It'd be soooo much easier for Aunt Millie to follow these instructions:...
Heh. I've watched a fair number of Aunt Millie analogs trying to use both Windows and Mac GUIs to do such things. I've watched as they get more and more frustrated, and tell the world what they think of the idiots who imposed such a godawful way of doing things on people like them. So no, I don't think that the existing GUI config stuff is any better than the CLI stuff. It just takes more keystrokes (most of the time), and changes more with every release.
2. Locate a line that looks like: HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Param eters\Interfaces\{GUID}...
3. Create a new line below the correct{GUID} and type DhcpConnDisableBcastFlagToggle = 1
One of the things I've seen with plain-text stuff is that the programmers know that such garbage names will be seen by users, and those users will mock the idiot programmers who imposed such names on the users. Especially things like setting a "Disable" flag to 1 to turn something off; even the dumbest user knows how perverse this is. Not that you don't see such things in plain-text stuff; you certainly do. But there is subtle pressure (based on the desire to not look too hostile to your users) to call it "BcastFlag" with 0=off and 1=on.
Another observation is that, while geeks like us certainly can handle such instructions, we often need several tries to get it right. I notice that there's a spurious blank in the "Param eters" field, and I'd expect that it will make the operation fail (or be ignored) if you type it that way. This is one more thing that helps encourage programmers to be less user-hostile and use a simpler heirarchy for such things. And not include blanks in names, as is done so often in GUI-based packages.
Not that things are really all that much better in the CLI world. We still have a long way to go before we can honestly claim that we're user friendly.
Of course, more or less. But that's only relevant the first time you have to config something. The second time, you know how to open a file, search for a keyword, and edit the line. You don't have to relearn these things a zillion times, once for each special-purpose config thingy.
Actually, the simple-minded sort of time-and-motion observations (that I've been doing mostly out of boredom) say there's often still an advantage to the plain-text, command-line approach. This is in great part because of something that can get really annoying with most current GUI tools: Every time you open a new window, it's back at your home directory, and you have to repeatedly wend your way through the stack of directories to find the file. There's nothing like the "cd" command to remember a point in the directory tree, or a "set" command to give a short name to the path. And the GUI tools have this annoying way of suddenly closing a window when they think a task is done, losing the directory information and forcing you to "drill down" to it again as you make repeated tries to get thing configged right. The GUI folks might be able to invent something like the "cd" concept, but they don't seem to have done it. Even the vaunted Mac interface is bad here, forcing you to waste time repeatedly getting yet another tool to the same directory.
(Not that the CLI folks have solved all the world's interface problems, either. But they've been more honest at facing up to the problems, and kludging up partial solutions to a lot of them.;-)
Hmm, and I thought getting stuff to work by having to edit a text file or copy&pasting something into the commandline was annoying.
Actually, this is typical. GUI config things are almost always much more complex and difficult to use than just editing a text file. But there has been a massive propaganda effort to convince people otherwise, and people fall for it.
Back in the 70s, when unix first came out, one of its widely-touted improvements over other computer systems was that you could configure everything with a single tool, a text editor. All the commercial systems had been moving over to fancy full-screen interactive stuff, which was done on character-based terminals, but were otherwise similar to the current GUI stuff. Every package had its own config tool, every such tool was different from every other, and you had to spend a lot of time learning each of them.
If you had to make a change 6 months later, you wouldn't remember that config tool, so you'd have to spend days relearning it just to change a couple of numbers that were hidden somewhere in a confusing mess of screens linked together in mysterious ways. And if you upgraded a package, the config tool would change, so you'd have to spend days relearning it. But unix-based packages mostly had a plain-text config file, editable with any editor. Conventionally, config files had lots of comments and examples of how to do common things. You could quickly scan a config file, searching for keywords, make the changes, write the file, restart the daemon if necessary, and a few minutes later, it'd be done. Sorta like how the apache server works now, y'know.
It is interesting that people fall for the complex GUI approach to configuring things. You'd think people could figure out for themselves that there's a much easier way, at least after they've seen a few editable config files. I wonder why people are so thick-skulled about such things?
People using Vista are very likely to just have bought a new computer since the beginning of the year, and have no idea why things don't work with it.
Leave out the brand name "Vista" and you've pinpointed the problem. Vista, like most new releases with major changes, has been reported to cause lots of problems for users. This has little if anything to do with the fact that Lund is using linux on their servers. It's because Vista is a new system. Similar problems happen with new releases of linux, and people don't try to blame their ISPs for such problems; they (usually) just report the problems calmly, and people try to find a fix.
Most people with any computer experience at all know that new things tend to be buggy. Actually, it's not something that computer people invented. Buying a new model of car involves the same sort of risk, and anyone with any sense knows this. You don't buy the first model. You wait for the suckers to buy it and find the bugs. After a while, when the bugs are (mostly) worked out, you consider buying it yourself.
The main "culprit" here is the crowd of suckers who bought Vista in the first year. Rather than ranting about some bogus Vista-vs-linux smackdown, we should be calmly explaining to people that this is exactly what you'd expect with such a major new release.
And since problems with Vista are already all over the press (at least the technical press), people who buy it can't reasonably expect it to "just work". And in particular, they can't reasonably point their fingers at someone else running different software, and say "You made my Vista not work right." They should just realize that they were suckered into being field testers for Microsoft, and they're suffering for that bit of gullibility.
I do sorta wonder if Lund is hiding the details on the bug, as the Microsoft people seem to be saying. Or maybe they put the details online, but the MS people are running Vista, and can't read those pages?;-)
If was a bad guy and I was worried that 'they' were on to me, receiving this trojan would be proof positive.
Nah; it would just mean that you had a computer (presumably one running MS Windows;-).
Note that they want the right to send it to any "terror suspect". The word suspect means anyone at all. If challenged, all they have to say is that they suspect you of something. Or they suspect a relative of yours. Or someone you knew in college 20 years ago. Or someone three houses down the street. Or someone with a name vaguely like yours. Or they learned that an ancestor of yours five generations ago wasn't German.
Such a law is really just a legal excuse to do nasty things to anyone at all, at any time.
The fun thing in this case is that you just know that their software would be isolated, probably within a week, and would soon be available at warez sites everywhere, for anyone's own private use. Someone annoying you? Send them a trojan that would start reporting all your keystrokes to the police.
I don't suppose you'd care to explain how the U.S. would go about enforcing domestic law in Sweden.
Easy. The US government's law-enforcement agencies, such as the Dept of Justice, the FBI, etc., have a long history of cooperating with corresponding agencies in other countries. It's quite common for such agencies to assist each other in investigations, collecting evidence that can be used in another country's legal proceedings. I'd guess that the Swedish government has already contacted several American government agencies on the topic of the recent vote buying. The ISO is probably also talking to NIST and the US Dept of State about the issue.
National borders aren't really all that serious a problem when government want to cooperate on something. The real question here is whether the Bush administration (and the Gonzales Justice Dept;-) want to cooperate when the suspect is one of their largest campaign contributors. That's really the important question. Whether the US and Swedish government cooperate isn't a question; they generally do. But the Bush people may well decide to block such cooperation in this instance.
Do you have an example of another productive, exporting industry in the US?
Historically, education has been one of the US's major "export industries".
There is a certain amount of irony here, since the US population as a whole has a rather deep antipathy towards education and educated people. It's easy to see this in American politics, where a college degree is a handicap that politicians tend to downplay. The recent fun over the video clip of a Miss Teen USA contestant explaining why Americans can't find things on a map is, sadly, all too typical of the general population.
But at the university level, the US has long attracted a large number of students from the rest of the world. And, again sadly, the recent "anti-terrorism" measures of the US government has seriously impacted this, by viewing most foreign students as potential terrorists and blocking their entry or seriously harassing them. I suppose this might be considered good news for educational institutions in the rest of the world, but it's not good news to people who just want good education to be available to anyone that wants it (and can pay for it;-).
Here in the Boston area, people are generally aware that education is now the most important local "industry". There are around 250,000 college-level students in the metro area (and most of them are arriving this weekend, making for some major traffic problems;-). My wife got a degree at BU (Boston University) some years back, and most of her friends there were foreign students. Actually, most of them were female, and the conventional explanation was that BU is where the wives of foreign Harvard and MIT students went. They liked to explain that their governments generally paid for their husbands' education, but women have to rely on their families for such funding. So the women usually go to a somewhat cheaper school in the area, while their husbands are off making business connections at the high-status schools.
