Oh, not to mention the positive effects that bootlegging has had on sales. People get a movie or two, a track or two from an artist or studio, like it, and then go find out what else has been done and often buy; very often!
Probably true, but not the whole story. Remember that a couple of years ago, Eminem's latest album was "pre-released" on the file-sharing sites several months before the actual release, and it was reported that there were some 5 million downloads. Then the album went on to become the best-selling album worldwide that year. Eminem joked that next time he'd pre-release it himself on the file-sharing sites. Or maybe he wasn't joking at all.
In this case, you can't argue that he needed the publicity or that people needed to know what his stuff was like. He was quite well known already. His fans pretty much knew what to expect, and they weren't disappointed. So the file sharers probably didn't really add publicity. It was more like people downloaded the album, and sent a track or two off to friends and acquaintances saying "Hey, listen to this!" Then lots of those people bought the album, after having been primed by the one or two tracks they'd heard (plus maybe being told that the rest was just as good).
In any case, given the top-seller status of the album, you can't make a very strong argument that the file sharers hurt sales. And his reputation means that you can't make a very strong argument that many new listeners were added by the file sharing. The biggest effect was probably the laughter (and warm fuzzies) when people heard his remark about the effect of the file sharing.
It might be interesting if we could find out how many people pre-ordered the album after hearing a "pirated" track. But I don't expect that the industry collects this sort of data.
At the level of nations, actual money means very little... [it's all just printed anyway, and everyone knows that].
Not to be overly picky, but this hasn't been true for several decades. Most of the money I've ever had has never been printed. It has existed only in electronic form. My paycheck (and my wife's) gets deposited electronically. I write a check or use a credit/debit card, and the money is transferred between banks in the form of data packets, with no physical money ever being involved.
So far, I haven't heard of much analysis of this fact by economists. Yes, it does get mentioned occasionally. But I'd think this should be making a lot of people nervous. After all, money can now be created or destroyed by editing an entry in a database. A bank can loan you money by simply adding a number to the balance of one of your accounts. If they're honest, they'll also subtract the same number from some other account, but we've been learning that the people running many banks are far from honest.
Of course, all this does sorta reinforce your main point, that actual money means very little. It's just ones and zeroes on some computer's disk drive. Some computer owned by someone other than you. Regulated by regulators that more and more don't believe in regulation or proper accounting practices.
Heh; I learned prolog when I was in college, several decades back. Since then, I keep running into situations where I think "This would be so much easier if I could just write it in prolog." But sadly, that's against the guidelines in most of the corporate world, which is still trying to adjust to the advent of things like C, perl, and python, and the fact that their beloved Cobol has been eclipsed. Actual powerful languages like prolog (or even lisp) would be just too much of a shock to The Way Things Are Done. One quickly learns that it doesn't help your career to suggest them.
But I don't think that even prolog programs could run backwards, or finish before they started, or whatever "118% slower" means. The obvious interpretation of this number would be that if, for example, one system ran a program at 100 mips, the other would run it at -18 mips. If we use the standard auto analogy, it's easy to understand this: One is going forward at 100 km/s, while the other is going in reverse at 18 km/s. But I'm a bit puzzled about how you could build a CPU with a "reverse gear".
You can't really fault them for wanting to keep on eye on people like Geraldo Rivera, can you?
Well, that discredits them right there. I stopped watching him years ago, and I'd question the intelligence of anyone who's keeping an eye on him now. Similarly for a long list of other pseudo-journalists.
Tom Tomorrow asks what other revelations about the Bush administration are likely to follow. Anyone have any ideas?
Well, one thing we can be pretty sure about is that, unless they come out with evidence of George W's sexual activity with staffers, nobody but a few wonks will care and there will be no repercussions.
So far, all the signs are that the Bush Gang has got away with it all, and nobody will ever be brought to account in the courts.
We might start thinking about what sort of precedent this sets for subsequent American administrations.
So the root of their complaint is that they want Google to pull in non-subscribers for them, indexing pages are not available to the browser unless they pay.
And, of course, google really can't do this unless the Googlebots can scan Britannica's pages and index them. Maybe someone should tell them that unless they let in the Googlebots, google will never send people to them.
So what they should do is check every incoming link for the ID string "Googlebot", and for those, return actual articles rather than flash ads. This will allow google to index their content and send us to Britannica. Then, since our browsers don't call themselves "Googlebot", we'll get the flash junk instead of the articles.
But it would end their complaints that google doesn't send people to them. Instead, they'd get a lot of visitors, who would be pissed off that they can't find the content that they were looking for, but just get flash junk instead.
The software developers will quickly undo all the speed advances that should result from multi-core CPUs. Software has a much shorter development time than hardware, so all the advantage in this contest is with the software.
Yeah, I was sorta wondering about that. Anyone know what the denominator could have been in this calculation? Are they really claiming that it runs the code backwards, undoing the calculations and going from a programs outputs to what its inputs had to be? If so, that could be a major technological advance all by itself. Imagine the useful things you could do with this capability...
Hmmm... I walked through the Barclays demo pages, and one thing I noticed was that the URL always started with "http://". So what's to prevent my ISP or anyone else along the data path from extracting all the data from the packets and adding it to their database? In particular, I noticed that the protocol involved typing in the recipient's account number and name, which could be useful data to anyone watching the conversation.
Why would I buy this thing and not and iPod Touch?
Maybe because of the bigger screen with lots more pixels, so you can see more at a time or not need to scroll back and forth so much.
OTOH, if the amount of stuff on screen isn't that important to you and fitting in your shirt pocket is important, then you wouldn't want this thing at all.
