"You can't blame immigrants for having instructions in multiple languages. You can blame globalization for that. It's cheaper to print a product in multiple languages than it is to print up separate packaging for each locale."
Ok, let's say I give you that one (but what about the products made in the US?)....what is the excuse for signs in a store, or a govt. agency (drivers licenses, etc) in any town in the US that is not a border town to have it in multiple languages?
So you object to American firms trying to sell their products in other countries? Or maybe you just object to American firms saving packaging costs (so they can sell their products at a lower price) by having a single package with labels and instructions in multiple languages.
And you obviously think that tourists in the US shouldn't be helped; they should be required to become fluent in English before they vacation here.
So why are you so opposed to Americans trying to attract foreign customers? Most of the world isn't fluent in English, and America's English-only practices are a good way to discourage the rest of the world from spending money here. Any sensible American businessman (and any government agency that wants to support foreign trade and tourism) has an obvious motive to support multilingual signs, documents, packaging, etc.
Well, I usually like to tell people that I'm "Mongrel American". I'm one of those mixed breeds with ancestors from nearly every country in Europe, plus a few that were never anywhere near Europe before they came over (some of them around 12,000 years ago;-).
Mostly, that phrase just gets grins. I once had someone look at me funny, and when I asked what's wrong, they said that I didn't look oriental. It took me a few seconds to realize what they meant.
Mac's are sluggish. There are plenty of theories as to why,...
Here's a little benchmark I've found that demos one of the reasons: On my Mac laptop (chosen for portable demos like this), I open a couple of Terminal windows and ssh to accounts that I have on a linux (Ubuntu heron right now) and a FreeBSD machine. On those, and in a third Terminal on the Mac, I cd to a directory that has a few thousand subdirectories and several hundred thousand files. I type a "time find..." command in one window, and copy it to the other two. In all three, it completes and prints out a few file names in maybe a minute. So they're about the same speed, right?
Then I hit the up-arrow and Return keys in all three windows. In the linux and FreeBSD windows, the repeated command prints out the same file names and terminates in around 2 or 3 seconds. In the Mac window, it takes about a minute.
I've demoed this on several occasions to mixed crowds of Mac, linux and BSD fanboys. The discussions, and attempts to tweak the behavior, were truly entertaining (in a geekish way). I've seen a number of similar demos in which the Mac loses spectacularly to linux and *BSD systems on similar hardware, doing other easy-to-understand tasks like the above.
I suspect that it's the result of the usual NIH syndrome at Apple. The linux and BSD crowds are fairly open to such criticisms, and respond by studying the situation, looking at other systems' approaches, and coming up with improvements. The Mac crowd tends to get indignant, and responds by insulting the demoer (is that a real word?) and chanting "It Just Works!" OK, I exaggerated a bit there, but you get the idea. Mac users are outsiders to the development process, which is controlled by Apple management, and very little OS X code comes from outsiders. Linux and the *BSD systems all contain large quantities of user-contributed code, and those systems' fanboys know that they can make real contributions to their favorite systems if they so desire. This not only affects users' attitudes; it says a lot about how easy it is to get problems fixed. I suspect that the problem described above is not even recognized as a problem by Apple's management.
What part of "surveillance of American citizens without judicial oversight is illegal" do they not understand?
I'd guess that they probably do understand that, where by "they" we presumably mean the top guys in the current US administration. But in their own words, such laws are "quaint" and "irrelevant".
To put it in some sort of perspective, such laws have historically only been relevant when the government actually wants to take you to court, since illegally collected evidence is usually not accepted by the courts. But if they're interested in you for reasons other than getting you convicted and sent to jail (or exported or executed), then the legality of their actions really isn't a very interesting point. After all, you can't realistically expect to file suit against the surveillance agencies and win anything. If a violation of the law can't end up in court, how is the law relevant to anything?
Yeah, we can post messages complaining about illegal government activities here and in other blogs. Lotta good that does.
Oops, looks like Slashdot doesn't like Chinese character entry. For those interested, the book's title is "Shang Xia Wu Qian Nian".
Yeah; slashdot only allows Latin1; none of those funny eastern characters allowed. Anyway, I did a bit of googling for the book, found lots of sites willing to sell me a copy, but I didn't find any trace of an online copy. Any idea how I might find it?
It'd be fun if there were a version that had the original next to the English translation, but I suppose that'd be too much to hope for.
The wifi mesh isn't useful for G1G1 users, unless maybe you GnGn for n > 1. The mesh is one way that application collaboration is enabled. From my reading, it seems to be problematic in the field, at least for fair sized meshes...
My wife and I got a couple, plus one for a grandkid, and we did a bunch of experimenting. We were never able to find anything that we could call a successful collaboration. We don't consider seeing each others' icons on the screen and "friending" each other to be collaboration, since it doesn't actually accomplish anything. I spent some time poking around in laptops.org to get more info, but that didn't lead to any collaboration. Anyone know where it might be documented?
The grandkid did find a few games that he liked, but we didn't find any way that the games could be played in a collaborative fashion. We keep thinking that we missed something important.
OTOH, we were impressed by how much better the OLPCs used any nearby wifi access points than any of our "grownups'" computers. It can see and use our (and our neighbors') Apple airports at 2 or 3 times the distance that a Mac laptop can, for instance, and it can use wifi APs that various Windows and linux laptops can't successfully use at all. I wonder if we could get the OLPC's wifi software and port it to RedHat and Ubuntu (and XP and Vista;-).
And on the third hand, I wish we could figure out how to use the OLPC's browser's bookmarks and history. It seems to have those features, but we've been defeated at every attempt. WTF is going on with those things? Are they documented somewhere that we can't find (in a way that makes sense to adults with computer experience)?
Imagine the consequences to a person who kept falsely telling millions of people your product would infect their computers. It would surely be grounds for libel.
No, it would be grounds for a charge of slander. This is a standard popular misunderstanding. The general explanation is that slander is spoken, libel is written.
Since software is written by a person or group of persons, if that software said that your product was doing something that it wasn't doing, it should be grounds for a charge of libel against the person or persons who wrote the software. There should be no material difference between this and their making the same claims in written English.
Have there been any court decisions saying that false accusations made by software is not libel on the part of the software's authors? Has any court decided that software isn't "writing"? Are there any legal experts on the topic lurking here who can tell us?
