Stories that support the concept that AT&T does not have the consumer's best interest in mind will of course get coverage, in fact, a disproportionate amount as compared to other providers or stories that reflect well on AT&T.
Such negative stories always get much more press everywhere than feel-good stories about how wonderful a company is.
Stories like this really helped block the initial adoption of DSL. When it first came out, there was a flood of stories about sleazy salesmen selling customers what they thought was a $50/month or so DSL plan. Then the customers got their first bill, often for several thousand dollars. The phone companies mostly replied "Hey, they signed the contract; they owe us the money." The word got out, and sales dragged for years. Finally, the telcos grudgingly started telling people the actual cost of the service, and sales slowly increased.
We had DSL in our house for several years, but not through Verizon, for exactly this sort of reason. After seeing lots of reports of how honest speakeasy was, and no reports of their pulling sleazy tricks like this, we finally decided to use them. We only switched to a cable company recently when we needed higher speed (for "telecommuting") than what speakeasy could supply. Even then, we did a lot of careful checking to verify that RCN's "home business" plan was actually what they said it was, and didn't have hidden charges or restrictions, just $X/month for "$Y up $Z down" service.
The telcos have a nasty history of stories like this. They've played such games with every new technology that comes along. Look at the early days of cell phones for lots of examples. The only way to fight it, really, is to publicise such stories to the max. If we can do enough damage to their sales by such publicity, they eventually get the hint and start acting a bit more honest.
This is just one more data point explaining why, ever since client-side scripting was first introduced into browsers, those of us who understand the Web have done most of our browsing with scripting turned off. If you permit strangers to download and run code on your machine, you're just inviting them to take advantage of you like this. And such injection attacks demonstrate that the dangers lie not just in the sites that you visit, but also with any machines along the path of your packets. Naturally, this would be no surprise to any network programmer.
Forging reset packets does not equal "throttling",...
Actually, "throttle" has been used for at least a century as a euphemism for "kill", at least among the criminal elements here in the US. So, if you view Comcast as part of the organized-crime community, their referring to RST packets as "throttling" is right in line with standard criminal jargon.
Are you saying that there are secret regulations not available to the public that they must still follow?
Yup; google for "secret law" to read about it. Here's an article about one recently-passed US secret law that might affect a few/. readers.
The wikipedia article on secret law is short, but has a few good links to other informative articles.
The old saw that "Ignorance of the law is no excuse" no longer applies in the US; it's more like "Ignorance of the law is legally enforced". We have laws on the books that you must obey, but you can't read. Unauthorized knowledge of the text of the secret laws is itself a crime, as is informing anyone else of the laws' text.
There have recently been reports of judges recusing themselves from a case on the grounds that they didn't have clearance to read the applicable laws.
Another interesting case, if you can find the history from a decade ago, is the newton.com domain. It was owned by Mark Newton for years, and used by his small software shop, until Apple came out with the Newton and wanted the domain. Apple didn't negotiate; they just gave him the choices of giving up the domain or being bankrupted by the lawyers that Apple would sic on him. There were a few stories recently about Apple renewing the registration for newton.com, although they no longer actually use it for anything. It now just redirects to apple.com.
So companies can be nice about it, or extremely nasty, or anything in between. I wonder if the US has any more legal protection from such things now than it did 10 or 15 years ago. Probably not; the big corporations can always bankrupt the little independents.
I'll show you. I'm going to make tiny.png's of each letter and lay them out as images.
Don't laugh too hard. When they first tried to get onto the Internet, that's more or less what the folks in Japan, Korea and China had to do. There are still Asian web sites that work that way. Recently, there has been a slow transition to UTF-8, but 10 years ago, hardly any software supported it. And it's still how a lot of printing software works, since most postscript and PDF printers still only use 8-bit characters internally.
Per-char images also have an advantage that they can be made big enough to legibly contain some of the more complex CJK characters, so they don't turn into a blur when the font size is too small. You can string a bunch of them on a line, and programs that handle images will space the lines to match the tallest chars; even when UTF-8 "works", you often get the chars drawn too small to read.
We still have a long way to go before we support all the world's languages sensibly. It shouldn't be so difficult, of course, since it's just a bunch of glyphs indexed by an integer. But when the organizations building the software are headed by people who are contemptuous of all languages but their own, it can take a very long time to do even the simplest things right.
(Well, except for Arabic-derived scripts, which are the worst case for everything.;-)
You're doing it wrong. Both yogh and wynn have unicode code points.
Yup, and they look fine in my firefox window.
So on to the next question: How do I get such non-Latin1 chars printed correctly on a postscript printer? Anyone know of a general solution?
We have a collection of different printers here, and also several OSs (linux, OSX, Windows XP and Vista). All of them tend to produce Latin1 gibberish (aka mojibake) for UTF-8 chars. I've been looking for ways to get text in languages like Arabic and Mandarin printed correctly, so far without notable success. I'd also like to print some Old English at times, too. Anyone know how to do it right?
I think the majority of responses will be: "Why do I need all these flashy fonts on the web anyway! I have my browser show every website in Courier 10,..."
Heh. Do you also force a white-on-green or orange-on-black color scheme?