If you're buying a laser printer with a view to Linux compatability in the future, look for one which supports Postscript. All this talk of "Drivers this... drivers that...." - it's cobblers. Postscript has been a perfectly good language for printers for something like 20 years, and postscript printers don't command anything like the same premium that they did 10 years ago.
What I'm looking for is a printer (or printer driver or whatever) that correctly prints files encoded as UTF-8. I have a bunch of multi-language postscript files that I can show contain the right text. I just edit them with vi(m) on my Mac or linux boxes, and you can see the Chinese or Arabic or whatever there on the screen. But when I send the files to any of the available printers, the Chinese or Arabic (or whatever) text comes out as Latin1 gibberish.
Anyone know of any printers and/or printing software that gets this right? Can you print files that are a mixture of English, Russian, and Japanese, for example? If so, how do you do it?
Presumably people in Asia have printers that can print their native languages. But so far, I haven't found out how to get it to work here in the US. Of course, "Customer Support" people around here don't find it interesting, because why would anyone want to use anything but English? OK, maybe in Spanish or French, and those do work. (But sometimes I wish I could get them to move their support to Asia.;-)
It's an old story. People in power find a way to prevent you and me from profiting from our own creations, by controlling the distribution channels.
The music recording industry is the poster child for this. Until recently, selling recordings required expensive production and distribution facilities. The owners of these facilities could say "You want people to hear and buy your music? Sign this contract and we'll make it happen." You might as well sign the contract, because it's the same for all the other distributors. And in the fine print, it says that you assign the copyright to the recording company. The result is that musicians can make a million-selling album and make no money from it at all. This is because the recording companies can say "If you want your stuff distributed, you must give it to us."
Almost all work "for hire" to corporations are of this nature. If you want to be paid, you have to assign ownership to your employer. If you're a university researcher, and you want to be paid, you usually must assign copyright and patents to the university. Unless, that is, you got funding on your own, in which case you must assign copyright and patents to the funder. And if you want your results published, almost all academic publishers have historically required that you assign the copyright to them. "If you want your stuff distributed, you must give it to us."
Now we have a new means of distribution, the Internet. That promised to give us a cheap distribution mechanism that wasn't controlled by the distributors. But For most of us, to use the Internet requires going through something called an "ISP". Those organizations, usually private companies, have a chokehold on your path to the Internet. Early on in the commercialization of the Internet, the ISPs started to realize what they had. Thus, a few years ago, we read the stories of msn.com (owned by Microsoft) using things from customers' web sites commercially. They mostly extracted images and used them in advertising. When customers discovered images of their children being used in ads, they understandably got upset. And msn.com pointed to the clause in the contract saying that any customer files stored on msn.com machines became the property of msn.com. After a bit of adverse publicity, the "gave in", in the sense that they publicly announced that they wouldn't do this again. But this was like any corporate promise: It was PR to mollify the current crowd of upset customers. After a while, people started noticing that that clause was still in the TOS. And it can be summarized as "If you want your stuff distributed, you must give it to us. You can't go to the competition, because their contracts say the same thing."
So this story is nothing new. ISPs are more and more realizing that they have a chokehold on customers' channel to the rest of the world. Most people have only one ISP, which is a legal monopoly. Even when there are two, they can easily make sure that their contracts are identical. Like various monopolists/oligopolists of the comm channels before them, they can say "If you want your stuff distributed, you must give it to us. You can't go to the competition, because their contracts say the same thing."
It now seems like google, the "Do no evil" company, has realized the same thing. They can provide customers useful tools that inprove people's access to the Internet. And they can hide "If you want your stuff distributed, you must give it to us" in their contract. You can't go to the competition, because their contracts say the same thing.
The only way around this is regulation that denies the controllers of the Internet any ownership of things that pass through their machines. But this sort of regulation has never been effective for any past distribution system. There's no reason to expect that it will be effective for the Internet.
So much for your rights to your own creations. Get used to it; it's the future. Just like the past.
If you dig around in the archives of a century ago, you'll find that this is essentially the same situation as with telephone and electrical service in rural areas.
The telephone story is the most comparable, obviously. 100 years ago, there were two sorts of problems. One was that you couldn't get phone service out in the sticks, because the phone companies didn't think it was profitable. Meanwhile, in the bigger cities there were often several competing phone companies, but you could usually only call someone who subscribed to the same phone company as yours. The ISPs haven't implemented this last problem, because the Internet software came with compatibility builtin. But they're working on it, and if the "Net Neutrality" goes the way they want, you'll find that you won't be able to see web sites of a lot of companies that your ISP doesn't like, especially their competitors, but more generally any sites on other ISPs' lines.
With electricity, the story was simpler. The electric companies didn't build out into the countryside, because it wasn't profitable enough.
Both of these effectively ended when the government stepped in and imposed regulations. The phone companies were told that they'd provide rural service and interoperability, or they wouldn't sell any service at all. This was done somewhat less with electricity, mostly because the Rural Electrification project discovered that they could create locally-owned electrical co-ops that could do the job. They did need regulation forcing the electrical companies to sell electricity to these "competitors" at a reasonable price, but the urban electrical companies weren't always forced to build out into rural areas.
So in effect, universal telephone and electrical service would never have happened without the government regulators stepping in and decreeing that the services be provided to everyone. There's no historical reason to expect universal Internet access to happen in any other way. The comm companies naturally want to sell to only the most lucrative market, and can't be bothered to deal with those rural yokels. If the rural folks want Internet service, they're just going to have to push for the same sort of solutions that gave them phones and electricity a century ago.
The best idea might be the rural co-op, as this would give them a semblance of control over their own service. Of course, the big comm companies are already fighting this, by bribing legislatures to ban anything that smells like government-provided Internet. But unless people start agitating for such things, nothing is going to happen any time soon.
I live in rural Ontario, Canada on a farm.... it's maybe $70/month when you factor everything in, but for that price I have a nominal 3Mbps/512kbps connection with a static IP, and no bandwidth caps or restrictions.
Well, I live in a fairly densely populated suburb of Boston, and the best we can do here for a static IP and no restrictions is $100 a month for a speakeasy DSL line that delivers 1.5/320 MB.
We actually have several providers, but once Verizon succeeds at persuading the FCC to "deregulate" us, speakeasy will be kicked out, and currently Verizon charges $200/month for a comparable line. We also have Comcast, but the last I checked, they wouldn't do static IP or promise not to block ports on a "residential" line for any price. Both Verizon and Comcast are locally documented to also block stuff like Skype packets, though their PR people look very innocent when claiming that they would never do such a thing.
Sounds like Canadian rules are a lot better for the customers than the rules around here.
What if the timeline were reversed, and we were moving from online apps to the desktop.... 'You can image the advertising push. "Now control your own data!" "Faster processing power now." "Cheaper!" "Everything at your fingertips." "No need to worry about network outages...."
I you look back to the early days of "minicomputers" and "desktop computers" around 1980, you'll see a lot of exactly the sort of arguments like the above. What was happening all over was that users of central time-shared "mainframe" computers were using whatever departmental funds they could to buy their own small computers, so they could abandon the central mainframes. The DP people were horrified, of course, and argued that the central machine provided much more cost-effective processing than a comparable flock of all those little machine, and the little ones could hardly communicate with each other.
But the users were fighting frustration at the difficulty of getting anything done on a big machine controlled by a different department. Their argument was that with their own small machine, they could install the software that they needed, without getting the approval of the bean counters, and the little machine would do what was needed when it was needed, not at the whims of the DP people. Yes, the little machine would eventually cost more, but they would answer to their users' needs, not to a remote disinterested bureaucracy's needs.
The current argument for central servers is really just a replay of this. And it'll mostly fail for the same reasons. A big central server will be controlled by a bureaucracy that has its own motives and needs, and won't be responsive to the lowly users' needs. Smart users will find ways of going around this and using a local machine for the things that the server just can't be made to do right. Management decrees won't work any better than they did with mainframes, because pressure to get your job done will always override pressure to hand your job over to the mercies of the remote central bureaucracy.