If this hasn't been tested in court, perhaps it's about time someone did. Good luck finding a judge or jury that understands what software is.
I'd bet good money that some voting machines are stored in church basements between elections.
And you'd win your bet, at least here in New England. This afternoon, I was at an event in a church which was a polling place last Tuesday. I happen to know which locked storeroom holds the voting equipment between elections; it's just down the hall from the room today's event was in.
Actually, I suppose they may not call that level the "basement", although it is the church's lowest floor. The church is built on a hillside, and it has doors on three different levels. On the north side, the lowest level's are 4 steps above above street level, so it doesn't look like a "basement" to people entering there. But it is the church's lowest floor, so it would be reasonable to say that the precinct stores the voting equipment in that church's basement.
Around here, there's a long tradition of voting in churches, which often serve as the neighborhood's public meeting hall. Most other voting places are in school gyms. I'd expect that a good fraction of the voting equipment is kept at the same site, in storerooms permanently leased to the local government.
DST costs us plenty in confusion and lost work hours, and in maintaining software that deals with 24x7 matters. All such software must deal with one 23 hour day an one 25 hour day each year. Especially when said software integrates with external software and people it is next to impossible to assure error free transition to or from DST.
Nah; sensible programmers have long understood that there's a simple solution: Software should always use UT internally. Preferably it would use the "second counter" that unix and VMS use internally, though ISO standard format will also work. If a human wants to see a data/time according to their current timezone and DST setting, that human can tell the software the timezone stuff, and the output routines can generate the local time for the human. But local times should never be stored inside computers anywhere. Input times should be translated to UT, using the user's UT/DST settings. Every library for every language has routines to do all this. We just have to use them, and stop storing local times internally.
If you do this, then you have very few software date/time problems. The entire issue is pushed out to the GUI, which simply asks the user for their preferred TZ/DST info and passes it to the date/time formatting routines. To prevent asking the user too often, the GUI can save this info in its per-user data area. But note that this doesn't require storing any local times, only the user id and the TZ/DST settings. Again, all computer systems and languages already have a way of doing this; the programmers just have to use it.
(It does get fun if a user insists on typing a date like 11/9/08. Is that today or two months ago? Or 2 years 9 months in the future? Maybe we can work on teaching them ISO standard date formats. Nah; that'll never be accepted.;-)
The whole idea of having to develop an entire infrastructure and spend so much effort (e.g. writing software, following changes in policies, synchronizing between different DST zones, even manually correcting clocks) just to supposedly save a little energy thanks to "using more sunlight" is beyond idiotic.
It has become even more idiotic here in the US in recent years. For example, our house now contains several "smart" clocks that automatically adjust for DST twice a year. But they use the change dates from before the last time that Congress changed the dates. So if we want them to be accurate, we have to change them four times a year now. There's no provision in any of them for altering the DST on/off dates (though of course the manufacturers would be happy to sell us new clocks with the current dates burnt into their ROMs;-).
both em-dash and hyphen are available on your keyboard btw
What you linked to does not prove what you said. Putting in codes to output the characters you want is not, in my mind, the same as 'available on your keyboard,' and Pedantic-Man isn't especially interested in such nonsense as outputting different types of dashes/hyphens when the 'minus key' on the keyboard will do for the sake of clarity. Pedantic-Man is also pretty lazy.:)
Pedantic-Man says, "Stay out of trouble!"
Good advice. I checked out those en- and em-dash inputs on this Mac Powerbook, and sure enough, I get three different-length dashes. But a hex dump showed me that the en-dash and em-dash are both UTF-8 encoded. So I'll predict that if I enter them here, they won't show up correctly on many readers' screens. Let's try:
- hyphen - en-dash -- em-dash
Now is there a way to find out what fraction of readers see all of those as the proper-length dashes on their screens? Hmmm... Let's try the Preview button and see if it even works on my own screen... Nope; the first two came back as hyphens, and the em-dash came back as a double hyphen. So the claim that I can input them from my keyboard failed spectacularly in this simple case.
The problem, of course, is that there is no universally-accepted encoding for the en-dash or the em-dash. Only the ASCII hyphen works reliably everywhere. If/. accepted UTF-8-encoded input and didn't damage it, AND if/. correctly labelled the text as charset="UTF-8", AND if everyone's browser correctly displayed UTF-8 text, it would have worked. But it's been more than 15 years since Ken T gave us the UTF-8 encoding, and most of the computer world (especially inside the US and Europe) has quietly ignored it.
(Yes, I know that by "most of the computer world" I meant Microsoft. But in this case, MS probably isn't involved; the damage was done by slashdot's software. MS isn't to blame for all of our communication problems. Both Apple and the linux crowd have made snafus out of their attempts to move to UTF-8 and Unicode, and much of the web runs software that damages UTF-8 text with malice aforethought, as/. did to my above test.;-)
"up until a short time ago, artists used to make most, if not all of their money through concerts"
Until a short time ago? That's the way it is now, as it always has been. Musicians don't make money from commercial recordings, unless they're in the top-selling 0.1% of recording artists. In the US, the current estimate is that the break-even point for commercially distributed recordings is around 1.5 million copies; if your recording doesn't sell that many, you lose money.
This is, of course, because the recording companies are able to use "industry standard" contracts that require assigning the copyright for to the recording company. If you don't sign away your rights to your creations, you don't have access to the commercial distribution system.
This is starting to change, with the Internet slowly becoming the distribution system for "minor" artists (i.e., 99.9% of them). But this is just starting, and it's still sufficiently complex that most musicians have to pay someone else to do it, so it's still pretty much true that musicians don't make money from recordings. The only way to make money as a musician is in public performances. And note that, even then, if you sell recordings produced by a big-name company, you won't make money from your CDs. Recordings are only profitable if you use a small local recording company to produce them, and you do all the selling and distribution yourself.
If you're a musician, you should view recordings as publicity. You pay for them, but you shouldn't expect to make any money from them.
Science can't prove that god exists, or that it doesn't exist. So it's a perfectly safe bet- it can never be won.
Uh... what?
They certainly CAN prove that God exists... if he really does.
Actualy, no. As any number of people (notably Karl Popper) have pointed out, scientific methods don't actually prove anything. Science use methods of disproof.