Actually, those who have minimally explored their browsers' capabilities have noticed that the browsers automatically fall back to a font that they have. (Whether this is a majority of Web users, we really don't know.) After all, you really can't rely on your machine having any particular font. The only sensible thing to do in that case is to pick some other font that your machine does have, and that's what most software does.
What we really should be doing is pointing out publicly that there is no reason whatsoever that anyone should ever install any specific font that some vendor supplies. You should just install the fonts that you like, and which you can read easily. Anything else is a waste of disk space and memory. Those people who insist that their font is the only one to use are violating the basic design goal of the Web, which was that a "page" should be sensibly viewable on whatever output device the user has, and will be formatted to fit that device. And if the user is blind or visually impaired, at least one of the "fonts" should be auditory (for those with hearing) or Braille-like. The web designer's desires don't matter, and neither do the vendor's; all that matters is what works well for the end user.
Pages that don't work that way are simply user-hostile, i.e., wrong.
(I rarely specify a font in any of my web pages. That makes the pages smaller, so they download faster, and they appear in the user's favorite font without any wasted cpu time. The only exceptions are when a boss explicitly orders use of a font, and I like to point out that many users will just override it or won't have the font. But then, I like to be nice to the visually handicapped, even if people are always complaining that my screen's fonts are too small for them to read.;-)
There isn't anything wrong with preferring fast incorrect results over slow correct results, but most people probably want slow and correct to be the default if given the choice.
Well, I'd be a bit careful about making such general statements. There is evidence that people aren't generally that intelligent.
I remember back in the 1970s, when I was at a large university that shall remain unnamed, and a bunch of CS people did a detailed study of the Fortran that accounted for fully half the runs on the campus's central mainframe (which shall also remain unnamed). They found that fully half the runs produced at least some incorrect output due to undetected integer overflows. The hardware gave interrupts for floating-point overflows, but for integers, it just set a flag bit, and you needed to test that flag to catch overflows. The compiler had an option to generate such tests, but it was off by default. The vendor said they did this because they had found that most customers preferred faster code.
The local gang didn't believe this, so they did a bit of a survey. They asked lots of users of the Fortran code whether they would prefer their programs to catch all arithmetic errors if this meant that the code ran slower, or if they would prefer faster code that sometimes didn't catch errors. Roughly 90% of the people they asked this said that they'd want the faster code. Later on, I ran across references to similar tests at other schools, with similar results.
Personally, I was shocked by this. This mainframe was used to do the computing for most of the scientific work on campus, and scientific computing was almost entirely done in Fortran. So half their data runs had undetected incorrect output. They now knew this, and they still preferred the faster speed to correct output.
Somehow, I suspect that this situation hasn't changed. I've dug into various programming languages since then, to learn how they handle this and other potential sources of erroneous results. Most current languages still ignore things like overflow flags by default. Some have no way of enabling the tests of such flags.
Yes, I know lots of ways of explicitly testing for such errors myself. I've done it a lot, because I know I can't rely on others to enable the builtin tests (when they exist) when they recompile the code. But when looking at other people's code, I almost never see anything that will detect overflows. When you're N levels deep in function calls, you usually have no way of verifying the possible range of the current function's args, so there's no way of proving that an overflow can't happen.
Sometimes I'm amazed that our systems run as well as they do, given this sort of nonchalant attitude towards known sources of hardware errors. And I do a lot of paranoid, defensive programming, even though I know that my employers probably don't want it because it slows down the software.
Home compiled software can easily be a source of security holes, as tracking what you have compiled versus what is patched using vendor updates adds significant management overhead and another point of failure.
Funny; pretty much all my past employers have had the opposite policy: Anything of importance, especially customer-facing servers, are rarely or never automatically upgraded from vendors' releases. They all seem to have learned this lesson the hard way, from having their servers down (or compromised) after installing a vendor's upgrade. So they all insist that upgrades be first tested in isolation in a lab setting. Usually this includes compiling from the sources when sources are available. I always recommend this so that we can enable and/or install logging/debugging features that vendors often suppress for the slight performance improvement this usually gives. But they usually seem to have lots of experience telling them to not trust vendors' upgrades without testing.
And, of course, the security folks have long been telling us all that if you want a secure system, one of the primary rules is that you never run binaries that you haven't compiled yourself. That's just inviting vendors to slip backdoors and spyware into your system. But, of course, this is irrelevant if you don't have employees who are able (and have the time) to study the code. In that case, you might as well just install your vendors' upgrades, and hope that they're honest. And you will get bitten by this sooner or later.
After all, just a year or so back, we had the entertaining example of Sony sneaking rootkits into customers' computers via an audio CD. Who ever thought that you'd have to watch out for a case like that?
Well, I'm anything but a hardcore Perl hacker -- just use it to pragmatically list some rubbish now and then -- and I've never even heard of compiling your own Perl.
You certainly don't qualify as a perl hacker! NTTAWWT.;-)
From the very beginning, the primary (and recommended) way to get perl has been to download the source and compile it yourself. It's true that most unix/linux distros have included perl for a decade or so, and of course it's usually not the current version. But this is only a minor time saver during installation. I've often upgraded to the current perl while installing systems, and I've always found it easy.
The idea that a perl user wouldn't even have heard about the easy availability of the source is sorta surprising. That's the way you're supposed to get it, after all. The standard textbooks start off telling you where to get it (or at least they did the last time I looked at one, which has been a while;-).