So which features of the iPod Touch do you rate as important? Make a list, and try to estimate which gadget does each of them better.
And be prepared for the marketers trying to obfuscate this as much as possible. For instance, I've found it quite impressive how difficult they make it to discover the actual pixel counts for various screens. Sometimes you can't even find this on the "specs" page.
Do the Unicode characters show up correctly when you print to PDF?
I don't know how to determine that.
I can say that when I send the PDF files to a printer, the non-ASCII chars come out as Latin1 gibberish. But this doesn't mean that they're wrong inside the PDF. Since PDF files are (usually) compressed, I can't examine the representation of the characters by any method that I know to determine whether they're PDF or not.
However, I have tested this for PS (PostScript) files. A number of tools that I have produce PS, and I get PDF by running/usr/bin/pstopdf. PS files are plain text, so I can edit them. I've used both the Edit program and vi inside a Terminal to examine PS files produced from UTF-8 text, and the non-ASCII characters (Chinese, Russian, Hebrew, Arabic, whatever) all show up on the screen correctly. With vi, I can cut out a line of text from the PS file that contains non-ASCII chars, feed them to any of several hexdump programs, and verify that they are encoded in UTF-8. But when I feed the PS files to Preview or lpr, those lines come out on the screen or paper as Latin1 gibberish.
So it's the rendering code in Preview or whatever lpr calls (or whatever is invoked by CMD-P or the File..Print menu items) that is turning UTF-8 text into Latin1 gibberish.
I'd guess that pstopdf also puts UTF-8 chars into its PDF output, but I know no way to test this. Is there a tool that will uncompress a PDF file, extract a line and output it unchanged, so I can feed it to a hexdump program? I don't know of anything that can do this, so I can't say anything about what's inside a PDF file. All I can say is that apparently only Latin1 encodings work with tools like Preview or lpr on the Macs that I can use for testing.
I'm willing to believe that I've screwed up something in the configuration of all these Macs. But I wouldn't know how to do it if I wanted to, and I have no idea what I might have done wrong. And it's curious that this happens on all of them, including a relatively new iMac (belonging to my wife) on which I've done very little work of any sort. Mostly I've just looked over her shoulder as she does things, and occasionally make suggestions to fix problems.
A funny thing on her Mac is that she has a virtual Windows XP installed (which lives on a VPN connection to her office network). From inside XP, she can print Arabic texts and they mostly come out fine on paper. I checked, and the ones I examined were UTF-8. But when she copies them to the OS X system and tries to display or print them, the Arabic often comes out as gibberish. Sometimes it's Latin1 gibberish; sometimes it is Arabic characters drawn left-to-right. Sometimes the order is correct, but they're all the "isolated" variants, which is hardly readable. How the Arabic is messed up seems to depend on the app, though, so we're assuming that different Mac apps use different routines to render Arabic, and some of the library routines are better than others.
Chinese is different, though. It usually displays fine on her new iMac, but it comes out as Latin1 gibberish on paper. This is using the same printer on the same machine where the XP system usually prints Arabic and Chinese correctly. It's sorta embarrasing for those of us who like Macs better than Microsoft junk.;-)
Well, I'm sad that UTF-8 doesn't work too well in linux, but in my experience, OS X is nearly as bad. And there's no obvious pattern to what works and what doesn't.
Where is the text coming from?
From wherever text comes from. Some I've typed in myself. Some comes out of email. Some comes from web sites. Some is produced by programs.
What format is the text in if it is coming from a file (plain, doc, tex etc)?
All of the above. Mostly plain text. Some PostScript (PS), which is of course plain text, and is an interesting case. I have some PS files which contain non-English text in UTF-8 form. If I open them with the vi editor in a Terminal window on my Mac, the Chinese or Japanese or whatever looks fine. I don't have to use any magical incantation; vi just writes the text to the screen and the glyphs appear as they should (after increasing the font size a bit for some of them;-). But when I feed the same PS file to Preview, for example, the Chinese or Russian or Hebrew text shows up on the screen in "mojibake" form, with latin1 gibberish instead of the characters that vi and Terminal display. I can also do "cat <file>", and the non-ASCII chars look correct. The command "more <file>" also works correctly. If I print the files with lpr, I get Latin1 gibberish instead of Chinese or Hebrew.
PDF files that contain non-ascii utf-8 chars almost always appear on both the screen and paper as gibberish, no matter how I print them. I can't verify that the PDF is correct, because I know of no way of viewing the internals. But with PS, I can verify that the non-ascci utf-8 text is correct, because I can view it with cat, more, less, vi, etc, and the text always renders correctly. I can also cut out a line, feed it to any of several hex-dump programs, and verify that the non-ascii chars are UTF-8 encodings. But Preview and several printers all render the text as gibberish.
Do you know how the text is being sent to the printer...
Nope; lpr, CMD-P, and the various File..Print thingies are all black magic. I have no idea how they work or what they do. But they never get it right on any of the Macs that I have available (which admittedly isn't very many). Here at home, we have 2 Macs, 2 linux systems, 2 Windows systems, and 3 printers, which I've hooked up to various computers in various configurations to try to solve the problems. None of them has succeeded for Chinese or Hebrew, or even Russian (which should be easy;-). My wife has had some success getting Arabic printed from one Windows system, but even then, she is always grumbling about things that come out garbaged in some way when they look good on the screen.