The idea is that you make up explanations ("hypotheses") of things that you and others have observed. Then you think up ways to test those hypotheses. The tests almost always either say that a given hypothesis is wrong (i.e., it predicted a different outcome of the test), or it's inconclusive.
After enough testing, you tend to find that you've disproved most of the hypotheses. The ones left standing eventually graduate to being called "theories". But you still keep trying to debunk them.
Nowhere in this is there ever anything that qualifies as "proof". A scientific theory is merely an explanation that has passed a lot of tests, so it seems the best theory at the moment.
It's fairly well understood among scientists that someone who talk about "scientific proof" isn't a scientist. They may be a journalist, or a mathematician, or theologian, or a bookie, but they don't understand scientific methodology, so they're not a scientist.
And, of course, this is also a tentative conclusion. They could be a (not very good) scientist who doesn't really understand the theoretical foundations of their field. You can test this by challenging them to develop and test explanations of things.
Remember when, just before the invasion of Iraq, and the media had published extensive debunking of all the excuses that Bush's people were using for the attack, so they finally settled on a new tactic that couldn't be refuted: We had to attack Iraq because they were capable of doing stuff to us in the future. This was quickly labeled "the Bush Doctrine", and was understood to mean that unless you were blind, deaf and quadriplegic, you were capable of attacking the US, so the US was justified in a preemptive attack on you. During the shocked silence that followed the public announcement of this doctrine, the US in fact did attack another country that had done nothing to the US. That was widely understood as meaning the Bush Doctrine was for real, and everyone in the world was being implicitly threatened.
You think those people should stop talking and start killing Americans then?
So far they seem to have kept talking, perhaps out of respect for things like the US's nuclear (or nucular) arsenal. Some have perhaps been a bit reluctant to talk too much, out of worry about how the Bush administration might interpret some offhand comment. Once Obama is in office, they might be willing to do a lot more talking, as he appears more amenable to talking that his predecessor.
But we should assume that people in many parts of the world have spent the past 5 years considering and preparing what they might do in case the US government's attention happens to fall on them at some time in the future. Whether any of this will result in physical actions against the US is something we don't know (yet). But things like the US's "extraordinary rendition" program can be expected to produce a desire for revenge, so we may have that to look forward to.
Stay tuned. The End of History was announced a bit prematurely.
he is actually the result of a MIXED black-white marriage
You would never know that from hearing him talk. He calls him self "black" not "bi-racia" or "multi-racial". Nor does he sometimes call himself "black" and other times "white".
For American black people, it would be pointless to point out that he has both African and European ancestors, because this it true for almost all American black people. Of course, he is a bit unusual, because his black+white ancestors actually were married. More often, the ancestral black+white pair was a white man who owned the black woman. This is sufficiently well understood by most Americans that there's no need to mention it.
But I did read an interesting take on it recently: Someone pointed out that, for obvious statistical reasons, black Americans usually have close white relatives, while most white Americans don't have close black relatives. This was presented as an explanation of why white Americans tend to both stereotype and fear black Americans, while black Americans tend to view white people as "just people" who are highly varied. You're a lot less likely to stereotype or fear a group of people if you've grown up close to a bunch of them.
In any case, because dark pigmentation is genetically dominant over light, people who have both dark- and light-skinned ancestors are usually on the dark side. In the US, "black" means having visibly dark skin. So most people of mixed ancestry tend to be considered "black", and any white ancestry is dismissed.
In the US, if you assume that a black person has white ancestors and relatives, you'll usually be right. Barrack Obama isn't the least bit unusual in that regard. What's unusual is that he's an "F1 hybrid".
Oh, and just so you know, most geneticist estimate there is only one human race, and that we all come from, you know, somewhere in Africa. So we all are "mono-racial" in a certain way, right? Like, we are all the same inside? Ebony and Ivory? Kumbaya, Oh Lord, Kumbaya?
Close, but the terminology could be adjusted a bit. We are all one species, with a number of subspecies. As usual, the subspecies interbreed at nearly every opportunity, and hybrids are common when subspecies intermingle. The term "race" isn't much used in biologist circles, but when it is, it's understood as a synonym for "subspecies". Horticulturalists tend to use "variety" for the same concept.
Actually, humans are one of the many species in which there isn't much real genetic difference between subspecies. Thus, in our closest relatives, the chimps, there is more genetic variation in the million or so that survive than there is in the 6 billion humans. A few human subspecies can be distinguished genetically, but they just barely qualify as subspecies, and the differences are rather superficial.
It's also common to point out that there is more genetic variability between African groups than there is between "Africans" and the "Caucasian", "Asiatic" and "Australian" subspecies. Those three are really just minor of within the greater African tree, genetically speaking. Modern white people are the descendants of a group of north Africans ("Cro-Magnons") who migrated to Europe around 40,000 years ago, plus later mixing with an influx of east Africans about 2000 years ago via the Romans' slave trading. The pale skin is a straightforward adaptation to Europe's cool, cloudy climate.
My favorite bit of silliness about "race" is that here in the US, one of the common terms in a list of racial groups is "Hispanic". That illustrates pretty clearly how nonsensical the whole concept is. But it doesn't really discredit the idea that there are human subspecies; it just says that the popular conception is all screwed up and doesn't match the genetic facts.
Makes me wonder if we can finally do away with race based affirmative action now.
Probably not, at least under one reasonable interpretation of Obama's win.
The first election I voted in was the Kennedy/Nixon one. Back then, lots of people said that Kennedy couldn't win, because there were too many Americans who would never vote for a Catholic. In the analysis afterward, people figured out that this was true; there were millions of Americans who didn't vote for him because he was Catholic. But those were mostly people who would never vote for a Democrat, so in fact being Catholic didn't hurt him.
We've already heard a few political analysts say this about Obama. It's probably true that millions of Americans wouldn't vote for him because he's black. But those people would mostly never vote for a Democrat, so it didn't matter.
In both cases, what's really going on is that most Americans will vote for a Catholic, or a black man, or a woman, etc. For example, here in Massachusetts we have a black governor, and there was little sign that his skin color was a problem for most white voters. The real problem has been that the two major parties that have a (media-supported) lock on the election process have refused to nominate such candidates. When the Democrats finally nominated a Catholic, he won. When they finally nominated a black man, he won. When they finally nominate a woman, she will likely win; being female won't hurt here any more than being black hurt Obama or being Catholic hurt Kennedy.