OTOH, perl does have a bit of a reputation for being solid and without any problems, so it's easy to see how someone might be lazy and just use whatever the vendor supplies. I wonder what went wrong with the RH release?
However, the fact that they've published is stopping anybody else than the authors (like a patent troll) to file for a patent, right?
Nope; that doesn't necessarily stop anyone from filing. All you have to do is fill out the papers and pay the application fee. Of course, if the Patent Office discovers that someone else published it earlier, they'll reject your application. But you can certainly file. If your lawyers can write it with terminology that's sufficiently obscure, the Patent Office's examiners might not notice that it's a dup of an earlier patent.
The really problematical aspect is that in the US, if you have sufficiently many millions of dollars to spend on court costs, even if your patent will be eventually rejected, you can use the legal system to block others (including the original inventors) from using the patent for the decade or so that it takes you to lose. If you're fighting a small start up, you can probably bankrupt them before you lose. And you stand a good chance of being able to buy up their rights during their bankruptcy proceedings.
The only real defense they have against this is if they have filed their own patent application before you did (and your lawyers didn't do a good enough job of obscuring your application). Then the Patent Office will probably notice what you're doing, and will deny your application very quickly.
So has the team applied for a patent? We wouldn't want just any ISP to be able to use this algorithm, would we? And if they don't patent it, one of the many patent-troll companies will, denying the researchers the right to use the results of their own work.
Funny thing: I checked with a few of the usual map sites, and found both "Neander Tal" and "Neandertal". One map (I forget which one) had both at different points. Another actually had the obsolete 'h'; I wonder if that was a subtle joke on the part of some obscure map worker. Anyhow, I'm not too surprised that you'd see such inconsistency. Despite their stereotype, Germans tend to be nearly as sloppy about spelling as the rest of the world. And the maps may not have been made by Germans.
I also read a few more of google's hits for "Neanderthal Neandertal", and found comments by a fair number of self-described paleoanthropologists and archaeologists. My impression is that they mostly consider it a somewhat silly discussion, sorta like how most astronomers view the "raging debate" over whether Pluto is a planet. Several admit to mixing the two spellings intentionally, just to keep the pot stirred. They do mostly seem to understand that the taxonomic name has the obsolete 'h', while the German place name doesn't. The main comments can probably be summarized as "Yeah, so?" I like to omit the 'h', on the grounds that it saves typing time and bandwidth. Of course, I've thrown away all those savings with a couple of posts here.
But it's always fun to have pseudo-debates on such inconsequentiae on forums like/.. (And do I really need that extra dot on the end of that sentence?;-)
One of my favorite t-shirts is the one that reads "Does anal retentive have a hyphen?" I like to point out that it really should have quotes around the "anal retentive".
With all the messages here so far, I'm surprised that nobody has posted the obvious auto analogy:
It's like an auto ad that says the car can drive anywhere on the public street system. So you buy it, and discover that it has a sensor that determines the road-surface type, and the engine turns off if you try to drive on any surface except asphalt. The car company excuses their misleading ad by saying that nobody needs to drive on concrete or (God forbid) gravel. You should be able to get anywhere you need to go on asphalt.
Now aren't you ashamed of not posting this analogy first?
That's great and all if you are an internet mechanic. But what if you just want to drive the damn car? For those people, who are the majority, those messages don't mean squat.
Y'know, this is one of the cases where the standard automotive analogy is a fairly good one.
There are a lot of people in the world who "just want to drive the damn car" and can't be bothered with safety. That's why we have so many thousands of traffic fatalities every year. It's also why most of the world has implemented at least minimal driver education before getting a driver's license. But this isn't done very well, and we still have too many traffic fatalities.
With autos, our century or so of experience has led to the legal requirement of a number of safety enhancements to the equipment. Maybe in another century or so, we'll also have laws mandating the use of various security enhancement to traffic on the public Internet.
Nah; it'll never happen. As with cars, we'll get token "security theater" laws that add only minimal real security, and mostly just produce a bogus safe feeling in the minds of non-technical users.
And if all your data is going through an MitM, in what sense is the initial handshake that establishes your connection secure?
And we might add that most of us are always going through a MitM [Man in the Middle], which we call an ISP.
I was tempted to add a smiley to that, but decided that it would be misleading, because there's really not much of a joke there. We recently read of Comcast getting into a bit of hot water because they were truly acting as a MitM, faking messages between the ends of a TCP connect to get them to believe that the other end had announced an end to the connection. They called it "traffic management", but it was pure classical MitM attacks.
We're now reading of not-too-secret discussions of the implementation of "deep packet inspection" by ISPs, to collect data on what's going over a connection. This is being implemented both for marketing purposes and to share the data with interested government agencies. Again, this is about as MitM as you can get.
The whole "Net Neutrality" issue is also an acknowledgement that ISPs can and do act as a MitM with the goal of controlling and sometimes interfering with their customers' data.
It might be a good idea if we were to rephrase our discussions so as to acknowledge that for most of the Internet, there's always a MitM, usually several of them, and any of them is likely to both examine and alter the data in transit. Anyone who thinks of a MitM as a sporadic, low-probability intruder is stuck back in the 1980s, and is ignoring the nature of the modern commercial Internet.