Here and there, a few DP (or IT or ??) departments will do a good job and really support their users. They will be used as evidence "proving" how good a central server can be. But most central servers won't be run by such departments, so most users will resist such centralization whenever they can.
Actually, the LoC is slowly putting stuff online. Unfortunately, it's mostly in the form of scans, since they don't have the personnel to do a proper job of transcribing. There is OCR software that (mostly) works for text data, but nothing much that works for music notation. This is a lot harder, because all the symbols have these lines going through them.
If you dig around in loc.gov you can find scans of all sorts of old sheet music. But it'll take you a while to find things, especially if you're looking like a musician would want to. Search for "a waltz in E minor with an average of 3 or more melody notes per measure". Yeah, right; OCR software is gonna do that for you.;-)
Why can't Microsoft compete without buying the outcome of the game? Are their products that poor?
Yeah, mostly, but that's irrelevant. They do have a few good products, but that's also irrelevant to sales.
Microsoft's entire history, and IBM's for the previous decodes, demonstrates quite well that sales in any computer-related field are determined almost entirely by marketing budget. Quality is nice, but it doesn't add materially to sales, so if you have the marketing clout, there's no financial reason to also invest heavily in quality.
Sorry to break the news to you. The best product doesn't win. The best-marketed product wins.
There's no (financial) reason that MS should care whether OOXML is good or bad. Their primary concern is that people use it, and this only requires that it be minimally usable. Investing what is for them a small amount to get their encoding declared a "standard" is just a (standard;-) marketing approach, and it would be puzzling if they didn't do it.
Turn that around: How do YOU sleep at night knowing that 'making available' a song that you don't own could wipe out your own savings? It's such a little thing, and SO easy to avoid... And yet, you do it anyhow.
Easy to avoid? People have been pointing out that we are rapidly approaching the day when, if you walk down a sidewalk whistling a tune, you'll be arrested and charged with unlicensed performance of a copyrighted work of music.
Fact is that the only practical way to avoid this now is to never say or do anything at all in public (which includes on the Internet). I've tested this a few times by asking a simple question: Suppose I have a tune in my head, and I'd like to discover whether it's something I "composed" myself or is a tune whose copyright is owned by someone. How do I do this?
I have asked reps of a couple of music publishers. Their answer, apparently said with a straight face as far as I can tell, is that I should buy a copy of everything they've published and search through it all. Of course, this only works with that one publisher; to actually answer the question, I'd have to buy a copy of every work of music ever published by anyone and search them all.
There is something of a shortcut. Here in the US, the Library of Congress (LoC) has copies of most of what has been published in the country. I could go to Washington and spend a few years searching through their archives. Then I could do the same in all other countries. This would only take a few lifetimes, not the thousands of lifetimes that the "buy and search everything" approach would take. But still, there's a certain limited practicality here.
Fact is, the only way I can determine in my lifetime whether that tune in my head is copyrighted is to publicly perform it, preferably in a recording, and wait to see whether anyone sues me.
Actually, there is a less reliable but more practical way that a number of musicians have been using: Put a copy online (either as music notation or a recording), accompanied by a note saying that you haven't determined whether it's copyrighted, and if anyone knows who owns the tune, send a message to <email-addr>. This isn't guaranteed to protect you, because the owner might be a bastard who will sue you for even this transgression.
And it still has the problem that, in practice, you get mostly copyright claims that turn out to be bogus. Publishers regularly claim copyright on music that's centuries old. If you can show this, they'll slink of to look for another victim. And sometimes this happens, because what happens is that someone sees your note and sends a message saying "That was published by So-And-So in London in 1793 as <title> in <book>." If you know this, you can use it as a weapon against the publisher.
But it's all very unreliable, and depends on the good will (or reasonable lawyers) of corporations, in addition to help from other musicians who stumble across your stuff. In general, there's no way to know that a random public utterance or idly whistling a tune won't be a copyright violation. The only really safe strategy is to be utterly silent in public. On the Internet, this includes learning enough about your computer's innards to guarantee that it isn't exposing any file to outsiders.
So basically, the enlightenment was due to Gutenberg?
No; it didn't have a single cause. Gutenberg certainly made a major contribution and helped to enable the Enlightenment. But his technical advance potentially made publication easier and cheaper for everyone in the world, not just in Europe. The really important advance wasn't in the hardware, but rather in the "software", i.e., in the social structure that developed the concept of open publication of scientific results. This could have been done anywhere, and could have happened before Gutenberg. It happened in western Europe, whose scientists took advantage of the improved publishing technology and put it at the core of the scientific enterprise. In much of the rest of the world, publishing was (and still is) controlled by the ruling classes, i.e., by politicians, so they lagged behind Europe.
There are other social innovations that were needed. I read an interesting comment a few years back, by a French researcher who explained why he always published in English. It had nothing to do with the size of the audience. His explanation was that doing good scientific work often requires that you invent terminology, and sometimes subtle differences in terminology can be the difference between a successful hypothesis and a failure. In French, there's a national language bureau that has the legal power to decide how the French language may be used, and it's full of people with no understanding of his scientific specialty. So he can't freely invent new terminology in French and use it in his publications. Well, he sorta can, but doing so risks legal harassment and possible revisions that would destroy the scientific usefulness of his work. The English language is an insane free-for-all without any official, centrally-controlled, legally-enforced rules. So in English, he and his colleagues can work out their own terminology without any official harassment. Once they think they've got it right, they can "borrow" the terminology into French, of course, but even then they sometimes get harassment for using Englishisms in French. So he does a good Gallic shrug, and publishes in English.
It's interesting to think about. The scientific revolution did depend on a number of independent developments. Some of these are negative, like the lack of official rules for the English language. Publishing technology is at the core of a lot of science. Right now, we're going through the pains of switching to new technology (the Internet) that's orders of magnitude faster and cheaper than what Gutenberg gave us. The societies that do this right will be the scientific leaders in a few decades. This might not be an English-speaking society, if the current "Intellectual Property" debate goes the way it has been going in the English-speaking parts of the world.
Actualy, the reason that the Internet is sucessfull is that there is a complete implementation of the main protocols available (BSD licensed) to anyone that want to use it.
Well, I wouldn't agree that this is the reason. It certainly helped get things rolling. But I'd argue that if the geeks at Berserkeley hadn't done the job, it would have been done by others. And it didn't have to be done at one place. The division of the design into layers and different specialized protocols makes it easy for independent developers to work on the pieces independently. But having it all in a single, freely downloadable package did a lot to encourage people to use it. At the time, our current Internet didn't exist, so there was an advantage to having it done by a group of people at a single university.
But all this modular design was true of, for example, the OSI protocols. On several projects, I saw first hand why OSI didn't take off. The OSI docs weren't published online (at first); you had to order printed copies. So I'd fill out a purchase order, hand it to the secretaries, and wait for the needed signatures. Finally it would get approved; they'd mail it off, and I'd wait some more. Finally, weeks or months after starting, I'd have the spec that I needed. In the meantime, to keep myself busy while waiting, I'd download a few RFCs, fire off questions to mailing lists, and work on some prototype code that I'd carefully keep modular with the idea of retargeting it to the OSI protocols. By the time the OSI docs arrived, I'd have a working version of what we wanted running on IP. You can guess what the group usually ended up using.
It didn't hurt that the Internet specs were usually simpler than the OSI specs. Thus, on one project (at Digital in the early 90s), we were getting frustrated with the problems getting SLIP to work over our modems. One day I decided to try a reimplementation. ("How hard can it be?";-) I downloaded the RFC, printed it out, and took it home for bedtime reading. Bright and early (ok; 8:30 am) the next morning, I started coding. By noon, I had a working version. We installed it in a few machines, ran our test suite, and everything worked. Less that 24 hours from the "Let's do it" decision to having a usable implementation.
Not that the IP-based stuff was always that easy. Some months later, we decided to give PPP a try. It was somewhat new at the time, and we didn't have a version yet on our machines. So I again downloaded an RFC, and went to work. This time, working on it about half time, it took me about a week, because PPP is a bit more complex than SLIP. We liked it better, because it wasn't prone to failure due to the assorted insanities built into all commercial modems.