If you think about this, it's actually fairly similar to employment discrimination. In both elections and hiring, the general population has little input into the decision of who to hire/nominate. That decision is made in great part by top management. If the people who make the hiring decisions "know" that a black man (or a Catholic or a woman) can't do the job, it doesn't matter what the general population thinks; no such person is offered the job. Jobs can usually only be done by the people who are hired for the job, and elections can usually only be won by the people nominated for the office.
Having a black man (with a "foreign" name) as president won't have much affect at all on hiring decisions in most companies. Or in most government agencies, for that matter. The people in charge of the hiring decisions will still decide based on the same prejudices, and many capable people will never be hired as a result. Or they may be hired, for low-level position, but never promoted to positions of more responsibility.
Denying this is simply ignoring facts that are fairly obvious to anyone who has ever been involved in the hiring process. And one political party finally deciding to offer the top job to a black man won't affect the general problem very much. Most of us white Americans would have voted for a black (or female or...) president decades ago, if one had ever been nominated. But there have always been powerful forces in the party structures that prevented such nominations. Just as there are powerful forces in most companies that restrict hiring and/or promotions to the "right kind of people".
It seems that "compositing" and "compositor" have become the buzzwords of the month. I can't count how many times they've appeared on my screen in the past few weeks. But attempts to determine what any particular writer means by them generally don't succeed. Any idea what might be meant in this case that's different from what, say, X Windows (or Y Windows or...) does?
Of course, I asked google to "define:compositing". Lots of definitions; none of them very helpful. Thus, I suspect that "A method of reducing the number of drill hole intervals within a database to intervals of equal length for a given drill hole" isn't what was meant here. Nor was "In deliberative procedure, compositing is the process of combining several motions into one composite motion", I'd guess. "The process of combining two or more images to yield a resulting, or 'composite' image" works, obviously, because every graphics app ever written does that. A dumb terminal that draws two characters on its screen does that. But I don't see any clues pointing to what's meant in TFA.
Are these just empty marketing buzzwords, or are they actually describing something that we should pay attention to?
(Of course, I expect a flood of responses with orthogonal definitions and/or vague descriptions that don't clarify much of anything.;-)
In any case, you should consider that we're talking about people who have in the past classified ICMP ("ping") packets as "hacking attacks".
Have you ever mistyped the name or IP address of a machine that you were trying to connect to? Did you try it a couple of times before you realized your mistake? If so, you may well be on a list of people who have "attacked" the actual machine that you were talking to.
When reading stories like this, you should always consider that, unless they tell you details of the attacks, they could be talking about such trivia, and you could be on their list of attackers.
Yes, there are real attackers out there doing scans to find new machines for their probes. There are also real network managers doing scans to map portions of the network and locate problems. And there are people mistyping machine names or addresses. To produce good numbers that get people's attention, it's fairly common for the security people to count all packets from unknown machines as "attempted attacks".
The ping times were a bit longer than most of us are accustomed to. But there's serious talk of extending the Internet to various space probes, and that will make the avian carrier protocol look speedy in comparison.
The more surveillance present on the internet the less useful it will be as a way to transmit information anonymously.
Actually, the Internet has always been highly susceptible to surveillance. This was done intentionally, but with different terminology that matches the motive. The intent was to make it reasonably easy to manage and troubleshoot. I.e., it's supposed to be easy for support people to examine the traffic, diagnose problems, and fix them. It's a large part of why the Internet has been so successful. And if the support crew can examine your packets, then anyone anywhere along the data path can do so.
This may seem odd considering that the early Internet was developed almost entirely with military funding. But it makes sense if you study their reasoning. The security people understood from the start that the only way you can get communication security is with end-to-end encryption.
Trying to push the security to a lower level is counterproductive, because the lower levels are inevitably close to invisible at the application level. This means that security breaches at lower levels will rarely be noticed for some time. And even when you notice a breach, digging into the lower levels of the protocols is inherently difficult for people who don't work with it every day. So they concluded that the IP layer should only worry about getting packets to their destination undamaged. That's difficult enough that you don't want the people working on it to be distracted by security issues; they'll just screw it up and block valid traffic. They don't need to know the contents of packets, just the headers, so if you encrypt all the contents, it doesn't affect the lower levels at all.
Or, more simply: Low-level encryption is a pure waste of cpu time and bandwidth, because you have to do it at the top level anyway. So don't bother. And nothing but top-level end-to-end encryption will give you secure communication.
Yes, this means that anyone can intercept your traffic and save it. If you are relying on this not happening, you can't ever be secure. You have to accept it, and make your data worthless to anyone but the intended recipients.
This was all understood decades ago by the folks who designed the Internet. Complaining about surveillance now really just shows poor understanding of the issues. You can't prevent surveillance on any network, so don't bother. You should be talking about making that surveillance a time and money sinkhole with no results. And you do that by encrypting stuff. There's a lot of research on this topic and most of it is pretty easy to find; go read some of it.
whitehouse.gov is the real official website of the executive branch, while whitehouse.org and whitehouse.com are not (though this example is a bit dated).
How so? Hasn't the White House been a commercial operation for the past 8 years, for sale to anyone for the right price?
Of course, the more cynical among us will claim that it has always been so. Others would suggest that at least whitehouse.org is inappropriate, though it might have been better to suggest that during the Clinton administration.
it works like this: You request a URL in Opera Mini. Opera Mini makes the request to a proxy server run by Opera. OperaÃ(TM)s proxy server connects to the web server hosting the requested URL,...
So you're saying that with Opera Mini, if you connect to your bank via HTTPS, the encryption is between Opera's server and the web site, so Opera sees things like your password and account number in the clear.
This is something that I've asked about a number of "smartphones", and usually I've gotten the old runaround rather than a straight answer. In any case, without access to the gadget's innards, it's difficult to tell how it really works.
This sort of question could be important, because smartphones are more and more being used for such financial purposes. Most people who use them have heard the assurances that "everything is encrypted", and don't understand that if the encryption is done by the ISP, all the data is available to the ISP. It thus takes only a small bribe to an ISP employee to totally defeat whatever security you and your bank think you are using.
This really looks like a disaster in the making. It could be interesting to read discussions of this question. Granted, such discussions would often degenerate to flamewars between sockpuppets, but whatever.