One could argue that the RIAA is a lot worse than the Communists were. Consider the Soviet Union: The Communists were voted out of power despite all the vote rigging that the party in power always does everywhere, and they accepted the vote. This wasn't a fluke; it has happened in most of the countries that had Communist governments. The RIAA and MPAA were never elected to their positions, and we can't vote them out. They'll be around longer than any Communist government, and there's not a whole lot we can do about it. They are private corporations created by other private corporations to "coordinate" their business so that no real market could develop. Judicious bribery, uh, I mean campaign contributions, led to the draconian copyright laws that helped prevent a market in music or movies.
Maybe the move to the Internet will end this whole centrally-controlled system. Or maybe they'll find a way to bring it under control. Stay tuned...
I clicked on that link, read through it, but didn't see anything about printing. So I hit CMD-F, typed in the search string "print", and there was one such string - the little print icon at the top of the article to print the page. There was no instance of "print" in the article. I tried a few of the links toward the bottom, but none of them seems to have any mention of printing from OS X, either.
So where is there actual documentation on how one makes a Mac print UTF-8 text correctly? Just saying that OS X handles Unicode doesn't answer this. Yes, I've seen it correctly generate non-ASCII file names. I have several directories that contain files with Chinese names (traditional chars, simplified chars, and pinyin with tone marks). But I've never got a correct printout of these directories' contents.
One bit of weirdness: the ls command prints Chinese, Hebrew and Russian filenames as "???????.txt", etc. However, if I type ls|cat, the names come out in Chinese, Hebrew and Russian. This proves that the Terminal window can correctly display those characters. But ls for some reason garbages them.
This might be explained by the "legacy" nature of the ls command (though the problem could probably be fixed by just commenting out a few lines of code in ls.c). But it doesn't explain why I keep seeing non-ASCII text turned to Latin1 gibberish all over the place. And so far, it has always happened with everything I've tried to print, on several different Macs, and with several different printers, including printers bought from an Apple store.
They were probably just ignoring you because you hadn't even tried to print a UTF-8 encoded document. If you had, you'd have found It Just Works.
Wrong. I've tried on several Macs, with several printers. In every case, when I've tried to print UTF-8 text (usually Chinese), it comes out as the usual "mojibake" Latin1 gibberish. This is using methods like CMD-P or the File..Print menu items with various windows that display the non-ASCII chars (e.g. Chinese characters) just fine. Every once in a while, there have been a few non-ASCII chars that print correctly, but usually all of them come out on paper as gibberish, even when they look good on the screen.
I've asked in various online forums, and the only replies I've gotten are either like yours, insulting me by telling me that it works without giving any clues as to how to make it work, or telling me that it misbehaves the same way on their Mac and if I figure out how to make it Just Work, I should tell them, too.
Earlier today, I accidentally typed CMD-P on this Mac in a Terminal window displaying a message that I wanted a hard copy of. Then, as it printed, I said "Duh!" as I realized my mistake. Half the message was in Hebrew, and I knew how it would come out. Sure enough, the Hebrew part was Latin1 gibberish, not Hebrew.
... I wouldn't be surprised if they told her to go to 192.168.0.1 on Firefox in order to get her internet connection configured.
My ears perked up at that, so I decided to try it. I have an Ubuntu (heron) box handy, where I fired up FF, typed in 192.168.0.1, hit Return... and it told me "Though the site seems valid, the browser was unable to establish a connection." Hmmm... Next, I opened a terminal window, typed in "ping 192.168.0.1", and got the results:
$ ping 192.168.0.1 PING 192.168.0.1 (192.168.0.1) 56(84) bytes of data. From 192.168.1.17: icmp_seq=2 Redirect Host(New nexthop: 192.168.0.1) From 192.168.1.17: icmp_seq=3 Redirect Host(New nexthop: 192.168.0.1) From 192.168.1.17 icmp_seq=1 Destination Host Unreachable From 192.168.1.17 icmp_seq=2 Destination Host Unreachable From 192.168.1.17 icmp_seq=3 Destination Host Unreachable...
So it looks like telling her to go to 192.168.0.1 on Firefox could be rather bad advice here. As a longtime unix user, with a couple decades of internet software development, I have no idea why this is supposed to do anything useful in general. Yes, I can see how I might install a machine with address 192.168.0.1, install a web server on it, and install software in htdocs that handles network configuration. That's easy. But I didn't expect anything like that was set up on this LAN, and sure enough, it isn't. I'm not sure what's going on with all those redirections, because the gateway machine at 192.168.0.17 (don't ask;-) shouldn't to my knowledge do that. But it definitely doesn't get you any net config tool.
But "Go to 192.168.0.1 on Firefox" is clearly not a viable approach on this Ubuntu machine. Are there large numbers of LANs (or out-of-the-box Ubuntu systems) where this would actually do something that a novice would find useful?
(Oh, and "Ask your network admin" isn't useful here. My answer would be "Hey, that's me!" Followed by "I have no idea.";-)
Unfortunately Linux users have the tendency to call people idiots unless you approach them with "xyz OS desktop/OS can do this but blahblah Linux distro can't".
Actually, this has little to do with fortune or with linux. It's a fundamental feature of human psychology. The computer industry has long been plagued by it, of course, but it's much more general. The best approach is to just the above as the best approach.
Thus, back in the 1980s, when I first got onto unix systems and faced the problem known as "emacs vs. vi", I found the same thing. If I asked in any forum "How do you do task T with editor E?", the answer inevitably was of the form "WTF? RTFM, n00b!"
But I quickly learned that the proper approach was to phrase the question as a negative statement: "Hey guys, emacs can do T, but vi can't." This would almost always be answered by one or more explanations of exactly how to do it in vi. Interchange "emacs" and "vi" to learn about doing T in emacs. Sometimes they would insult me, too, but I learned to dig the facts out of the insults and go on my way. Yes, I learned the basic mechanics of both editors from smallish manuals. But the real learning was done by claiming that an editor couldn't do something, and waiting for the experts to show what a dummy I was.