Funny thing is that most management doesn't get any of this. I've been to a number of interviews where they ask about driver or protocol experience. I mention a few cases like the above, and invariably the reaction is "That's only a few days' experience; we're looking for someone with several years experience." So I don't get the job, and sometimes they make it clear that they think I'm lying about the time it took. I've even told them that I can make it take several years, if that's a project requirement, but somehow that doesn't convince them. (Of course, I only say that after it's clear that I'm not getting the job.;-)
[N]obody should be surprised that Bush appointees (here and in Iraq) have consistently proved themselves incompetent.
Well, that depends on what you think they're trying to do. Others have concluded that many of those supposed "incompetents" are actually highly competent, and have succeeded admirably at their goals.
For example, consider that Gonzales may have been appointed Attorney General specifically to block the Justice Department's natural tendency to investigate and prosecute corruption. Various people have pointed out that many of those fired attorneys had something in common: They were investigating corruption in government contracting. By firing them, Gonzales terminated those investigations (and got the message across to others who might be tempted to start their own investigations). The result has been things like the billions of dollars that have disappeared without a trace in Iraq. Companies owned by the Bushes and their associates, such as Haliburton's subsidiaries, have profited tremendously from the lack of Justice Dept attention.
If you surmise that this was the intended result, then appointing Gonzales was a very competent decision by Bush, and Gonzales was in fact quite good at doing his (actual) job.
Their immediate problem might be that their overreaching has gotten to the point that even their so-called "conservative" supporters are starting to get offended by the audacity of the corruption, so they had to ease Gonzales out. Very conveniently, they waited until Congress wasn't in session, so Bush can replace him with a recess appointment and avoid congressional oversight once again. And this could be viewed as a very competent action on their part, demonstrating that they know how to exploit the fine print to relieve Congress of any bothersome need to get involved in presidential appointments.
[T]here are actual people in the world, who PAY for some research through their taxes, and then does not want the ownership of it through public domain.
I've read quite a number of histories of science in which the authors point out that the Western "scientific revolution" during the last few centuries has nothing to do with discovering "the scientific method". Scientific methods have been independently discovered in nearly all societies, going back into prehistory. One example I ran across recently was the comment that from what we know of their methods, the North American Indians' "medicine men" had better medicine than Europe did until sometime in the 1800s. The reason was that the medicine men actually had better scientific methodology than European doctors did. But then things changed
So what caused the big advances in Western science in the past few centuries? The historians answer to this is simple: open publication. In all other societies, such knowledge has almost always been strictly controlled by small "guilds". The knowledge was secret, and discoveries were usually not even shared with colleagues outside the discoverer's immediate circle of professionals. This meant that everything had to be rediscovered over and over again. You could only learn what your mentors knew, and you could only build on what they passed on to you, because everyone else's knowledge was unavailable to you.
But a few hundred years ago, some researchers in Europe developed a curious new approach: They published their discoveries openly, making them available for others to read, use, and build on. This led to the explosive growth of knowledge that we're familiar with.
In most of the West, such open publication is historically done only by government researchers. Before the 20th century, this meant the few idle rich such as Isaac Newton, who had brains and curiosity. So it took a while to really get going. But in the 1900s, various governments slowly got it through their thick skulls that funding research was one of the things that was building other countries' economies (and militaries), so maybe they should be funding research too. Then things really started ramping up.
But there is still one major drag on scientific advance: A lot of funding still goes into "private" (i.e., corporate) research. This is, scientifically speaking, usually a dead end, because the results of such research is kept private, and as of with the guilds of old, it isn't available for others to build on. The legal system cooperates in this, by prosecuting people who get access to private research results and try to build on it. In recent years, this has been happening more and more in the US, as the corporate world consolidates its control over the government and determines how most research is funded. Some universities also contribute to the problem, by claiming ownership of research results when they can and keeping it secret (or usable only under high license fees). I've read a few predictions that useful American research may be ending now, as the corporate world takes most of it private. And it's curious to see the publishers of scientific journals jumping in to block the advent of cheap open publication via the Internet.
Anyway, at least according to these histories, we should be supporting the open-publication people, because they're the ones pushing for more of what really made Western science the success it has been. If we really want Science to continue to improve our lives, we should be pushing for full access to all research results.
(And people have pointed out that the Internet is an example of the same phenomenon. It hasn't succeeded because it's such a great network. The IP/UDP/TCP/DNS/SMTP/HTTP/... protocols aren't all that good, actually. They're good enough to do the job, but anyone who knows them can tell you lots of ways to improve them. The reason for the Internet's success is mostly that all the specs have been published openly from the start. Anyone can download them for free, read them and implement them without legal restrictions. This gave the Internet a huge advantage over other privately-developed protocols that were often technically better but weren't available to any developer that was curious.)
You do _not_ talk about Linux crashing on Slashdot.
Heh. One question that came to mind immediately was: Did those people actually see linux crash, or did they see linux rebooting. These are two very different things, of course, but I've found that even experienced users can be rather sloppy about such insignificant details.
I'm sure that most people who've taken commercial flights have noticed things like the cabin lights all flickering at times, especially during takeoff. No big deal for lights, but this will "crash" pretty much any computer using that power system. And the crew obviously has control of the lights; why would you be surprised if they could also control all the computers? Just as the control panel has buttons to turn all the lights on and off, I'd expect that you could reboot all the seatback computers from the control panel. You could also override what they're doing and show some particular video clip.
I'd think that the ability to tell all these little computers to do something simultaneously would be a good selling point. And considering how the power-supply systems work in commercial airliners, I'd expect that it would be normal to do a net reboot of the little buggers frequently. The ability to easily do such things from a central control system would be a major selling point.
So, rather than us _not_ talking about linux "crashing", I'd think we'd be interested in information about how these seatback computers are configured and how they're managed by the crew.
Not that I expect to see much real information here. And I'd guess that much of the airlines' control software is proprietary, for "security" (by obscurity) reasons.
I'd imagine that one of the driving forces here is that listening to music is now an outdoor activity.
Nah; that can't explain it. After all, if all the recordings were at half the current level, then you'd just turn the volume dial to get the same loudness.
To understand the escalation, it works better to think of someone listening to their Walkman or boombox or iPod, which is set to "shuffle" (or to a music radio station, which is similar). What the recording companies want is for the track you're listening to now to be slightly louder than the one you just finished listening to (and to the one you're going to hear next), but not enough louder that you turn down the volume.
This leads them to sample the competition, and set their tracks to a bit louder than the average. Over time, they all get louder, until they're all at the max intensity with no dynamic range. Of course, this will be indistinguishable from near silence, since listeners will have the volume at minimum.
I could swear this program has been "killed" twice, and by "killed" i mean the government's definition: proclaiming a project discontinued while continuing it under a new name.
This sort of thing has been reported for US government agencies for decades. Back when all the data was all in paper archives, there were lots of reports of agencies that "obeyed" orders to destroy them by first running them through copiers, transporting the copies to some other site, and then destroying the originals. I recall during the Vietnam War, when the DoD was ordered to destroy their records of spying on anti-war groups, and the DoD announced that it had done so. But some time later, reporters published reports that at least five backup copies had been located at five different sites.
No matter how dumb you think government (or corporate) employees are, fact is that most of them are smart enough to figure out this ruse.
There's also the reverse version that's been making the news: We're now reading reports of managers (government and corporate) finding the contents of their email used against them in court, when they thought that the email messages had been deleted. The messages had been deleted from disk, of course, but the backups could still be read. Oops!
When you read stories like this, you should always ask yourself "How many copies were made?" That will put you into the proper cynical perspective, which you'll later remember when it turns out you were right to ask.
Good advice, for those who can. But here in the US, most people have exactly one ISP to choose from. If you don't like it, you can move somewhere else where there's a different ISP.
Some people do have two ISPs. But if you investigate, you discover that they both have almost exactly the same policies and restrictions (and prices), so it won't do you a bit of good to change.
A couple of years back, while discussing such things with our then ISP (RCN), the support guy made it clear that to them, "Internet Service" covers web browsing and email. End of story. Nothing else was supported, and by supported they meant permitted. Oh, and by email they meant access via their own email servers; no other email was supported. All incoming and outgoing email must be stored on their own servers, and no, they won't tell you why they want this. We also have Comcast as an alternative, but I suppose most people here know what sort of alternative that is.