We could start by trying to find answers for specific systems. If someone uses an iPhone to communate with their bank when far from home, does Apple now know all your bank contact info? What about if I use a BlackBerry; what info is now available to the telco or RIM or whoever? And consider Google's new Android phone; does their library do something similar that puts all my bank info into a google.com database? What about a Windows Mobile phone?
A valid answer for any of these could be "We can't know; the low-level code's behavior is hidden." Such an answer should be grounds for ignoring the rest of the discussion until that answer changes.
Slashdot Mirror
So you object to American firms trying to sell their products in other countries? Or maybe you just object to American firms saving packaging costs (so they can sell their products at a lower price) by having a single package with labels and instructions in multiple languages.
And you obviously think that tourists in the US shouldn't be helped; they should be required to become fluent in English before they vacation here.
So why are you so opposed to Americans trying to attract foreign customers? Most of the world isn't fluent in English, and America's English-only practices are a good way to discourage the rest of the world from spending money here. Any sensible American businessman (and any government agency that wants to support foreign trade and tourism) has an obvious motive to support multilingual signs, documents, packaging, etc.
Well, I usually like to tell people that I'm "Mongrel American". I'm one of those mixed breeds with ancestors from nearly every country in Europe, plus a few that were never anywhere near Europe before they came over (some of them around 12,000 years ago ;-).
Mostly, that phrase just gets grins. I once had someone look at me funny, and when I asked what's wrong, they said that I didn't look oriental. It took me a few seconds to realize what they meant.
Mac's are sluggish. There are plenty of theories as to why, ...
Here's a little benchmark I've found that demos one of the reasons: On my Mac laptop (chosen for portable demos like this), I open a couple of Terminal windows and ssh to accounts that I have on a linux (Ubuntu heron right now) and a FreeBSD machine. On those, and in a third Terminal on the Mac, I cd to a directory that has a few thousand subdirectories and several hundred thousand files. I type a "time find ..." command in one window, and copy it to the other two. In all three, it completes and prints out a few file names in maybe a minute. So they're about the same speed, right?
Then I hit the up-arrow and Return keys in all three windows. In the linux and FreeBSD windows, the repeated command prints out the same file names and terminates in around 2 or 3 seconds. In the Mac window, it takes about a minute.
I've demoed this on several occasions to mixed crowds of Mac, linux and BSD fanboys. The discussions, and attempts to tweak the behavior, were truly entertaining (in a geekish way). I've seen a number of similar demos in which the Mac loses spectacularly to linux and *BSD systems on similar hardware, doing other easy-to-understand tasks like the above.
I suspect that it's the result of the usual NIH syndrome at Apple. The linux and BSD crowds are fairly open to such criticisms, and respond by studying the situation, looking at other systems' approaches, and coming up with improvements. The Mac crowd tends to get indignant, and responds by insulting the demoer (is that a real word?) and chanting "It Just Works!" OK, I exaggerated a bit there, but you get the idea. Mac users are outsiders to the development process, which is controlled by Apple management, and very little OS X code comes from outsiders. Linux and the *BSD systems all contain large quantities of user-contributed code, and those systems' fanboys know that they can make real contributions to their favorite systems if they so desire. This not only affects users' attitudes; it says a lot about how easy it is to get problems fixed. I suspect that the problem described above is not even recognized as a problem by Apple's management.
What part of "surveillance of American citizens without judicial oversight is illegal" do they not understand?
I'd guess that they probably do understand that, where by "they" we presumably mean the top guys in the current US administration. But in their own words, such laws are "quaint" and "irrelevant".
To put it in some sort of perspective, such laws have historically only been relevant when the government actually wants to take you to court, since illegally collected evidence is usually not accepted by the courts. But if they're interested in you for reasons other than getting you convicted and sent to jail (or exported or executed), then the legality of their actions really isn't a very interesting point. After all, you can't realistically expect to file suit against the surveillance agencies and win anything. If a violation of the law can't end up in court, how is the law relevant to anything?
Yeah, we can post messages complaining about illegal government activities here and in other blogs. Lotta good that does.
Oops, looks like Slashdot doesn't like Chinese character entry. For those interested, the book's title is "Shang Xia Wu Qian Nian".
Yeah; slashdot only allows Latin1; none of those funny eastern characters allowed. Anyway, I did a bit of googling for the book, found lots of sites willing to sell me a copy, but I didn't find any trace of an online copy. Any idea how I might find it?
It'd be fun if there were a version that had the original next to the English translation, but I suppose that'd be too much to hope for.
The wifi mesh isn't useful for G1G1 users, unless maybe you GnGn for n > 1. The mesh is one way that application collaboration is enabled. From my reading, it seems to be problematic in the field, at least for fair sized meshes ...
My wife and I got a couple, plus one for a grandkid, and we did a bunch of experimenting. We were never able to find anything that we could call a successful collaboration. We don't consider seeing each others' icons on the screen and "friending" each other to be collaboration, since it doesn't actually accomplish anything. I spent some time poking around in laptops.org to get more info, but that didn't lead to any collaboration. Anyone know where it might be documented?
The grandkid did find a few games that he liked, but we didn't find any way that the games could be played in a collaborative fashion. We keep thinking that we missed something important.
OTOH, we were impressed by how much better the OLPCs used any nearby wifi access points than any of our "grownups'" computers. It can see and use our (and our neighbors') Apple airports at 2 or 3 times the distance that a Mac laptop can, for instance, and it can use wifi APs that various Windows and linux laptops can't successfully use at all. I wonder if we could get the OLPC's wifi software and port it to RedHat and Ubuntu (and XP and Vista ;-).
And on the third hand, I wish we could figure out how to use the OLPC's browser's bookmarks and history. It seems to have those features, but we've been defeated at every attempt. WTF is going on with those things? Are they documented somewhere that we can't find (in a way that makes sense to adults with computer experience)?
Imagine the consequences to a person who kept falsely telling millions of people your product would infect their computers. It would surely be grounds for libel.
No, it would be grounds for a charge of slander. This is a standard popular misunderstanding. The general explanation is that slander is spoken, libel is written.
Since software is written by a person or group of persons, if that software said that your product was doing something that it wasn't doing, it should be grounds for a charge of libel against the person or persons who wrote the software. There should be no material difference between this and their making the same claims in written English.
Have there been any court decisions saying that false accusations made by software is not libel on the part of the software's authors? Has any court decided that software isn't "writing"? Are there any legal experts on the topic lurking here who can tell us?