I did eventually settle on vi, mostly because I found myself doing a lot of work on machines that were borked somehow, or were on the other end of a semi-sane network link that ate control characters, or whatever. I'd make the claim that "Emacs/vi can't be used unless you have a totally sane link to a totally sane system". The vi experts responded by explaining things like the "ex" interface and other tricks that made it usable in most cases. They also suggested learning "ed" and "sed" for extreme failure situations, but I'd already done that. OTOH, the emacs experts never answered such charges, telling me that emacs really wasn't usable on a damaged system. So I went with vi, while envying the folks that never had to deal with recovering a brain-damaged system or negotiating with comm gadgets that ate control characters.
So maybe our tactic should be to teach people to approach linux experts with the "Linux can't do T" sort of charge. Don't ask the linux experts how to do T; challenge them to teach a dummy like you that linux can do task T. This will also help greatly with getting info about Windows, of course, but that's probably not what most readers here want to do. I've also found that this tactic does wonders with getting Mac fanboys to loosen up, stop chanting "It Just Works!", and explain how to get the little things to do the job right.
Claiming that "System S can't do task T" is a general tactic that the public should know about when dealing with any tech experts. If we educate people about it, it might go a long way toward getting them on our side by making their computers do what they want.
My current favorite claim is that linux software can't print UTF-8-encoded text with stuff like Arabic or Chinese text. I suspect that this probably isn't true, because linux is widely used in Asia. But so far, it seems true, because about all I've found is assertions that it's possible, without any coherent explanation of how people (in Asia?) are doing it. (I've also collected a list of email addresses of people around the world saying that if I find a good answer, I should forward a copy to them.;-)
(The Mac crowd also doesn't answer this charge, either, so apparently Macs can't print UTF-8 text, either. I'd be happy to be proved an idiot here...)
Actually, I am still using an old Powerbook. We had two of them here; one died and one still lives (but is getting a bit feeble). I'll probably order a new Mac laptop of some sort within the month.
Seriously, am I the only person in the entire world who runs strings or emacs on binary files just to see what might be in them?
Yes. Yes you are.
No, he (she) isn't. The first thing I did after reading the summary was to pick up my Mac Powerbook, cd into my Music/Itunes directory, find a couple of.m4p files, and run the strings command on them. Adding a few greps to filter out the printable binary junk, I quickly found my name and email address.
As for someone writing a tool to replace them, I found that I already had one. Years ago, I wrote a little command-line app that just does a simple string substitution and writes the result to stdout. It's quite handy, and I use it all the time. I told it to copy one of the.m4p files, with my email address replaced by a fake email address of the same length. I then told iTunes to load that file - and it played fine.
Then, of course, I did the same trick, replacing my name with a different name of the same length. As I expected, iTunes popped up a little window saying that it needed to check the tune's registration, showing me the name, and asking for a password. Presumably when DRM goes away, that little window will also go away, and I'll bet that the tunes will play.
I don't think I'll bother posting the program. Any semi-competent beginning C programmer should be able to type it in under a minute. Probably most perl and python programmers can do the same, a bit faster, as could any moderately experienced emacs user. 25 years ago, when I first picked up the C bible, I wouldn't have found it a challenge after my second day with the language.
Just make sure the replacement strings have the same byte count as the old name.
In some nations this would be illegal as Word is no ISO standard. You cannotn require someone to buy a specific product. It would contravene competition laws.
Actually, it's not that simple, even if your country has such laws. Note that the school teacher and administrators in question didn't explicitly require that the student (or his parents) buy anything. What they demanded was that assignments be handed in in a specific format (Word), using a specific font (Times New Roman). These are the names of Microsoft products, but I'd guess that the name "Microsoft" wasn't mentioned by the teach or or the admins. Most of them probably didn't even know that they were demanding that the child use Microsoft products. They certainly didn't insist that the parents buy anything from Microsoft.
This means that the laws in question don't apply. Or at least, that's what the school will argue. Getting them to stop doing this will almost certainly require a court case in most jurisdictions. And this will be a huge expense for most parents. They're more likely to take the easy way out, and buy a cheap Microsoft-based system, rather than face the time and expense of taking the school system to court.
If you don't get anywhere with the teacher, you should definately ask the school board to put the topic on your agenda.
Or maybe you should point out that the teacher has required a name-brand file format and font that are proprietary, and you don't have a license for them. Suggest that if these are required, then the school should pay for your child's computer with license to use such proprietary products. Mention that if they refuse to pay for your child's computer, you know some lawyers that will help you get a court order for reimbursement for the price.
(And you really should check with any lawyer friends about the legality of a school requiring that a student bring a name-brand product to school. It'd be more fun to push for a legal precedent that the school must pay for any such proprietary material required for classwork.;-)
What happened to good old "Scientist"? It's a nice, nine letters long, and respected. "Experimentalist"... It sounds like what a social deviant might call themselves....
Of course, the more common term is "experimental scientist", as opposed to someone like Albert Einstein or Stephen Hawking, who were/are mostly known as theoretical scientists.
But "experimentalist" is a valid English word, makes sense in context, and has fewer syllables than "experimental scientist" while still emphasizing the experimental nature of their work.
Slashdot Mirror
Oh, not to mention the positive effects that bootlegging has had on sales. People get a movie or two, a track or two from an artist or studio, like it, and then go find out what else has been done and often buy; very often!