When speakeasy became available in our neighborhood, we switched. It was a bit slower than the "blinding fast" service advertised by RCN, but they allowed (and promise to always allow) unblocked Internet access. They actively encourage things like servers. In fact, they even encourage redistributing your service to neighbors, and they'll even handle the billing for you, so you effectively become a local distributor.
Too bad they aren't allowed access to most potential customers. Maybe we need to be pushing for regulations that only permit real IP service, i.e., just deliver the packets at whatever rate I've paid for.
(The really funny part of the story is that we looked into having Verizon, our phone monopoly, supply DSL service. They couldn't do it; we're too far away. So we got speakeasy DSL instead, over the same Verizon line. We've been happy with them.;-)
Slashdot Mirror
Let's try a google fight:
google evil about 65,900,000
pope evil about 2,080,000
Looks like google has the lead here, folks.
It'd be soooo much easier for Aunt Millie to follow these instructions: ...
m eters\Interfaces\{GUID} ...
Heh. I've watched a fair number of Aunt Millie analogs trying to use both
Windows and Mac GUIs to do such things. I've watched as they get more and
more frustrated, and tell the world what they think of the idiots who imposed
such a godawful way of doing things on people like them. So no, I don't think
that the existing GUI config stuff is any better than the CLI stuff. It just
takes more keystrokes (most of the time), and changes more with every release.
2. Locate a line that looks like:
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Para
3. Create a new line below the correct{GUID} and type DhcpConnDisableBcastFlagToggle = 1
One of the things I've seen with plain-text stuff is that the programmers know
that such garbage names will be seen by users, and those users will mock the
idiot programmers who imposed such names on the users. Especially things like
setting a "Disable" flag to 1 to turn something off; even the dumbest user
knows how perverse this is. Not that you don't see such things in plain-text
stuff; you certainly do. But there is subtle pressure (based on the desire
to not look too hostile to your users) to call it "BcastFlag" with 0=off
and 1=on.
Another observation is that, while geeks like us certainly can handle such
instructions, we often need several tries to get it right. I notice that
there's a spurious blank in the "Param eters" field, and I'd expect that it
will make the operation fail (or be ignored) if you type it that way. This
is one more thing that helps encourage programmers to be less user-hostile
and use a simpler heirarchy for such things. And not include blanks in names,
as is done so often in GUI-based packages.
Not that things are really all that much better in the CLI world. We still
have a long way to go before we can honestly claim that we're user friendly.
Of course, more or less. But that's only relevant the first time you have to config something. The second time, you know how to open a file, search for a keyword, and edit the line. You don't have to relearn these things a zillion times, once for each special-purpose config thingy.
;-)
Actually, the simple-minded sort of time-and-motion observations (that I've been doing mostly out of boredom) say there's often still an advantage to the plain-text, command-line approach. This is in great part because of something that can get really annoying with most current GUI tools: Every time you open a new window, it's back at your home directory, and you have to repeatedly wend your way through the stack of directories to find the file. There's nothing like the "cd" command to remember a point in the directory tree, or a "set" command to give a short name to the path. And the GUI tools have this annoying way of suddenly closing a window when they think a task is done, losing the directory information and forcing you to "drill down" to it again as you make repeated tries to get thing configged right. The GUI folks might be able to invent something like the "cd" concept, but they don't seem to have done it. Even the vaunted Mac interface is bad here, forcing you to waste time repeatedly getting yet another tool to the same directory.
(Not that the CLI folks have solved all the world's interface problems, either. But they've been more honest at facing up to the problems, and kludging up partial solutions to a lot of them.
Hmm, and I thought getting stuff to work by having to edit a text file or copy&pasting something into the commandline was annoying.
Actually, this is typical. GUI config things are almost always much more complex and difficult to use than just editing a text file. But there has been a massive propaganda effort to convince people otherwise, and people fall for it.
Back in the 70s, when unix first came out, one of its widely-touted improvements over other computer systems was that you could configure everything with a single tool, a text editor. All the commercial systems had been moving over to fancy full-screen interactive stuff, which was done on character-based terminals, but were otherwise similar to the current GUI stuff. Every package had its own config tool, every such tool was different from every other, and you had to spend a lot of time learning each of them.
If you had to make a change 6 months later, you wouldn't remember that config tool, so you'd have to spend days relearning it just to change a couple of numbers that were hidden somewhere in a confusing mess of screens linked together in mysterious ways. And if you upgraded a package, the config tool would change, so you'd have to spend days relearning it. But unix-based packages mostly had a plain-text config file, editable with any editor. Conventionally, config files had lots of comments and examples of how to do common things. You could quickly scan a config file, searching for keywords, make the changes, write the file, restart the daemon if necessary, and a few minutes later, it'd be done. Sorta like how the apache server works now, y'know.
It is interesting that people fall for the complex GUI approach to configuring things. You'd think people could figure out for themselves that there's a much easier way, at least after they've seen a few editable config files. I wonder why people are so thick-skulled about such things?
People using Vista are very likely to just have bought a new computer since the beginning of the year, and have no idea why things don't work with it.
;-)
Leave out the brand name "Vista" and you've pinpointed the problem. Vista, like most new releases with major changes, has been reported to cause lots of problems for users. This has little if anything to do with the fact that Lund is using linux on their servers. It's because Vista is a new system. Similar problems happen with new releases of linux, and people don't try to blame their ISPs for such problems; they (usually) just report the problems calmly, and people try to find a fix.
Most people with any computer experience at all know that new things tend to be buggy. Actually, it's not something that computer people invented. Buying a new model of car involves the same sort of risk, and anyone with any sense knows this. You don't buy the first model. You wait for the suckers to buy it and find the bugs. After a while, when the bugs are (mostly) worked out, you consider buying it yourself.
The main "culprit" here is the crowd of suckers who bought Vista in the first year. Rather than ranting about some bogus Vista-vs-linux smackdown, we should be calmly explaining to people that this is exactly what you'd expect with such a major new release.
And since problems with Vista are already all over the press (at least the technical press), people who buy it can't reasonably expect it to "just work". And in particular, they can't reasonably point their fingers at someone else running different software, and say "You made my Vista not work right." They should just realize that they were suckered into being field testers for Microsoft, and they're suffering for that bit of gullibility.
I do sorta wonder if Lund is hiding the details on the bug, as the Microsoft people seem to be saying. Or maybe they put the details online, but the MS people are running Vista, and can't read those pages?
If was a bad guy and I was worried that 'they' were on to me, receiving this trojan would be proof positive.
;-).
Nah; it would just mean that you had a computer (presumably one running MS Windows
Note that they want the right to send it to any "terror suspect". The word suspect means anyone at all. If challenged, all they have to say is that they suspect you of something. Or they suspect a relative of yours. Or someone you knew in college 20 years ago. Or someone three houses down the street. Or someone with a name vaguely like yours. Or they learned that an ancestor of yours five generations ago wasn't German.
Such a law is really just a legal excuse to do nasty things to anyone at all, at any time.
The fun thing in this case is that you just know that their software would be isolated, probably within a week, and would soon be available at warez sites everywhere, for anyone's own private use. Someone annoying you? Send them a trojan that would start reporting all your keystrokes to the police.
I don't suppose you'd care to explain how the U.S. would go about enforcing domestic law in Sweden.
;-) want to cooperate when the suspect is one of their largest campaign contributors. That's really the important question. Whether the US and Swedish government cooperate isn't a question; they generally do. But the Bush people may well decide to block such cooperation in this instance.
Easy. The US government's law-enforcement agencies, such as the Dept of Justice, the FBI, etc., have a long history of cooperating with corresponding agencies in other countries. It's quite common for such agencies to assist each other in investigations, collecting evidence that can be used in another country's legal proceedings. I'd guess that the Swedish government has already contacted several American government agencies on the topic of the recent vote buying. The ISO is probably also talking to NIST and the US Dept of State about the issue.
National borders aren't really all that serious a problem when government want to cooperate on something. The real question here is whether the Bush administration (and the Gonzales Justice Dept
Do you have an example of another productive, exporting industry in the US?
;-).
;-). My wife got a degree at BU (Boston University) some years back, and most of her friends there were foreign students. Actually, most of them were female, and the conventional explanation was that BU is where the wives of foreign Harvard and MIT students went. They liked to explain that their governments generally paid for their husbands' education, but women have to rely on their families for such funding. So the women usually go to a somewhat cheaper school in the area, while their husbands are off making business connections at the high-status schools.