If this hasn't been tested in court, perhaps it's about time someone did. Good luck finding a judge or jury that understands what software is.
I'd bet good money that some voting machines are stored in church basements between elections.
And you'd win your bet, at least here in New England. This afternoon, I was at an event in a church which was a polling place last Tuesday. I happen to know which locked storeroom holds the voting equipment between elections; it's just down the hall from the room today's event was in.
Actually, I suppose they may not call that level the "basement", although it is the church's lowest floor. The church is built on a hillside, and it has doors on three different levels. On the north side, the lowest level's are 4 steps above above street level, so it doesn't look like a "basement" to people entering there. But it is the church's lowest floor, so it would be reasonable to say that the precinct stores the voting equipment in that church's basement.
Around here, there's a long tradition of voting in churches, which often serve as the neighborhood's public meeting hall. Most other voting places are in school gyms. I'd expect that a good fraction of the voting equipment is kept at the same site, in storerooms permanently leased to the local government.
DST costs us plenty in confusion and lost work hours, and in maintaining software that deals with 24x7 matters. All such software must deal with one 23 hour day an one 25 hour day each year. Especially when said software integrates with external software and people it is next to impossible to assure error free transition to or from DST.
Nah; sensible programmers have long understood that there's a simple solution: Software should always use UT internally. Preferably it would use the "second counter" that unix and VMS use internally, though ISO standard format will also work. If a human wants to see a data/time according to their current timezone and DST setting, that human can tell the software the timezone stuff, and the output routines can generate the local time for the human. But local times should never be stored inside computers anywhere. Input times should be translated to UT, using the user's UT/DST settings. Every library for every language has routines to do all this. We just have to use them, and stop storing local times internally.
If you do this, then you have very few software date/time problems. The entire issue is pushed out to the GUI, which simply asks the user for their preferred TZ/DST info and passes it to the date/time formatting routines. To prevent asking the user too often, the GUI can save this info in its per-user data area. But note that this doesn't require storing any local times, only the user id and the TZ/DST settings. Again, all computer systems and languages already have a way of doing this; the programmers just have to use it.
(It does get fun if a user insists on typing a date like 11/9/08. Is that today or two months ago? Or 2 years 9 months in the future? Maybe we can work on teaching them ISO standard date formats. Nah; that'll never be accepted. ;-)
The whole idea of having to develop an entire infrastructure and spend so much effort (e.g. writing software, following changes in policies, synchronizing between different DST zones, even manually correcting clocks) just to supposedly save a little energy thanks to "using more sunlight" is beyond idiotic.
It has become even more idiotic here in the US in recent years. For example, our house now contains several "smart" clocks that automatically adjust for DST twice a year. But they use the change dates from before the last time that Congress changed the dates. So if we want them to be accurate, we have to change them four times a year now. There's no provision in any of them for altering the DST on/off dates (though of course the manufacturers would be happy to sell us new clocks with the current dates burnt into their ROMs ;-).
Good advice. I checked out those en- and em-dash inputs on this Mac Powerbook, and sure enough, I get three different-length dashes. But a hex dump showed me that the en-dash and em-dash are both UTF-8 encoded. So I'll predict that if I enter them here, they won't show up correctly on many readers' screens. Let's try:
- hyphen
- en-dash
-- em-dash
Now is there a way to find out what fraction of readers see all of those as the proper-length dashes on their screens? Hmmm ... Let's try the Preview button and see if it even works on my own screen ... Nope; the first two came back as hyphens, and the em-dash came back as a double hyphen. So the claim that I can input them from my keyboard failed spectacularly in this simple case.
The problem, of course, is that there is no universally-accepted encoding for the en-dash or the em-dash. Only the ASCII hyphen works reliably everywhere. If /. accepted UTF-8-encoded input and didn't damage it, AND if /. correctly labelled the text as charset="UTF-8", AND if everyone's browser correctly displayed UTF-8 text, it would have worked. But it's been more than 15 years since Ken T gave us the UTF-8 encoding, and most of the computer world (especially inside the US and Europe) has quietly ignored it.
(Yes, I know that by "most of the computer world" I meant Microsoft. But in this case, MS probably isn't involved; the damage was done by slashdot's software. MS isn't to blame for all of our communication problems. Both Apple and the linux crowd have made snafus out of their attempts to move to UTF-8 and Unicode, and much of the web runs software that damages UTF-8 text with malice aforethought, as /. did to my above test. ;-)
"up until a short time ago, artists used to make most, if not all of their money through concerts"
Until a short time ago? That's the way it is now, as it always has been. Musicians don't make money from commercial recordings, unless they're in the top-selling 0.1% of recording artists. In the US, the current estimate is that the break-even point for commercially distributed recordings is around 1.5 million copies; if your recording doesn't sell that many, you lose money.
This is, of course, because the recording companies are able to use "industry standard" contracts that require assigning the copyright for to the recording company. If you don't sign away your rights to your creations, you don't have access to the commercial distribution system.
This is starting to change, with the Internet slowly becoming the distribution system for "minor" artists (i.e., 99.9% of them). But this is just starting, and it's still sufficiently complex that most musicians have to pay someone else to do it, so it's still pretty much true that musicians don't make money from recordings. The only way to make money as a musician is in public performances. And note that, even then, if you sell recordings produced by a big-name company, you won't make money from your CDs. Recordings are only profitable if you use a small local recording company to produce them, and you do all the selling and distribution yourself.
If you're a musician, you should view recordings as publicity. You pay for them, but you shouldn't expect to make any money from them.
Actualy, no. As any number of people (notably Karl Popper) have pointed out, scientific methods don't actually prove anything. Science use methods of disproof.
The idea is that you make up explanations ("hypotheses") of things that you and others have observed. Then you think up ways to test those hypotheses. The tests almost always either say that a given hypothesis is wrong (i.e., it predicted a different outcome of the test), or it's inconclusive.
After enough testing, you tend to find that you've disproved most of the hypotheses. The ones left standing eventually graduate to being called "theories". But you still keep trying to debunk them.
Nowhere in this is there ever anything that qualifies as "proof". A scientific theory is merely an explanation that has passed a lot of tests, so it seems the best theory at the moment.