Maybe we could subtly start replacing "MitM" with "ISP" in our discussions. That might get the idea across to readers that this isn't a purely hypothetical topic.
Well, actually (;-), I've read statements from a number of scientific organizations that have given their official OK to both "Neandertal" and "Neanderthal". (And MS's spell checker is hardly an authority for either scientific or German spelling.;-)
The difference dates back to the official spelling reforms in German in the late 1800s and early 1900s. One of the many changes was to eliminate most unpronounced letters, including the 'h' in "th" combinations. So the noun "Thal" (valley) became "Tal" in standard German.
But, of course, the scientific name of the Neandert[h]al (sub)species had been established decades earlier, using the old spelling. Also, Germans often like to use what are now deprecated spellings like "Thal", much like we do in English with words like "ye olde" instead of "the old".
The official taxonomic spelling remains "Neanderthal". Taxonomic names almost always keep the same spelling even when they're known to be incorrect, because in scientific writing, the exact sequence of letters is important and must remain stable. But you often see "Neandertal" and "neandertal" used in informal scientific writing, often in the same paragraph as the taxonomic spelling with the 'h'. Some writers like to omit the 'h' in English text to encourage the correct pronunciation, since to German-speaking ears (;-) the English "th" sounds distinctly odd. ("Why do Americans lisp like that?") The folks in the Neander Valley don't use that weird English phoneme, and many English-speaking scientists pronounce "Neanderthal" in the German fashion with a/t/ phoneme.
And if you dig around, you'll find that the appropriate scientific societies either don't care, or officially consider the modern German spelling acceptable anywhere but in the (sub)species' official name. And if you're writing about the valley where the first skeleton was found, the correct spelling is "Neantertal" (unless you're in marketing and are aiming for an olde-tyme effect;-).
If you google for "Neandertal Neanderthal" (in either order), you'll find around 74,000 hits, and a lot of them are discussions of the issue. A quick check of the first few shows a lot of explanation of the difference, and why the scientific world generally doesn't make a fuss over it unless you're using the italicized (sub)species name. This may be partly because archaeologists and paleoanthropologists tend to be familiar with the German language, and consider standard German spelling to be correct in an obvious sense.
-- a gene that causes your saliva to dissolve clam shells is great if you are a starfish. If the same trait arose in a clam, however, it would likely not be passed on.
That's a great example! I hope you don't mind if I steal it. (And I'll bet that others will, too.)
The American Museum of Natural History has a now decades-old depiction of a Neanderthal in a suit & tie as part of an exhibit debunking the old popular-science depiction of Neanderthals as unsavoury brutes.
True. And it's common to repeat the suggestion that if you were to bring a Neandertal person forward in your time machine, give him a shave and a haircut, dress him up in modern clothes, and set him down in a city street anywhere in modern Europe, nobody would notice anything at all odd about his looks.
(I don't recall who first wrote that. Anyone know?)
Pretty much all the Neandertal physical features are within the "normal" range for modern Europeans, though some features would be considered outliers. This says nothing about relatedness, of course, because those features could just be what's adaptive in the European climate, and could have evolved multiple times in groups that migrated to Europe.
They did have somewhat larger brains on average, but this goes along with their general larger "robust" stature. We have lots of people of similar size in the British Isles and Scandinavia, because large size is adaptive in a cold climate.
The whole topic is a primary example of over-generalization from far too little data.
"Recently I read that there appears to be no purely Neanderthal genes in the genome of modern man tends to discount the interbreeding theories."
It's a long way from discounting them, because we don't actually know whether any of the genes from the European hunter-gatherer cultures who lived alongside Neanderthals for thousands of years exist in modern humans.
Well, I was hoping someone would point this out.
We should also be emphasizing that the studies that purportedly "prove" that the Neandertal and Cro Magnon populations didn't interbreed were only looking at mitochondrial DNA (mtDNA), for which we have a few Neandertal samples. This is non-nuclear DNA that is inherited only through the female line, and is much less than 1% of our total DNA. Nothing at all is known about Neandertal genes in the other 99% of the genes in our cells' nuclei.
Drawing such conclusions from mtDNA alone is an egregious misinterpretation of the data. This DNA says little or nothing about who our ancestors could have been. For all we know, we could have several genes on chromosomes 3, 9 and 13 that derive from Neandertal ancestors. The mtDNA studies couldn't possibly have said anything about such possibilities.
For all we know, our Y chromosomes could be entirely of Neandertal origin. Imagine a small, slight Cro Magnon chick swooning at the sight of a big, hulking Neandertal guy. He would have passed on no mtDNA genes to their offspring, of course, so the mtDNA studies would see no sign of him at all.
In any case, anyone drawing general conclusions from mtDNA data should be simply dismissed as too ignorant of how inheritance works to be part of any serious conversation on the topic.
(And I don't think the scientist who did those studies made a general claim that we have no Neandertal ancestors. They said that our maternal lines are unlikely to be Neandertal. But at a distance of 30,000 years or more, the maternal - and paternal - lines are a tiny portion of your ancestry, and we have no good clues about the rest of the tree.)
(And I was duly amused to see firefox's spell checker suggest "Netherlander" as the correct spelling for "Neandertal".;-)
How long until there's a terrorist named Robert'); DROP DATABASE; --?"