Historically, education has been one of the US's major "export industries".
There is a certain amount of irony here, since the US population as a whole has a rather deep antipathy towards education and educated people. It's easy to see this in American politics, where a college degree is a handicap that politicians tend to downplay. The recent fun over the video clip of a Miss Teen USA contestant explaining why Americans can't find things on a map is, sadly, all too typical of the general population.
But at the university level, the US has long attracted a large number of students from the rest of the world. And, again sadly, the recent "anti-terrorism" measures of the US government has seriously impacted this, by viewing most foreign students as potential terrorists and blocking their entry or seriously harassing them. I suppose this might be considered good news for educational institutions in the rest of the world, but it's not good news to people who just want good education to be available to anyone that wants it (and can pay for it
Here in the Boston area, people are generally aware that education is now the most important local "industry". There are around 250,000 college-level students in the metro area (and most of them are arriving this weekend, making for some major traffic problems
If you're buying a laser printer with a view to Linux compatability in the future, look for one which supports Postscript. All this talk of "Drivers this... drivers that...." - it's cobblers. Postscript has been a perfectly good language for printers for something like 20 years, and postscript printers don't command anything like the same premium that they did 10 years ago.
;-)
What I'm looking for is a printer (or printer driver or whatever) that correctly prints files encoded as UTF-8. I have a bunch of multi-language postscript files that I can show contain the right text. I just edit them with vi(m) on my Mac or linux boxes, and you can see the Chinese or Arabic or whatever there on the screen. But when I send the files to any of the available printers, the Chinese or Arabic (or whatever) text comes out as Latin1 gibberish.
Anyone know of any printers and/or printing software that gets this right? Can you print files that are a mixture of English, Russian, and Japanese, for example? If so, how do you do it?
Presumably people in Asia have printers that can print their native languages. But so far, I haven't found out how to get it to work here in the US. Of course, "Customer Support" people around here don't find it interesting, because why would anyone want to use anything but English? OK, maybe in Spanish or French, and those do work. (But sometimes I wish I could get them to move their support to Asia.
It's an old story. People in power find a way to prevent you and me from profiting from our own creations, by controlling the distribution channels.
The music recording industry is the poster child for this. Until recently, selling recordings required expensive production and distribution facilities. The owners of these facilities could say "You want people to hear and buy your music? Sign this contract and we'll make it happen." You might as well sign the contract, because it's the same for all the other distributors. And in the fine print, it says that you assign the copyright to the recording company. The result is that musicians can make a million-selling album and make no money from it at all. This is because the recording companies can say "If you want your stuff distributed, you must give it to us."
Almost all work "for hire" to corporations are of this nature. If you want to be paid, you have to assign ownership to your employer. If you're a university researcher, and you want to be paid, you usually must assign copyright and patents to the university. Unless, that is, you got funding on your own, in which case you must assign copyright and patents to the funder. And if you want your results published, almost all academic publishers have historically required that you assign the copyright to them. "If you want your stuff distributed, you must give it to us."
Now we have a new means of distribution, the Internet. That promised to give us a cheap distribution mechanism that wasn't controlled by the distributors. But For most of us, to use the Internet requires going through something called an "ISP". Those organizations, usually private companies, have a chokehold on your path to the Internet. Early on in the commercialization of the Internet, the ISPs started to realize what they had. Thus, a few years ago, we read the stories of msn.com (owned by Microsoft) using things from customers' web sites commercially. They mostly extracted images and used them in advertising. When customers discovered images of their children being used in ads, they understandably got upset. And msn.com pointed to the clause in the contract saying that any customer files stored on msn.com machines became the property of msn.com. After a bit of adverse publicity, the "gave in", in the sense that they publicly announced that they wouldn't do this again. But this was like any corporate promise: It was PR to mollify the current crowd of upset customers. After a while, people started noticing that that clause was still in the TOS. And it can be summarized as "If you want your stuff distributed, you must give it to us. You can't go to the competition, because their contracts say the same thing."
So this story is nothing new. ISPs are more and more realizing that they have a chokehold on customers' channel to the rest of the world. Most people have only one ISP, which is a legal monopoly. Even when there are two, they can easily make sure that their contracts are identical. Like various monopolists/oligopolists of the comm channels before them, they can say "If you want your stuff distributed, you must give it to us. You can't go to the competition, because their contracts say the same thing."
It now seems like google, the "Do no evil" company, has realized the same thing. They can provide customers useful tools that inprove people's access to the Internet. And they can hide "If you want your stuff distributed, you must give it to us" in their contract. You can't go to the competition, because their contracts say the same thing.
The only way around this is regulation that denies the controllers of the Internet any ownership of things that pass through their machines. But this sort of regulation has never been effective for any past distribution system. There's no reason to expect that it will be effective for the Internet.
So much for your rights to your own creations. Get used to it; it's the future. Just like the past.
If you dig around in the archives of a century ago, you'll find that this is essentially the same situation as with telephone and electrical service in rural areas.
The telephone story is the most comparable, obviously. 100 years ago, there were two sorts of problems. One was that you couldn't get phone service out in the sticks, because the phone companies didn't think it was profitable. Meanwhile, in the bigger cities there were often several competing phone companies, but you could usually only call someone who subscribed to the same phone company as yours. The ISPs haven't implemented this last problem, because the Internet software came with compatibility builtin. But they're working on it, and if the "Net Neutrality" goes the way they want, you'll find that you won't be able to see web sites of a lot of companies that your ISP doesn't like, especially their competitors, but more generally any sites on other ISPs' lines.
With electricity, the story was simpler. The electric companies didn't build out into the countryside, because it wasn't profitable enough.
Both of these effectively ended when the government stepped in and imposed regulations. The phone companies were told that they'd provide rural service and interoperability, or they wouldn't sell any service at all. This was done somewhat less with electricity, mostly because the Rural Electrification project discovered that they could create locally-owned electrical co-ops that could do the job. They did need regulation forcing the electrical companies to sell electricity to these "competitors" at a reasonable price, but the urban electrical companies weren't always forced to build out into rural areas.
So in effect, universal telephone and electrical service would never have happened without the government regulators stepping in and decreeing that the services be provided to everyone. There's no historical reason to expect universal Internet access to happen in any other way. The comm companies naturally want to sell to only the most lucrative market, and can't be bothered to deal with those rural yokels. If the rural folks want Internet service, they're just going to have to push for the same sort of solutions that gave them phones and electricity a century ago.
The best idea might be the rural co-op, as this would give them a semblance of control over their own service. Of course, the big comm companies are already fighting this, by bribing legislatures to ban anything that smells like government-provided Internet. But unless people start agitating for such things, nothing is going to happen any time soon.
I live in rural Ontario, Canada on a farm. ... it's maybe $70/month when you factor everything in, but for that price I have a nominal 3Mbps/512kbps connection with a static IP, and no bandwidth caps or restrictions.
Well, I live in a fairly densely populated suburb of Boston, and the best we can do here for a static IP and no restrictions is $100 a month for a speakeasy DSL line that delivers 1.5/320 MB.
We actually have several providers, but once Verizon succeeds at persuading the FCC to "deregulate" us, speakeasy will be kicked out, and currently Verizon charges $200/month for a comparable line. We also have Comcast, but the last I checked, they wouldn't do static IP or promise not to block ports on a "residential" line for any price. Both Verizon and Comcast are locally documented to also block stuff like Skype packets, though their PR people look very innocent when claiming that they would never do such a thing.
Sounds like Canadian rules are a lot better for the customers than the rules around here.
What if the timeline were reversed, and we were moving from online apps to the desktop. ... 'You can image the advertising push. "Now control your own data!" "Faster processing power now." "Cheaper!" "Everything at your fingertips." "No need to worry about network outages. ..."
I you look back to the early days of "minicomputers" and "desktop computers" around 1980, you'll see a lot of exactly the sort of arguments like the above. What was happening all over was that users of central time-shared "mainframe" computers were using whatever departmental funds they could to buy their own small computers, so they could abandon the central mainframes. The DP people were horrified, of course, and argued that the central machine provided much more cost-effective processing than a comparable flock of all those little machine, and the little ones could hardly communicate with each other.