Probably true, but not the whole story. Remember that a couple of years ago, Eminem's latest album was "pre-released" on the file-sharing sites several months before the actual release, and it was reported that there were some 5 million downloads. Then the album went on to become the best-selling album worldwide that year. Eminem joked that next time he'd pre-release it himself on the file-sharing sites. Or maybe he wasn't joking at all.
In this case, you can't argue that he needed the publicity or that people needed to know what his stuff was like. He was quite well known already. His fans pretty much knew what to expect, and they weren't disappointed. So the file sharers probably didn't really add publicity. It was more like people downloaded the album, and sent a track or two off to friends and acquaintances saying "Hey, listen to this!" Then lots of those people bought the album, after having been primed by the one or two tracks they'd heard (plus maybe being told that the rest was just as good).
In any case, given the top-seller status of the album, you can't make a very strong argument that the file sharers hurt sales. And his reputation means that you can't make a very strong argument that many new listeners were added by the file sharing. The biggest effect was probably the laughter (and warm fuzzies) when people heard his remark about the effect of the file sharing.
It might be interesting if we could find out how many people pre-ordered the album after hearing a "pirated" track. But I don't expect that the industry collects this sort of data.
At the level of nations, actual money means very little... [it's all just printed anyway, and everyone knows that].
Not to be overly picky, but this hasn't been true for several decades. Most of the money I've ever had has never been printed. It has existed only in electronic form. My paycheck (and my wife's) gets deposited electronically. I write a check or use a credit/debit card, and the money is transferred between banks in the form of data packets, with no physical money ever being involved.
So far, I haven't heard of much analysis of this fact by economists. Yes, it does get mentioned occasionally. But I'd think this should be making a lot of people nervous. After all, money can now be created or destroyed by editing an entry in a database. A bank can loan you money by simply adding a number to the balance of one of your accounts. If they're honest, they'll also subtract the same number from some other account, but we've been learning that the people running many banks are far from honest.
Of course, all this does sorta reinforce your main point, that actual money means very little. It's just ones and zeroes on some computer's disk drive. Some computer owned by someone other than you. Regulated by regulators that more and more don't believe in regulation or proper accounting practices.
Heh; I learned prolog when I was in college, several decades back. Since then, I keep running into situations where I think "This would be so much easier if I could just write it in prolog." But sadly, that's against the guidelines in most of the corporate world, which is still trying to adjust to the advent of things like C, perl, and python, and the fact that their beloved Cobol has been eclipsed. Actual powerful languages like prolog (or even lisp) would be just too much of a shock to The Way Things Are Done. One quickly learns that it doesn't help your career to suggest them.
But I don't think that even prolog programs could run backwards, or finish before they started, or whatever "118% slower" means. The obvious interpretation of this number would be that if, for example, one system ran a program at 100 mips, the other would run it at -18 mips. If we use the standard auto analogy, it's easy to understand this: One is going forward at 100 km/s, while the other is going in reverse at 18 km/s. But I'm a bit puzzled about how you could build a CPU with a "reverse gear".
Maybe my imagination just isn't up to the job.
You can't really fault them for wanting to keep on eye on people like Geraldo Rivera, can you?
Well, that discredits them right there. I stopped watching him years ago, and I'd question the intelligence of anyone who's keeping an eye on him now. Similarly for a long list of other pseudo-journalists.
Tom Tomorrow asks what other revelations about the Bush administration are likely to follow. Anyone have any ideas?
Well, one thing we can be pretty sure about is that, unless they come out with evidence of George W's sexual activity with staffers, nobody but a few wonks will care and there will be no repercussions.
So far, all the signs are that the Bush Gang has got away with it all, and nobody will ever be brought to account in the courts.
We might start thinking about what sort of precedent this sets for subsequent American administrations.
So the root of their complaint is that they want Google to pull in non-subscribers for them, indexing pages are not available to the browser unless they pay.
And, of course, google really can't do this unless the Googlebots can scan Britannica's pages and index them. Maybe someone should tell them that unless they let in the Googlebots, google will never send people to them.
So what they should do is check every incoming link for the ID string "Googlebot", and for those, return actual articles rather than flash ads. This will allow google to index their content and send us to Britannica. Then, since our browsers don't call themselves "Googlebot", we'll get the flash junk instead of the articles.
But it would end their complaints that google doesn't send people to them. Instead, they'd get a lot of visitors, who would be pissed off that they can't find the content that they were looking for, but just get flash junk instead.
The software developers will quickly undo all the speed advances that should result from multi-core CPUs. Software has a much shorter development time than hardware, so all the advantage in this contest is with the software.
Yeah, I was sorta wondering about that. Anyone know what the denominator could have been in this calculation? Are they really claiming that it runs the code backwards, undoing the calculations and going from a programs outputs to what its inputs had to be? If so, that could be a major technological advance all by itself. Imagine the useful things you could do with this capability ...
Hmmm ... I walked through the Barclays demo pages, and one thing I noticed was that the URL always started with "http://". So what's to prevent my ISP or anyone else along the data path from extracting all the data from the packets and adding it to their database? In particular, I noticed that the protocol involved typing in the recipient's account number and name, which could be useful data to anyone watching the conversation.
We have a name for a mixture of water and carbon dioxide. It's called "seltzer water". With added impurities, it's sold as "soft drinks".
Mmmm ... Martian dust cola. Satisfies your body's need for hundreds of trace minerals.
Why would I buy this thing and not and iPod Touch?
Maybe because of the bigger screen with lots more pixels, so you can see more at a time or not need to scroll back and forth so much.
OTOH, if the amount of stuff on screen isn't that important to you and fitting in your shirt pocket is important, then you wouldn't want this thing at all.