I have a friend from Southeast Asia whose name, when you zero out the high-order bits of the usual UTF-8 encoding, comes out to just that. It'll be fun seeing what happens the next time he flies home to visit his family.
Slashdot Mirror
Stories that support the concept that AT&T does not have the consumer's best interest in mind will of course get coverage, in fact, a disproportionate amount as compared to other providers or stories that reflect well on AT&T.
Such negative stories always get much more press everywhere than feel-good stories about how wonderful a company is.
Stories like this really helped block the initial adoption of DSL. When it first came out, there was a flood of stories about sleazy salesmen selling customers what they thought was a $50/month or so DSL plan. Then the customers got their first bill, often for several thousand dollars. The phone companies mostly replied "Hey, they signed the contract; they owe us the money." The word got out, and sales dragged for years. Finally, the telcos grudgingly started telling people the actual cost of the service, and sales slowly increased.
We had DSL in our house for several years, but not through Verizon, for exactly this sort of reason. After seeing lots of reports of how honest speakeasy was, and no reports of their pulling sleazy tricks like this, we finally decided to use them. We only switched to a cable company recently when we needed higher speed (for "telecommuting") than what speakeasy could supply. Even then, we did a lot of careful checking to verify that RCN's "home business" plan was actually what they said it was, and didn't have hidden charges or restrictions, just $X/month for "$Y up $Z down" service.
The telcos have a nasty history of stories like this. They've played such games with every new technology that comes along. Look at the early days of cell phones for lots of examples. The only way to fight it, really, is to publicise such stories to the max. If we can do enough damage to their sales by such publicity, they eventually get the hint and start acting a bit more honest.
I do take issue with JavaScript injection that amounts to a man-in-the-middle attack http://www.theregister.co.uk/2008/06/23/topolski_takes_on_nebuad/
This is just one more data point explaining why, ever since client-side scripting was first introduced into browsers, those of us who understand the Web have done most of our browsing with scripting turned off. If you permit strangers to download and run code on your machine, you're just inviting them to take advantage of you like this. And such injection attacks demonstrate that the dangers lie not just in the sites that you visit, but also with any machines along the path of your packets. Naturally, this would be no surprise to any network programmer.
Forging reset packets does not equal "throttling", ...
Actually, "throttle" has been used for at least a century as a euphemism for "kill", at least among the criminal elements here in the US. So, if you view Comcast as part of the organized-crime community, their referring to RST packets as "throttling" is right in line with standard criminal jargon.
Are you saying that there are secret regulations not available to the public that they must still follow?
Yup; google for "secret law" to read about it. Here's an article about one recently-passed US secret law that might affect a few /. readers.
The wikipedia article on secret law is short, but has a few good links to other informative articles.
The old saw that "Ignorance of the law is no excuse" no longer applies in the US; it's more like "Ignorance of the law is legally enforced". We have laws on the books that you must obey, but you can't read. Unauthorized knowledge of the text of the secret laws is itself a crime, as is informing anyone else of the laws' text.
There have recently been reports of judges recusing themselves from a case on the grounds that they didn't have clearance to read the applicable laws.
Another interesting case, if you can find the history from a decade ago, is the newton.com domain. It was owned by Mark Newton for years, and used by his small software shop, until Apple came out with the Newton and wanted the domain. Apple didn't negotiate; they just gave him the choices of giving up the domain or being bankrupted by the lawyers that Apple would sic on him. There were a few stories recently about Apple renewing the registration for newton.com, although they no longer actually use it for anything. It now just redirects to apple.com.
So companies can be nice about it, or extremely nasty, or anything in between. I wonder if the US has any more legal protection from such things now than it did 10 or 15 years ago. Probably not; the big corporations can always bankrupt the little independents.
Hey, one of the fonts I've had installed for years is the Group Sex font. Talk about cutesy ...
I've used it for headers in a few of my pages, and the people who have the font have told me how much they liked it.
It does have one problem: It doesn't distinguish case. But for headers, that's OK.
I'll show you. I'm going to make tiny .png's of each letter and lay them out as images.
Don't laugh too hard. When they first tried to get onto the Internet, that's more or less what the folks in Japan, Korea and China had to do. There are still Asian web sites that work that way. Recently, there has been a slow transition to UTF-8, but 10 years ago, hardly any software supported it. And it's still how a lot of printing software works, since most postscript and PDF printers still only use 8-bit characters internally.
Per-char images also have an advantage that they can be made big enough to legibly contain some of the more complex CJK characters, so they don't turn into a blur when the font size is too small. You can string a bunch of them on a line, and programs that handle images will space the lines to match the tallest chars; even when UTF-8 "works", you often get the chars drawn too small to read.
We still have a long way to go before we support all the world's languages sensibly. It shouldn't be so difficult, of course, since it's just a bunch of glyphs indexed by an integer. But when the organizations building the software are headed by people who are contemptuous of all languages but their own, it can take a very long time to do even the simplest things right.
(Well, except for Arabic-derived scripts, which are the worst case for everything. ;-)
You're doing it wrong. Both yogh and wynn have unicode code points.
Yup, and they look fine in my firefox window.
So on to the next question: How do I get such non-Latin1 chars printed correctly on a postscript printer? Anyone know of a general solution?