But the users were fighting frustration at the difficulty of getting anything done on a big machine controlled by a different department. Their argument was that with their own small machine, they could install the software that they needed, without getting the approval of the bean counters, and the little machine would do what was needed when it was needed, not at the whims of the DP people. Yes, the little machine would eventually cost more, but they would answer to their users' needs, not to a remote disinterested bureaucracy's needs.
The current argument for central servers is really just a replay of this. And it'll mostly fail for the same reasons. A big central server will be controlled by a bureaucracy that has its own motives and needs, and won't be responsive to the lowly users' needs. Smart users will find ways of going around this and using a local machine for the things that the server just can't be made to do right. Management decrees won't work any better than they did with mainframes, because pressure to get your job done will always override pressure to hand your job over to the mercies of the remote central bureaucracy.
Here and there, a few DP (or IT or ??) departments will do a good job and really support their users. They will be used as evidence "proving" how good a central server can be. But most central servers won't be run by such departments, so most users will resist such centralization whenever they can.
Actually, the LoC is slowly putting stuff online. Unfortunately, it's mostly in the form of scans, since they don't have the personnel to do a proper job of transcribing. There is OCR software that (mostly) works for text data, but nothing much that works for music notation. This is a lot harder, because all the symbols have these lines going through them.
;-)
If you dig around in loc.gov you can find scans of all sorts of old sheet music. But it'll take you a while to find things, especially if you're looking like a musician would want to.
Search for "a waltz in E minor with an average of 3 or more melody notes per measure". Yeah, right; OCR software is gonna do that for you.
Why can't Microsoft compete without buying the outcome of the game? Are their products that poor?
Yeah, mostly, but that's irrelevant. They do have a few good products, but that's also irrelevant to sales.
Microsoft's entire history, and IBM's for the previous decodes, demonstrates quite well that sales in any computer-related field are determined almost entirely by marketing budget. Quality is nice, but it doesn't add materially to sales, so if you have the marketing clout, there's no financial reason to also invest heavily in quality.
Sorry to break the news to you. The best product doesn't win. The best-marketed product wins.
There's no (financial) reason that MS should care whether OOXML is good or bad. Their primary concern is that people use it, and this only requires that it be minimally usable. Investing what is for them a small amount to get their encoding declared a "standard" is just a (standard;-) marketing approach, and it would be puzzling if they didn't do it.
Turn that around: How do YOU sleep at night knowing that 'making available' a song that you don't own could wipe out your own savings? It's such a little thing, and SO easy to avoid... And yet, you do it anyhow.
Easy to avoid? People have been pointing out that we are rapidly approaching the day when, if you walk down a sidewalk whistling a tune, you'll be arrested and charged with unlicensed performance of a copyrighted work of music.
Fact is that the only practical way to avoid this now is to never say or do anything at all in public (which includes on the Internet). I've tested this a few times by asking a simple question: Suppose I have a tune in my head, and I'd like to discover whether it's something I "composed" myself or is a tune whose copyright is owned by someone. How do I do this?
I have asked reps of a couple of music publishers. Their answer, apparently said with a straight face as far as I can tell, is that I should buy a copy of everything they've published and search through it all. Of course, this only works with that one publisher; to actually answer the question, I'd have to buy a copy of every work of music ever published by anyone and search them all.
There is something of a shortcut. Here in the US, the Library of Congress (LoC) has copies of most of what has been published in the country. I could go to Washington and spend a few years searching through their archives. Then I could do the same in all other countries. This would only take a few lifetimes, not the thousands of lifetimes that the "buy and search everything" approach would take. But still, there's a certain limited practicality here.
Fact is, the only way I can determine in my lifetime whether that tune in my head is copyrighted is to publicly perform it, preferably in a recording, and wait to see whether anyone sues me.
Actually, there is a less reliable but more practical way that a number of musicians have been using: Put a copy online (either as music notation or a recording), accompanied by a note saying that you haven't determined whether it's copyrighted, and if anyone knows who owns the tune, send a message to <email-addr>. This isn't guaranteed to protect you, because the owner might be a bastard who will sue you for even this transgression.
And it still has the problem that, in practice, you get mostly copyright claims that turn out to be bogus. Publishers regularly claim copyright on music that's centuries old. If you can show this, they'll slink of to look for another victim. And sometimes this happens, because what happens is that someone sees your note and sends a message saying "That was published by So-And-So in London in 1793 as <title> in <book>." If you know this, you can use it as a weapon against the publisher.
But it's all very unreliable, and depends on the good will (or reasonable lawyers) of corporations, in addition to help from other musicians who stumble across your stuff. In general, there's no way to know that a random public utterance or idly whistling a tune won't be a copyright violation. The only really safe strategy is to be utterly silent in public. On the Internet, this includes learning enough about your computer's innards to guarantee that it isn't exposing any file to outsiders.
So basically, the enlightenment was due to Gutenberg?
No; it didn't have a single cause. Gutenberg certainly made a major contribution and helped to enable the Enlightenment. But his technical advance potentially made publication easier and cheaper for everyone in the world, not just in Europe. The really important advance wasn't in the hardware, but rather in the "software", i.e., in the social structure that developed the concept of open publication of scientific results. This could have been done anywhere, and could have happened before Gutenberg. It happened in western Europe, whose scientists took advantage of the improved publishing technology and put it at the core of the scientific enterprise. In much of the rest of the world, publishing was (and still is) controlled by the ruling classes, i.e., by politicians, so they lagged behind Europe.
There are other social innovations that were needed. I read an interesting comment a few years back, by a French researcher who explained why he always published in English. It had nothing to do with the size of the audience. His explanation was that doing good scientific work often requires that you invent terminology, and sometimes subtle differences in terminology can be the difference between a successful hypothesis and a failure. In French, there's a national language bureau that has the legal power to decide how the French language may be used, and it's full of people with no understanding of his scientific specialty. So he can't freely invent new terminology in French and use it in his publications. Well, he sorta can, but doing so risks legal harassment and possible revisions that would destroy the scientific usefulness of his work. The English language is an insane free-for-all without any official, centrally-controlled, legally-enforced rules. So in English, he and his colleagues can work out their own terminology without any official harassment. Once they think they've got it right, they can "borrow" the terminology into French, of course, but even then they sometimes get harassment for using Englishisms in French. So he does a good Gallic shrug, and publishes in English.
It's interesting to think about. The scientific revolution did depend on a number of independent developments. Some of these are negative, like the lack of official rules for the English language. Publishing technology is at the core of a lot of science. Right now, we're going through the pains of switching to new technology (the Internet) that's orders of magnitude faster and cheaper than what Gutenberg gave us. The societies that do this right will be the scientific leaders in a few decades. This might not be an English-speaking society, if the current "Intellectual Property" debate goes the way it has been going in the English-speaking parts of the world.
Actualy, the reason that the Internet is sucessfull is that there is a complete implementation of the main protocols available (BSD licensed) to anyone that want to use it.
;-) I downloaded the RFC, printed it out, and took it home for bedtime reading. Bright and early (ok; 8:30 am) the next morning, I started coding. By noon, I had a working version. We installed it in a few machines, ran our test suite, and everything worked. Less that 24 hours from the "Let's do it" decision to having a usable implementation.
;-)
Well, I wouldn't agree that this is the reason. It certainly helped get things rolling. But I'd argue that if the geeks at Berserkeley hadn't done the job, it would have been done by others. And it didn't have to be done at one place. The division of the design into layers and different specialized protocols makes it easy for independent developers to work on the pieces independently. But having it all in a single, freely downloadable package did a lot to encourage people to use it. At the time, our current Internet didn't exist, so there was an advantage to having it done by a group of people at a single university.
But all this modular design was true of, for example, the OSI protocols. On several projects, I saw first hand why OSI didn't take off. The OSI docs weren't published online (at first); you had to order printed copies. So I'd fill out a purchase order, hand it to the secretaries, and wait for the needed signatures. Finally it would get approved; they'd mail it off, and I'd wait some more. Finally, weeks or months after starting, I'd have the spec that I needed. In the meantime, to keep myself busy while waiting, I'd download a few RFCs, fire off questions to mailing lists, and work on some prototype code that I'd carefully keep modular with the idea of retargeting it to the OSI protocols. By the time the OSI docs arrived, I'd have a working version of what we wanted running on IP. You can guess what the group usually ended up using.