So which features of the iPod Touch do you rate as important? Make a list, and try to estimate which gadget does each of them better.
And be prepared for the marketers trying to obfuscate this as much as possible. For instance, I've found it quite impressive how difficult they make it to discover the actual pixel counts for various screens. Sometimes you can't even find this on the "specs" page.
Do the Unicode characters show up correctly when you print to PDF?
I don't know how to determine that.
I can say that when I send the PDF files to a printer, the non-ASCII chars come out as Latin1 gibberish. But this doesn't mean that they're wrong inside the PDF. Since PDF files are (usually) compressed, I can't examine the representation of the characters by any method that I know to determine whether they're PDF or not.
However, I have tested this for PS (PostScript) files. A number of tools that I have produce PS, and I get PDF by running /usr/bin/pstopdf. PS files are plain text, so I can edit them. I've used both the Edit program and vi inside a Terminal to examine PS files produced from UTF-8 text, and the non-ASCII characters (Chinese, Russian, Hebrew, Arabic, whatever) all show up on the screen correctly. With vi, I can cut out a line of text from the PS file that contains non-ASCII chars, feed them to any of several hexdump programs, and verify that they are encoded in UTF-8. But when I feed the PS files to Preview or lpr, those lines come out on the screen or paper as Latin1 gibberish.
So it's the rendering code in Preview or whatever lpr calls (or whatever is invoked by CMD-P or the File..Print menu items) that is turning UTF-8 text into Latin1 gibberish.
I'd guess that pstopdf also puts UTF-8 chars into its PDF output, but I know no way to test this. Is there a tool that will uncompress a PDF file, extract a line and output it unchanged, so I can feed it to a hexdump program? I don't know of anything that can do this, so I can't say anything about what's inside a PDF file. All I can say is that apparently only Latin1 encodings work with tools like Preview or lpr on the Macs that I can use for testing.
I'm willing to believe that I've screwed up something in the configuration of all these Macs. But I wouldn't know how to do it if I wanted to, and I have no idea what I might have done wrong. And it's curious that this happens on all of them, including a relatively new iMac (belonging to my wife) on which I've done very little work of any sort. Mostly I've just looked over her shoulder as she does things, and occasionally make suggestions to fix problems.
A funny thing on her Mac is that she has a virtual Windows XP installed (which lives on a VPN connection to her office network). From inside XP, she can print Arabic texts and they mostly come out fine on paper. I checked, and the ones I examined were UTF-8. But when she copies them to the OS X system and tries to display or print them, the Arabic often comes out as gibberish. Sometimes it's Latin1 gibberish; sometimes it is Arabic characters drawn left-to-right. Sometimes the order is correct, but they're all the "isolated" variants, which is hardly readable. How the Arabic is messed up seems to depend on the app, though, so we're assuming that different Mac apps use different routines to render Arabic, and some of the library routines are better than others.
Chinese is different, though. It usually displays fine on her new iMac, but it comes out as Latin1 gibberish on paper. This is using the same printer on the same machine where the XP system usually prints Arabic and Chinese correctly. It's sorta embarrasing for those of us who like Macs better than Microsoft junk. ;-)
(Mostly, she totally loves her new iMac. )
Well, I'm sad that UTF-8 doesn't work too well in linux, but in my experience, OS X is nearly as bad. And there's no obvious pattern to what works and what doesn't.
Where is the text coming from?
From wherever text comes from. Some I've typed in myself. Some comes out of email. Some comes from web sites. Some is produced by programs.
What format is the text in if it is coming from a file (plain, doc, tex etc)?
All of the above. Mostly plain text. Some PostScript (PS), which is of course plain text, and is an interesting case. I have some PS files which contain non-English text in UTF-8 form. If I open them with the vi editor in a Terminal window on my Mac, the Chinese or Japanese or whatever looks fine. I don't have to use any magical incantation; vi just writes the text to the screen and the glyphs appear as they should (after increasing the font size a bit for some of them ;-). But when I feed the same PS file to Preview, for example, the Chinese or Russian or Hebrew text shows up on the screen in "mojibake" form, with latin1 gibberish instead of the characters that vi and Terminal display. I can also do "cat <file>", and the non-ASCII chars look correct. The command "more <file>" also works correctly. If I print the files with lpr, I get Latin1 gibberish instead of Chinese or Hebrew.
PDF files that contain non-ascii utf-8 chars almost always appear on both the screen and paper as gibberish, no matter how I print them. I can't verify that the PDF is correct, because I know of no way of viewing the internals. But with PS, I can verify that the non-ascci utf-8 text is correct, because I can view it with cat, more, less, vi, etc, and the text always renders correctly. I can also cut out a line, feed it to any of several hex-dump programs, and verify that the non-ascii chars are UTF-8 encodings. But Preview and several printers all render the text as gibberish.
Do you know how the text is being sent to the printer ...
Nope; lpr, CMD-P, and the various File..Print thingies are all black magic. I have no idea how they work or what they do. But they never get it right on any of the Macs that I have available (which admittedly isn't very many). Here at home, we have 2 Macs, 2 linux systems, 2 Windows systems, and 3 printers, which I've hooked up to various computers in various configurations to try to solve the problems. None of them has succeeded for Chinese or Hebrew, or even Russian (which should be easy ;-). My wife has had some success getting Arabic printed from one Windows system, but even then, she is always grumbling about things that come out garbaged in some way when they look good on the screen.
They are no better than communists.