We have a collection of different printers here, and also several OSs (linux, OSX, Windows XP and Vista). All of them tend to produce Latin1 gibberish (aka mojibake) for UTF-8 chars. I've been looking for ways to get text in languages like Arabic and Mandarin printed correctly, so far without notable success. I'd also like to print some Old English at times, too. Anyone know how to do it right?
I think the majority of responses will be: "Why do I need all these flashy fonts on the web anyway! I have my browser show every website in Courier 10, ..."
Heh. Do you also force a white-on-green or orange-on-black color scheme?
Actually, those who have minimally explored their browsers' capabilities have noticed that the browsers automatically fall back to a font that they have. (Whether this is a majority of Web users, we really don't know.) After all, you really can't rely on your machine having any particular font. The only sensible thing to do in that case is to pick some other font that your machine does have, and that's what most software does.
What we really should be doing is pointing out publicly that there is no reason whatsoever that anyone should ever install any specific font that some vendor supplies. You should just install the fonts that you like, and which you can read easily. Anything else is a waste of disk space and memory. Those people who insist that their font is the only one to use are violating the basic design goal of the Web, which was that a "page" should be sensibly viewable on whatever output device the user has, and will be formatted to fit that device. And if the user is blind or visually impaired, at least one of the "fonts" should be auditory (for those with hearing) or Braille-like. The web designer's desires don't matter, and neither do the vendor's; all that matters is what works well for the end user.
Pages that don't work that way are simply user-hostile, i.e., wrong.
(I rarely specify a font in any of my web pages. That makes the pages smaller, so they download faster, and they appear in the user's favorite font without any wasted cpu time. The only exceptions are when a boss explicitly orders use of a font, and I like to point out that many users will just override it or won't have the font. But then, I like to be nice to the visually handicapped, even if people are always complaining that my screen's fonts are too small for them to read. ;-)
...but eventually, all the 31337 hax0rs get caught.
And how exactly do we know this? ;-)
Hmmm ... I tried that example on my Mac running 10.4.11 and perl 5.8.6, and the two sort times differed only by 1 in the third decimal place.
Maybe I should update the perl on that machine and try it again.
There isn't anything wrong with preferring fast incorrect results over slow correct results, but most people probably want slow and correct to be the default if given the choice.
Well, I'd be a bit careful about making such general statements. There is evidence that people aren't generally that intelligent.
I remember back in the 1970s, when I was at a large university that shall remain unnamed, and a bunch of CS people did a detailed study of the Fortran that accounted for fully half the runs on the campus's central mainframe (which shall also remain unnamed). They found that fully half the runs produced at least some incorrect output due to undetected integer overflows. The hardware gave interrupts for floating-point overflows, but for integers, it just set a flag bit, and you needed to test that flag to catch overflows. The compiler had an option to generate such tests, but it was off by default. The vendor said they did this because they had found that most customers preferred faster code.
The local gang didn't believe this, so they did a bit of a survey. They asked lots of users of the Fortran code whether they would prefer their programs to catch all arithmetic errors if this meant that the code ran slower, or if they would prefer faster code that sometimes didn't catch errors. Roughly 90% of the people they asked this said that they'd want the faster code. Later on, I ran across references to similar tests at other schools, with similar results.
Personally, I was shocked by this. This mainframe was used to do the computing for most of the scientific work on campus, and scientific computing was almost entirely done in Fortran. So half their data runs had undetected incorrect output. They now knew this, and they still preferred the faster speed to correct output.
Somehow, I suspect that this situation hasn't changed. I've dug into various programming languages since then, to learn how they handle this and other potential sources of erroneous results. Most current languages still ignore things like overflow flags by default. Some have no way of enabling the tests of such flags.
Yes, I know lots of ways of explicitly testing for such errors myself. I've done it a lot, because I know I can't rely on others to enable the builtin tests (when they exist) when they recompile the code. But when looking at other people's code, I almost never see anything that will detect overflows. When you're N levels deep in function calls, you usually have no way of verifying the possible range of the current function's args, so there's no way of proving that an overflow can't happen.
Sometimes I'm amazed that our systems run as well as they do, given this sort of nonchalant attitude towards known sources of hardware errors. And I do a lot of paranoid, defensive programming, even though I know that my employers probably don't want it because it slows down the software.
Well, I'm anything but a hardcore Perl hacker -- just use it to pragmatically list some rubbish now and then -- and I've never even heard of compiling your own Perl.
You certainly don't qualify as a perl hacker! NTTAWWT. ;-)
From the very beginning, the primary (and recommended) way to get perl has been to download the source and compile it yourself. It's true that most unix/linux distros have included perl for a decade or so, and of course it's usually not the current version. But this is only a minor time saver during installation. I've often upgraded to the current perl while installing systems, and I've always found it easy.
The idea that a perl user wouldn't even have heard about the easy availability of the source is sorta surprising. That's the way you're supposed to get it, after all. The standard textbooks start off telling you where to get it (or at least they did the last time I looked at one, which has been a while ;-).
OTOH, perl does have a bit of a reputation for being solid and without any problems, so it's easy to see how someone might be lazy and just use whatever the vendor supplies. I wonder what went wrong with the RH release?