It's fairly well understood among scientists that someone who talk about "scientific proof" isn't a scientist. They may be a journalist, or a mathematician, or theologian, or a bookie, but they don't understand scientific methodology, so they're not a scientist.
And, of course, this is also a tentative conclusion. They could be a (not very good) scientist who doesn't really understand the theoretical foundations of their field. You can test this by challenging them to develop and test explanations of things.
How many countries has Bush threatened so far?
All of them, actually, back in 2003.
Remember when, just before the invasion of Iraq, and the media had published extensive debunking of all the excuses that Bush's people were using for the attack, so they finally settled on a new tactic that couldn't be refuted: We had to attack Iraq because they were capable of doing stuff to us in the future. This was quickly labeled "the Bush Doctrine", and was understood to mean that unless you were blind, deaf and quadriplegic, you were capable of attacking the US, so the US was justified in a preemptive attack on you. During the shocked silence that followed the public announcement of this doctrine, the US in fact did attack another country that had done nothing to the US. That was widely understood as meaning the Bush Doctrine was for real, and everyone in the world was being implicitly threatened.
You think those people should stop talking and start killing Americans then?
So far they seem to have kept talking, perhaps out of respect for things like the US's nuclear (or nucular) arsenal. Some have perhaps been a bit reluctant to talk too much, out of worry about how the Bush administration might interpret some offhand comment. Once Obama is in office, they might be willing to do a lot more talking, as he appears more amenable to talking that his predecessor.
But we should assume that people in many parts of the world have spent the past 5 years considering and preparing what they might do in case the US government's attention happens to fall on them at some time in the future. Whether any of this will result in physical actions against the US is something we don't know (yet). But things like the US's "extraordinary rendition" program can be expected to produce a desire for revenge, so we may have that to look forward to.
Stay tuned. The End of History was announced a bit prematurely.
For American black people, it would be pointless to point out that he has both African and European ancestors, because this it true for almost all American black people. Of course, he is a bit unusual, because his black+white ancestors actually were married. More often, the ancestral black+white pair was a white man who owned the black woman. This is sufficiently well understood by most Americans that there's no need to mention it.
But I did read an interesting take on it recently: Someone pointed out that, for obvious statistical reasons, black Americans usually have close white relatives, while most white Americans don't have close black relatives. This was presented as an explanation of why white Americans tend to both stereotype and fear black Americans, while black Americans tend to view white people as "just people" who are highly varied. You're a lot less likely to stereotype or fear a group of people if you've grown up close to a bunch of them.
In any case, because dark pigmentation is genetically dominant over light, people who have both dark- and light-skinned ancestors are usually on the dark side. In the US, "black" means having visibly dark skin. So most people of mixed ancestry tend to be considered "black", and any white ancestry is dismissed.
In the US, if you assume that a black person has white ancestors and relatives, you'll usually be right. Barrack Obama isn't the least bit unusual in that regard. What's unusual is that he's an "F1 hybrid".
Oh, and just so you know, most geneticist estimate there is only one human race, and that we all come from, you know, somewhere in Africa. So we all are "mono-racial" in a certain way, right? Like, we are all the same inside? Ebony and Ivory? Kumbaya, Oh Lord, Kumbaya?
Close, but the terminology could be adjusted a bit. We are all one species, with a number of subspecies. As usual, the subspecies interbreed at nearly every opportunity, and hybrids are common when subspecies intermingle. The term "race" isn't much used in biologist circles, but when it is, it's understood as a synonym for "subspecies". Horticulturalists tend to use "variety" for the same concept.
Actually, humans are one of the many species in which there isn't much real genetic difference between subspecies. Thus, in our closest relatives, the chimps, there is more genetic variation in the million or so that survive than there is in the 6 billion humans. A few human subspecies can be distinguished genetically, but they just barely qualify as subspecies, and the differences are rather superficial.
It's also common to point out that there is more genetic variability between African groups than there is between "Africans" and the "Caucasian", "Asiatic" and "Australian" subspecies. Those three are really just minor of within the greater African tree, genetically speaking. Modern white people are the descendants of a group of north Africans ("Cro-Magnons") who migrated to Europe around 40,000 years ago, plus later mixing with an influx of east Africans about 2000 years ago via the Romans' slave trading. The pale skin is a straightforward adaptation to Europe's cool, cloudy climate.
My favorite bit of silliness about "race" is that here in the US, one of the common terms in a list of racial groups is "Hispanic". That illustrates pretty clearly how nonsensical the whole concept is. But it doesn't really discredit the idea that there are human subspecies; it just says that the popular conception is all screwed up and doesn't match the genetic facts.
Makes me wonder if we can finally do away with race based affirmative action now.
Probably not, at least under one reasonable interpretation of Obama's win.
The first election I voted in was the Kennedy/Nixon one. Back then, lots of people said that Kennedy couldn't win, because there were too many Americans who would never vote for a Catholic. In the analysis afterward, people figured out that this was true; there were millions of Americans who didn't vote for him because he was Catholic. But those were mostly people who would never vote for a Democrat, so in fact being Catholic didn't hurt him.
We've already heard a few political analysts say this about Obama. It's probably true that millions of Americans wouldn't vote for him because he's black. But those people would mostly never vote for a Democrat, so it didn't matter.
In both cases, what's really going on is that most Americans will vote for a Catholic, or a black man, or a woman, etc. For example, here in Massachusetts we have a black governor, and there was little sign that his skin color was a problem for most white voters. The real problem has been that the two major parties that have a (media-supported) lock on the election process have refused to nominate such candidates. When the Democrats finally nominated a Catholic, he won. When they finally nominated a black man, he won. When they finally nominate a woman, she will likely win; being female won't hurt here any more than being black hurt Obama or being Catholic hurt Kennedy.
If you think about this, it's actually fairly similar to employment discrimination. In both elections and hiring, the general population has little input into the decision of who to hire/nominate. That decision is made in great part by top management. If the people who make the hiring decisions "know" that a black man (or a Catholic or a woman) can't do the job, it doesn't matter what the general population thinks; no such person is offered the job. Jobs can usually only be done by the people who are hired for the job, and elections can usually only be won by the people nominated for the office.
Having a black man (with a "foreign" name) as president won't have much affect at all on hiring decisions in most companies. Or in most government agencies, for that matter. The people in charge of the hiring decisions will still decide based on the same prejudices, and many capable people will never be hired as a result. Or they may be hired, for low-level position, but never promoted to positions of more responsibility.