One could argue that the RIAA is a lot worse than the Communists were. Consider the Soviet Union: The Communists were voted out of power despite all the vote rigging that the party in power always does everywhere, and they accepted the vote. This wasn't a fluke; it has happened in most of the countries that had Communist governments. The RIAA and MPAA were never elected to their positions, and we can't vote them out. They'll be around longer than any Communist government, and there's not a whole lot we can do about it. They are private corporations created by other private corporations to "coordinate" their business so that no real market could develop. Judicious bribery, uh, I mean campaign contributions, led to the draconian copyright laws that helped prevent a market in music or movies.
Maybe the move to the Internet will end this whole centrally-controlled system. Or maybe they'll find a way to bring it under control. Stay tuned ...
I clicked on that link, read through it, but didn't see anything about printing. So I hit CMD-F, typed in the search string "print", and there was one such string - the little print icon at the top of the article to print the page. There was no instance of "print" in the article. I tried a few of the links toward the bottom, but none of them seems to have any mention of printing from OS X, either.
So where is there actual documentation on how one makes a Mac print UTF-8 text correctly? Just saying that OS X handles Unicode doesn't answer this. Yes, I've seen it correctly generate non-ASCII file names. I have several directories that contain files with Chinese names (traditional chars, simplified chars, and pinyin with tone marks). But I've never got a correct printout of these directories' contents.
One bit of weirdness: the ls command prints Chinese, Hebrew and Russian filenames as "???????.txt", etc. However, if I type ls|cat, the names come out in Chinese, Hebrew and Russian. This proves that the Terminal window can correctly display those characters. But ls for some reason garbages them.
This might be explained by the "legacy" nature of the ls command (though the problem could probably be fixed by just commenting out a few lines of code in ls.c). But it doesn't explain why I keep seeing non-ASCII text turned to Latin1 gibberish all over the place. And so far, it has always happened with everything I've tried to print, on several different Macs, and with several different printers, including printers bought from an Apple store.
They were probably just ignoring you because you hadn't even tried to print a UTF-8 encoded document. If you had, you'd have found It Just Works.
Wrong. I've tried on several Macs, with several printers. In every case, when I've tried to print UTF-8 text (usually Chinese), it comes out as the usual "mojibake" Latin1 gibberish. This is using methods like CMD-P or the File..Print menu items with various windows that display the non-ASCII chars (e.g. Chinese characters) just fine. Every once in a while, there have been a few non-ASCII chars that print correctly, but usually all of them come out on paper as gibberish, even when they look good on the screen.
I've asked in various online forums, and the only replies I've gotten are either like yours, insulting me by telling me that it works without giving any clues as to how to make it work, or telling me that it misbehaves the same way on their Mac and if I figure out how to make it Just Work, I should tell them, too.
Earlier today, I accidentally typed CMD-P on this Mac in a Terminal window displaying a message that I wanted a hard copy of. Then, as it printed, I said "Duh!" as I realized my mistake. Half the message was in Hebrew, and I knew how it would come out. Sure enough, the Hebrew part was Latin1 gibberish, not Hebrew.
This is a very bizarre sense of "Just Works".
Maybe you should try reading the fucking article BEFORE you post.
Ob: You must be new here.
... I wouldn't be surprised if they told her to go to 192.168.0.1 on Firefox in order to get her internet connection configured.
My ears perked up at that, so I decided to try it. I have an Ubuntu (heron) box handy, where I fired up FF, typed in 192.168.0.1, hit Return ... and it told me "Though the site seems valid, the browser was unable to establish a connection." Hmmm ... Next, I opened a terminal window, typed in "ping 192.168.0.1", and got the results:
$ ping 192.168.0.1 ...
PING 192.168.0.1 (192.168.0.1) 56(84) bytes of data.
From 192.168.1.17: icmp_seq=2 Redirect Host(New nexthop: 192.168.0.1)
From 192.168.1.17: icmp_seq=3 Redirect Host(New nexthop: 192.168.0.1)
From 192.168.1.17 icmp_seq=1 Destination Host Unreachable
From 192.168.1.17 icmp_seq=2 Destination Host Unreachable
From 192.168.1.17 icmp_seq=3 Destination Host Unreachable
So it looks like telling her to go to 192.168.0.1 on Firefox could be rather bad advice here. As a longtime unix user, with a couple decades of internet software development, I have no idea why this is supposed to do anything useful in general. Yes, I can see how I might install a machine with address 192.168.0.1, install a web server on it, and install software in htdocs that handles network configuration. That's easy. But I didn't expect anything like that was set up on this LAN, and sure enough, it isn't. I'm not sure what's going on with all those redirections, because the gateway machine at 192.168.0.17 (don't ask ;-) shouldn't to my knowledge do that. But it definitely doesn't get you any net config tool.
But "Go to 192.168.0.1 on Firefox" is clearly not a viable approach on this Ubuntu machine. Are there large numbers of LANs (or out-of-the-box Ubuntu systems) where this would actually do something that a novice would find useful?
(Oh, and "Ask your network admin" isn't useful here. My answer would be "Hey, that's me!" Followed by "I have no idea." ;-)
Unfortunately Linux users have the tendency to call people idiots unless you approach them with "xyz OS desktop/OS can do this but blahblah Linux distro can't".
Actually, this has little to do with fortune or with linux. It's a fundamental feature of human psychology. The computer industry has long been plagued by it, of course, but it's much more general. The best approach is to just the above as the best approach.
Thus, back in the 1980s, when I first got onto unix systems and faced the problem known as "emacs vs. vi", I found the same thing. If I asked in any forum "How do you do task T with editor E?", the answer inevitably was of the form "WTF? RTFM, n00b!"