However, the fact that they've published is stopping anybody else than the authors (like a patent troll) to file for a patent, right?
Nope; that doesn't necessarily stop anyone from filing. All you have to do is fill out the papers and pay the application fee. Of course, if the Patent Office discovers that someone else published it earlier, they'll reject your application. But you can certainly file. If your lawyers can write it with terminology that's sufficiently obscure, the Patent Office's examiners might not notice that it's a dup of an earlier patent.
The really problematical aspect is that in the US, if you have sufficiently many millions of dollars to spend on court costs, even if your patent will be eventually rejected, you can use the legal system to block others (including the original inventors) from using the patent for the decade or so that it takes you to lose. If you're fighting a small start up, you can probably bankrupt them before you lose. And you stand a good chance of being able to buy up their rights during their bankruptcy proceedings.
The only real defense they have against this is if they have filed their own patent application before you did (and your lawyers didn't do a good enough job of obscuring your application). Then the Patent Office will probably notice what you're doing, and will deny your application very quickly.
So has the team applied for a patent? We wouldn't want just any ISP to be able to use this algorithm, would we? And if they don't patent it, one of the many patent-troll companies will, denying the researchers the right to use the results of their own work.
Funny thing: I checked with a few of the usual map sites, and found both "Neander Tal" and "Neandertal". One map (I forget which one) had both at different points. Another actually had the obsolete 'h'; I wonder if that was a subtle joke on the part of some obscure map worker. Anyhow, I'm not too surprised that you'd see such inconsistency. Despite their stereotype, Germans tend to be nearly as sloppy about spelling as the rest of the world. And the maps may not have been made by Germans.
I also read a few more of google's hits for "Neanderthal Neandertal", and found comments by a fair number of self-described paleoanthropologists and archaeologists. My impression is that they mostly consider it a somewhat silly discussion, sorta like how most astronomers view the "raging debate" over whether Pluto is a planet. Several admit to mixing the two spellings intentionally, just to keep the pot stirred. They do mostly seem to understand that the taxonomic name has the obsolete 'h', while the German place name doesn't. The main comments can probably be summarized as "Yeah, so?" I like to omit the 'h', on the grounds that it saves typing time and bandwidth. Of course, I've thrown away all those savings with a couple of posts here.
But it's always fun to have pseudo-debates on such inconsequentiae on forums like /.. (And do I really need that extra dot on the end of that sentence? ;-)
One of my favorite t-shirts is the one that reads "Does anal retentive have a hyphen?" I like to point out that it really should have quotes around the "anal retentive".
With all the messages here so far, I'm surprised that nobody has posted the obvious auto analogy:
It's like an auto ad that says the car can drive anywhere on the public street system. So you buy it, and discover that it has a sensor that determines the road-surface type, and the engine turns off if you try to drive on any surface except asphalt. The car company excuses their misleading ad by saying that nobody needs to drive on concrete or (God forbid) gravel. You should be able to get anywhere you need to go on asphalt.
Now aren't you ashamed of not posting this analogy first?
That's great and all if you are an internet mechanic. But what if you just want to drive the damn car? For those people, who are the majority, those messages don't mean squat.
Y'know, this is one of the cases where the standard automotive analogy is a fairly good one.
There are a lot of people in the world who "just want to drive the damn car" and can't be bothered with safety. That's why we have so many thousands of traffic fatalities every year. It's also why most of the world has implemented at least minimal driver education before getting a driver's license. But this isn't done very well, and we still have too many traffic fatalities.
With autos, our century or so of experience has led to the legal requirement of a number of safety enhancements to the equipment. Maybe in another century or so, we'll also have laws mandating the use of various security enhancement to traffic on the public Internet.
Nah; it'll never happen. As with cars, we'll get token "security theater" laws that add only minimal real security, and mostly just produce a bogus safe feeling in the minds of non-technical users.
And if all your data is going through an MitM, in what sense is the initial handshake that establishes your connection secure?
And we might add that most of us are always going through a MitM [Man in the Middle], which we call an ISP.
I was tempted to add a smiley to that, but decided that it would be misleading, because there's really not much of a joke there. We recently read of Comcast getting into a bit of hot water because they were truly acting as a MitM, faking messages between the ends of a TCP connect to get them to believe that the other end had announced an end to the connection. They called it "traffic management", but it was pure classical MitM attacks.
We're now reading of not-too-secret discussions of the implementation of "deep packet inspection" by ISPs, to collect data on what's going over a connection. This is being implemented both for marketing purposes and to share the data with interested government agencies. Again, this is about as MitM as you can get.
The whole "Net Neutrality" issue is also an acknowledgement that ISPs can and do act as a MitM with the goal of controlling and sometimes interfering with their customers' data.
It might be a good idea if we were to rephrase our discussions so as to acknowledge that for most of the Internet, there's always a MitM, usually several of them, and any of them is likely to both examine and alter the data in transit. Anyone who thinks of a MitM as a sporadic, low-probability intruder is stuck back in the 1980s, and is ignoring the nature of the modern commercial Internet.
Maybe we could subtly start replacing "MitM" with "ISP" in our discussions. That might get the idea across to readers that this isn't a purely hypothetical topic.
It's actually "Neanderthal", ...