Denying this is simply ignoring facts that are fairly obvious to anyone who has ever been involved in the hiring process. And one political party finally deciding to offer the top job to a black man won't affect the general problem very much. Most of us white Americans would have voted for a black (or female or ...) president decades ago, if one had ever been nominated. But there have always been powerful forces in the party structures that prevented such nominations. Just as there are powerful forces in most companies that restrict hiring and/or promotions to the "right kind of people".
I have a friend named Melanie who is known as "Mel", and I suspect that she'd just think it funny if a computer decided she's male.
I'd guess that the typical hiker or cyclist would prefer something a bit lighter than that.
It seems that "compositing" and "compositor" have become the buzzwords of the month. I can't count how many times they've appeared on my screen in the past few weeks. But attempts to determine what any particular writer means by them generally don't succeed. Any idea what might be meant in this case that's different from what, say, X Windows (or Y Windows or ...) does?
Of course, I asked google to "define:compositing". Lots of definitions; none of them very helpful. Thus, I suspect that "A method of reducing the number of drill hole intervals within a database to intervals of equal length for a given drill hole" isn't what was meant here. Nor was "In deliberative procedure, compositing is the process of combining several motions into one composite motion", I'd guess. "The process of combining two or more images to yield a resulting, or 'composite' image" works, obviously, because every graphics app ever written does that. A dumb terminal that draws two characters on its screen does that. But I don't see any clues pointing to what's meant in TFA.
Are these just empty marketing buzzwords, or are they actually describing something that we should pay attention to?
(Of course, I expect a flood of responses with orthogonal definitions and/or vague descriptions that don't clarify much of anything. ;-)
In any case, you should consider that we're talking about people who have in the past classified ICMP ("ping") packets as "hacking attacks".
Have you ever mistyped the name or IP address of a machine that you were trying to connect to? Did you try it a couple of times before you realized your mistake? If so, you may well be on a list of people who have "attacked" the actual machine that you were talking to.
When reading stories like this, you should always consider that, unless they tell you details of the attacks, they could be talking about such trivia, and you could be on their list of attackers.
Yes, there are real attackers out there doing scans to find new machines for their probes. There are also real network managers doing scans to map portions of the network and locate problems. And there are people mistyping machine names or addresses. To produce good numbers that get people's attention, it's fairly common for the security people to count all packets from unknown machines as "attempted attacks".
Note that RFC 1149 has been implemented and publicly demoed.
The ping times were a bit longer than most of us are accustomed to. But there's serious talk of extending the Internet to various space probes, and that will make the avian carrier protocol look speedy in comparison.
The more surveillance present on the internet the less useful it will be as a way to transmit information anonymously.
Actually, the Internet has always been highly susceptible to surveillance. This was done intentionally, but with different terminology that matches the motive. The intent was to make it reasonably easy to manage and troubleshoot. I.e., it's supposed to be easy for support people to examine the traffic, diagnose problems, and fix them. It's a large part of why the Internet has been so successful. And if the support crew can examine your packets, then anyone anywhere along the data path can do so.
This may seem odd considering that the early Internet was developed almost entirely with military funding. But it makes sense if you study their reasoning. The security people understood from the start that the only way you can get communication security is with end-to-end encryption.
Trying to push the security to a lower level is counterproductive, because the lower levels are inevitably close to invisible at the application level. This means that security breaches at lower levels will rarely be noticed for some time. And even when you notice a breach, digging into the lower levels of the protocols is inherently difficult for people who don't work with it every day. So they concluded that the IP layer should only worry about getting packets to their destination undamaged. That's difficult enough that you don't want the people working on it to be distracted by security issues; they'll just screw it up and block valid traffic. They don't need to know the contents of packets, just the headers, so if you encrypt all the contents, it doesn't affect the lower levels at all.
Or, more simply: Low-level encryption is a pure waste of cpu time and bandwidth, because you have to do it at the top level anyway. So don't bother. And nothing but top-level end-to-end encryption will give you secure communication.
Yes, this means that anyone can intercept your traffic and save it. If you are relying on this not happening, you can't ever be secure. You have to accept it, and make your data worthless to anyone but the intended recipients.
This was all understood decades ago by the folks who designed the Internet. Complaining about surveillance now really just shows poor understanding of the issues. You can't prevent surveillance on any network, so don't bother. You should be talking about making that surveillance a time and money sinkhole with no results. And you do that by encrypting stuff. There's a lot of research on this topic and most of it is pretty easy to find; go read some of it.
whitehouse.gov is the real official website of the executive branch, while whitehouse.org and whitehouse.com are not (though this example is a bit dated).
How so? Hasn't the White House been a commercial operation for the past 8 years, for sale to anyone for the right price?
Of course, the more cynical among us will claim that it has always been so. Others would suggest that at least whitehouse.org is inappropriate, though it might have been better to suggest that during the Clinton administration.
it works like this: You request a URL in Opera Mini. Opera Mini makes the request to a proxy server run by Opera. OperaÃ(TM)s proxy server connects to the web server hosting the requested URL, ...
So you're saying that with Opera Mini, if you connect to your bank via HTTPS, the encryption is between Opera's server and the web site, so Opera sees things like your password and account number in the clear.
This is something that I've asked about a number of "smartphones", and usually I've gotten the old runaround rather than a straight answer. In any case, without access to the gadget's innards, it's difficult to tell how it really works.
This sort of question could be important, because smartphones are more and more being used for such financial purposes. Most people who use them have heard the assurances that "everything is encrypted", and don't understand that if the encryption is done by the ISP, all the data is available to the ISP. It thus takes only a small bribe to an ISP employee to totally defeat whatever security you and your bank think you are using.
This really looks like a disaster in the making. It could be interesting to read discussions of this question. Granted, such discussions would often degenerate to flamewars between sockpuppets, but whatever.
We could start by trying to find answers for specific systems. If someone uses an iPhone to communate with their bank when far from home, does Apple now know all your bank contact info? What about if I use a BlackBerry; what info is now available to the telco or RIM or whoever? And consider Google's new Android phone; does their library do something similar that puts all my bank info into a google.com database? What about a Windows Mobile phone?
A valid answer for any of these could be "We can't know; the low-level code's behavior is hidden." Such an answer should be grounds for ignoring the rest of the discussion until that answer changes.