "https://", so the email is encrypted, and they can't tell that it's email. Even if they suspected this (which I'm sure they do), there's not much they can do. They can't block all https pages, because online commerce depends on them.
It's much like the shutdown of Napster. The only real result was to replace it with a whole flock of other P2P services, no two alike. Shutting them down will just annoy your customers and encourage replacements that your blocking/taxing code can't recognize.
Heh; fun argument.
Way back when the US government passed the 55 mph national speed limit, there were a number of people who pointed out the same sort of problem. The usual reason was for saving lives. But a straightforward cost-benefit analysis, using pulicly-available actuarial and highway data, showed that when you added up the extra travel time from the lower speed limit, it would take roughly three lifetimes of travel to match a single traffic death. Since the extra travel time is "dead" time that can be considered stolen from your life, the same reasoning could be applied.
Not that this convinced much of anyone. But it was a fun bit of reasoning to read about.
We do have a problem that not just spammers, but pretty much anyone in a position of authority generally considers wasting your and my time to be not a problem. The only way I've ever found to fight this is when I'm working on an hourly basis, and I can tell someone "This discussion we have just had, which has accomplished little other than prevent me from doing the job I'm being paid to do, has cost your company $N. Should we continue the discussion, or should I get back to work?"
Unfortunately, even this isn't always effective. But sometimes it does result in me not getting invited to meetings.