Well, actually (;-), I've read statements from a number of scientific organizations that have given their official OK to both "Neandertal" and "Neanderthal". (And MS's spell checker is hardly an authority for either scientific or German spelling. ;-)
The difference dates back to the official spelling reforms in German in the late 1800s and early 1900s. One of the many changes was to eliminate most unpronounced letters, including the 'h' in "th" combinations. So the noun "Thal" (valley) became "Tal" in standard German.
But, of course, the scientific name of the Neandert[h]al (sub)species had been established decades earlier, using the old spelling. Also, Germans often like to use what are now deprecated spellings like "Thal", much like we do in English with words like "ye olde" instead of "the old".
The official taxonomic spelling remains "Neanderthal". Taxonomic names almost always keep the same spelling even when they're known to be incorrect, because in scientific writing, the exact sequence of letters is important and must remain stable. But you often see "Neandertal" and "neandertal" used in informal scientific writing, often in the same paragraph as the taxonomic spelling with the 'h'. Some writers like to omit the 'h' in English text to encourage the correct pronunciation, since to German-speaking ears (;-) the English "th" sounds distinctly odd. ("Why do Americans lisp like that?") The folks in the Neander Valley don't use that weird English phoneme, and many English-speaking scientists pronounce "Neanderthal" in the German fashion with a /t/ phoneme.
And if you dig around, you'll find that the appropriate scientific societies either don't care, or officially consider the modern German spelling acceptable anywhere but in the (sub)species' official name. And if you're writing about the valley where the first skeleton was found, the correct spelling is "Neantertal" (unless you're in marketing and are aiming for an olde-tyme effect ;-).
If you google for "Neandertal Neanderthal" (in either order), you'll find around 74,000 hits, and a lot of them are discussions of the issue. A quick check of the first few shows a lot of explanation of the difference, and why the scientific world generally doesn't make a fuss over it unless you're using the italicized (sub)species name. This may be partly because archaeologists and paleoanthropologists tend to be familiar with the German language, and consider standard German spelling to be correct in an obvious sense.
-- a gene that causes your saliva to dissolve clam shells is great if you are a starfish. If the same trait arose in a clam, however, it would likely not be passed on.
That's a great example! I hope you don't mind if I steal it. (And I'll bet that others will, too.)
The American Museum of Natural History has a now decades-old depiction of a Neanderthal in a suit & tie as part of an exhibit debunking the old popular-science depiction of Neanderthals as unsavoury brutes.
True. And it's common to repeat the suggestion that if you were to bring a Neandertal person forward in your time machine, give him a shave and a haircut, dress him up in modern clothes, and set him down in a city street anywhere in modern Europe, nobody would notice anything at all odd about his looks.
(I don't recall who first wrote that. Anyone know?)
Pretty much all the Neandertal physical features are within the "normal" range for modern Europeans, though some features would be considered outliers. This says nothing about relatedness, of course, because those features could just be what's adaptive in the European climate, and could have evolved multiple times in groups that migrated to Europe.
They did have somewhat larger brains on average, but this goes along with their general larger "robust" stature. We have lots of people of similar size in the British Isles and Scandinavia, because large size is adaptive in a cold climate.
The whole topic is a primary example of over-generalization from far too little data.
"Recently I read that there appears to be no purely Neanderthal genes in the genome of modern man tends to discount the interbreeding theories."
It's a long way from discounting them, because we don't actually know whether any of the genes from the European hunter-gatherer cultures who lived alongside Neanderthals for thousands of years exist in modern humans.
Well, I was hoping someone would point this out.
We should also be emphasizing that the studies that purportedly "prove" that the Neandertal and Cro Magnon populations didn't interbreed were only looking at mitochondrial DNA (mtDNA), for which we have a few Neandertal samples. This is non-nuclear DNA that is inherited only through the female line, and is much less than 1% of our total DNA. Nothing at all is known about Neandertal genes in the other 99% of the genes in our cells' nuclei.
Drawing such conclusions from mtDNA alone is an egregious misinterpretation of the data. This DNA says little or nothing about who our ancestors could have been. For all we know, we could have several genes on chromosomes 3, 9 and 13 that derive from Neandertal ancestors. The mtDNA studies couldn't possibly have said anything about such possibilities.
For all we know, our Y chromosomes could be entirely of Neandertal origin. Imagine a small, slight Cro Magnon chick swooning at the sight of a big, hulking Neandertal guy. He would have passed on no mtDNA genes to their offspring, of course, so the mtDNA studies would see no sign of him at all.
In any case, anyone drawing general conclusions from mtDNA data should be simply dismissed as too ignorant of how inheritance works to be part of any serious conversation on the topic.
(And I don't think the scientist who did those studies made a general claim that we have no Neandertal ancestors. They said that our maternal lines are unlikely to be Neandertal. But at a distance of 30,000 years or more, the maternal - and paternal - lines are a tiny portion of your ancestry, and we have no good clues about the rest of the tree.)
(And I was duly amused to see firefox's spell checker suggest "Netherlander" as the correct spelling for "Neandertal". ;-)
How long until there's a terrorist named Robert'); DROP DATABASE; --?"
I have a friend from Southeast Asia whose name, when you zero out the high-order bits of the usual UTF-8 encoding, comes out to just that. It'll be fun seeing what happens the next time he flies home to visit his family.