But I quickly learned that the proper approach was to phrase the question as a negative statement: "Hey guys, emacs can do T, but vi can't." This would almost always be answered by one or more explanations of exactly how to do it in vi. Interchange "emacs" and "vi" to learn about doing T in emacs. Sometimes they would insult me, too, but I learned to dig the facts out of the insults and go on my way. Yes, I learned the basic mechanics of both editors from smallish manuals. But the real learning was done by claiming that an editor couldn't do something, and waiting for the experts to show what a dummy I was.
I did eventually settle on vi, mostly because I found myself doing a lot of work on machines that were borked somehow, or were on the other end of a semi-sane network link that ate control characters, or whatever. I'd make the claim that "Emacs/vi can't be used unless you have a totally sane link to a totally sane system". The vi experts responded by explaining things like the "ex" interface and other tricks that made it usable in most cases. They also suggested learning "ed" and "sed" for extreme failure situations, but I'd already done that. OTOH, the emacs experts never answered such charges, telling me that emacs really wasn't usable on a damaged system. So I went with vi, while envying the folks that never had to deal with recovering a brain-damaged system or negotiating with comm gadgets that ate control characters.
So maybe our tactic should be to teach people to approach linux experts with the "Linux can't do T" sort of charge. Don't ask the linux experts how to do T; challenge them to teach a dummy like you that linux can do task T. This will also help greatly with getting info about Windows, of course, but that's probably not what most readers here want to do. I've also found that this tactic does wonders with getting Mac fanboys to loosen up, stop chanting "It Just Works!", and explain how to get the little things to do the job right.
Claiming that "System S can't do task T" is a general tactic that the public should know about when dealing with any tech experts. If we educate people about it, it might go a long way toward getting them on our side by making their computers do what they want.
My current favorite claim is that linux software can't print UTF-8-encoded text with stuff like Arabic or Chinese text. I suspect that this probably isn't true, because linux is widely used in Asia. But so far, it seems true, because about all I've found is assertions that it's possible, without any coherent explanation of how people (in Asia?) are doing it. (I've also collected a list of email addresses of people around the world saying that if I find a good answer, I should forward a copy to them. ;-)
(The Mac crowd also doesn't answer this charge, either, so apparently Macs can't print UTF-8 text, either. I'd be happy to be proved an idiot here ...)
Actually, I am still using an old Powerbook. We had two of them here; one died and one still lives (but is getting a bit feeble). I'll probably order a new Mac laptop of some sort within the month.
No, he (she) isn't. The first thing I did after reading the summary was to pick up my Mac Powerbook, cd into my Music/Itunes directory, find a couple of .m4p files, and run the strings command on them. Adding a few greps to filter out the printable binary junk, I quickly found my name and email address.
As for someone writing a tool to replace them, I found that I already had one. Years ago, I wrote a little command-line app that just does a simple string substitution and writes the result to stdout. It's quite handy, and I use it all the time. I told it to copy one of the .m4p files, with my email address replaced by a fake email address of the same length. I then told iTunes to load that file - and it played fine.
Then, of course, I did the same trick, replacing my name with a different name of the same length. As I expected, iTunes popped up a little window saying that it needed to check the tune's registration, showing me the name, and asking for a password. Presumably when DRM goes away, that little window will also go away, and I'll bet that the tunes will play.
I don't think I'll bother posting the program. Any semi-competent beginning C programmer should be able to type it in under a minute. Probably most perl and python programmers can do the same, a bit faster, as could any moderately experienced emacs user. 25 years ago, when I first picked up the C bible, I wouldn't have found it a challenge after my second day with the language.
Just make sure the replacement strings have the same byte count as the old name.
In some nations this would be illegal as Word is no ISO standard. You cannotn require someone to buy a specific product. It would contravene competition laws.
Actually, it's not that simple, even if your country has such laws. Note that the school teacher and administrators in question didn't explicitly require that the student (or his parents) buy anything. What they demanded was that assignments be handed in in a specific format (Word), using a specific font (Times New Roman). These are the names of Microsoft products, but I'd guess that the name "Microsoft" wasn't mentioned by the teach or or the admins. Most of them probably didn't even know that they were demanding that the child use Microsoft products. They certainly didn't insist that the parents buy anything from Microsoft.
This means that the laws in question don't apply. Or at least, that's what the school will argue. Getting them to stop doing this will almost certainly require a court case in most jurisdictions. And this will be a huge expense for most parents. They're more likely to take the easy way out, and buy a cheap Microsoft-based system, rather than face the time and expense of taking the school system to court.
Huh? You used scientific algebraic notation? On slashdot? And you expected people here to understand you?
You must be new here ...
If you don't get anywhere with the teacher, you should definately ask the school board to put the topic on your agenda.
Or maybe you should point out that the teacher has required a name-brand file format and font that are proprietary, and you don't have a license for them. Suggest that if these are required, then the school should pay for your child's computer with license to use such proprietary products. Mention that if they refuse to pay for your child's computer, you know some lawyers that will help you get a court order for reimbursement for the price.
(And you really should check with any lawyer friends about the legality of a school requiring that a student bring a name-brand product to school. It'd be more fun to push for a legal precedent that the school must pay for any such proprietary material required for classwork. ;-)
What happened to good old "Scientist"? It's a nice, nine letters long, and respected. "Experimentalist"... It sounds like what a social deviant might call themselves. ...
Of course, the more common term is "experimental scientist", as opposed to someone like Albert Einstein or Stephen Hawking, who were/are mostly known as theoretical scientists.
But "experimentalist" is a valid English word, makes sense in context, and has fewer syllables than "experimental scientist" while still emphasizing the experimental nature of their work.