It didn't hurt that the Internet specs were usually simpler than the OSI specs. Thus, on one project (at Digital in the early 90s), we were getting frustrated with the problems getting SLIP to work over our modems. One day I decided to try a reimplementation. ("How hard can it be?"
Not that the IP-based stuff was always that easy. Some months later, we decided to give PPP a try. It was somewhat new at the time, and we didn't have a version yet on our machines. So I again downloaded an RFC, and went to work. This time, working on it about half time, it took me about a week, because PPP is a bit more complex than SLIP. We liked it better, because it wasn't prone to failure due to the assorted insanities built into all commercial modems.
Funny thing is that most management doesn't get any of this. I've been to a number of interviews where they ask about driver or protocol experience. I mention a few cases like the above, and invariably the reaction is "That's only a few days' experience; we're looking for someone with several years experience." So I don't get the job, and sometimes they make it clear that they think I'm lying about the time it took. I've even told them that I can make it take several years, if that's a project requirement, but somehow that doesn't convince them. (Of course, I only say that after it's clear that I'm not getting the job.
[N]obody should be surprised that Bush appointees (here and in Iraq) have consistently proved themselves incompetent.
;-)
Well, that depends on what you think they're trying to do. Others have concluded that many of those supposed "incompetents" are actually highly competent, and have succeeded admirably at their goals.
For example, consider that Gonzales may have been appointed Attorney General specifically to block the Justice Department's natural tendency to investigate and prosecute corruption. Various people have pointed out that many of those fired attorneys had something in common: They were investigating corruption in government contracting. By firing them, Gonzales terminated those investigations (and got the message across to others who might be tempted to start their own investigations). The result has been things like the billions of dollars that have disappeared without a trace in Iraq. Companies owned by the Bushes and their associates, such as Haliburton's subsidiaries, have profited tremendously from the lack of Justice Dept attention.
If you surmise that this was the intended result, then appointing Gonzales was a very competent decision by Bush, and Gonzales was in fact quite good at doing his (actual) job.
Their immediate problem might be that their overreaching has gotten to the point that even their so-called "conservative" supporters are starting to get offended by the audacity of the corruption, so they had to ease Gonzales out. Very conveniently, they waited until Congress wasn't in session, so Bush can replace him with a recess appointment and avoid congressional oversight once again. And this could be viewed as a very competent action on their part, demonstrating that they know how to exploit the fine print to relieve Congress of any bothersome need to get involved in presidential appointments.
(Hmmm; am I cynical enough yet?
[T]here are actual people in the world, who PAY for some research through their taxes, and then does not want the ownership of it through public domain.
I've read quite a number of histories of science in which the authors point out that the Western "scientific revolution" during the last few centuries has nothing to do with discovering "the scientific method". Scientific methods have been independently discovered in nearly all societies, going back into prehistory. One example I ran across recently was the comment that from what we know of their methods, the North American Indians' "medicine men" had better medicine than Europe did until sometime in the 1800s. The reason was that the medicine men actually had better scientific methodology than European doctors did. But then things changed
So what caused the big advances in Western science in the past few centuries? The historians answer to this is simple: open publication. In all other societies, such knowledge has almost always been strictly controlled by small "guilds". The knowledge was secret, and discoveries were usually not even shared with colleagues outside the discoverer's immediate circle of professionals. This meant that everything had to be rediscovered over and over again. You could only learn what your mentors knew, and you could only build on what they passed on to you, because everyone else's knowledge was unavailable to you.
But a few hundred years ago, some researchers in Europe developed a curious new approach: They published their discoveries openly, making them available for others to read, use, and build on. This led to the explosive growth of knowledge that we're familiar with.
In most of the West, such open publication is historically done only by government researchers. Before the 20th century, this meant the few idle rich such as Isaac Newton, who had brains and curiosity. So it took a while to really get going. But in the 1900s, various governments slowly got it through their thick skulls that funding research was one of the things that was building other countries' economies (and militaries), so maybe they should be funding research too. Then things really started ramping up.
But there is still one major drag on scientific advance: A lot of funding still goes into "private" (i.e., corporate) research. This is, scientifically speaking, usually a dead end, because the results of such research is kept private, and as of with the guilds of old, it isn't available for others to build on. The legal system cooperates in this, by prosecuting people who get access to private research results and try to build on it. In recent years, this has been happening more and more in the US, as the corporate world consolidates its control over the government and determines how most research is funded. Some universities also contribute to the problem, by claiming ownership of research results when they can and keeping it secret (or usable only under high license fees). I've read a few predictions that useful American research may be ending now, as the corporate world takes most of it private. And it's curious to see the publishers of scientific journals jumping in to block the advent of cheap open publication via the Internet.
Anyway, at least according to these histories, we should be supporting the open-publication people, because they're the ones pushing for more of what really made Western science the success it has been. If we really want Science to continue to improve our lives, we should be pushing for full access to all research results.
(And people have pointed out that the Internet is an example of the same phenomenon. It hasn't succeeded because it's such a great network. The IP/UDP/TCP/DNS/SMTP/HTTP/... protocols aren't all that good, actually. They're good enough to do the job, but anyone who knows them can tell you lots of ways to improve them. The reason for the Internet's success is mostly that all the specs have been published openly from the start. Anyone can download them for free, read them and implement them without legal restrictions. This gave the Internet a huge advantage over other privately-developed protocols that were often technically better but weren't available to any developer that was curious.)
You do _not_ talk about Linux crashing on Slashdot.
Heh. One question that came to mind immediately was: Did those people actually see linux crash, or did they see linux rebooting. These are two very different things, of course, but I've found that even experienced users can be rather sloppy about such insignificant details.
I'm sure that most people who've taken commercial flights have noticed things like the cabin lights all flickering at times, especially during takeoff. No big deal for lights, but this will "crash" pretty much any computer using that power system. And the crew obviously has control of the lights; why would you be surprised if they could also control all the computers? Just as the control panel has buttons to turn all the lights on and off, I'd expect that you could reboot all the seatback computers from the control panel. You could also override what they're doing and show some particular video clip.
I'd think that the ability to tell all these little computers to do something simultaneously would be a good selling point. And considering how the power-supply systems work in commercial airliners, I'd expect that it would be normal to do a net reboot of the little buggers frequently. The ability to easily do such things from a central control system would be a major selling point.
So, rather than us _not_ talking about linux "crashing", I'd think we'd be interested in information about how these seatback computers are configured and how they're managed by the crew.
Not that I expect to see much real information here. And I'd guess that much of the airlines' control software is proprietary, for "security" (by obscurity) reasons.
I'd imagine that one of the driving forces here is that listening to music is now an outdoor activity.
Nah; that can't explain it. After all, if all the recordings were at half the current level, then you'd just turn the volume dial to get the same loudness.
To understand the escalation, it works better to think of someone listening to their Walkman or boombox or iPod, which is set to "shuffle" (or to a music radio station, which is similar). What the recording companies want is for the track you're listening to now to be slightly louder than the one you just finished listening to (and to the one you're going to hear next), but not enough louder that you turn down the volume.
This leads them to sample the competition, and set their tracks to a bit louder than the average. Over time, they all get louder, until they're all at the max intensity with no dynamic range. Of course, this will be indistinguishable from near silence, since listeners will have the volume at minimum.
Maybe you should just set yours there now.
I could swear this program has been "killed" twice, and by "killed" i mean the government's definition: proclaiming a project discontinued while continuing it under a new name.
This sort of thing has been reported for US government agencies for decades. Back when all the data was all in paper archives, there were lots of reports of agencies that "obeyed" orders to destroy them by first running them through copiers, transporting the copies to some other site, and then destroying the originals. I recall during the Vietnam War, when the DoD was ordered to destroy their records of spying on anti-war groups, and the DoD announced that it had done so. But some time later, reporters published reports that at least five backup copies had been located at five different sites.
No matter how dumb you think government (or corporate) employees are, fact is that most of them are smart enough to figure out this ruse.
There's also the reverse version that's been making the news: We're now reading reports of managers (government and corporate) finding the contents of their email used against them in court, when they thought that the email messages had been deleted. The messages had been deleted from disk, of course, but the backups could still be read. Oops!
When you read stories like this, you should always ask yourself "How many copies were made?" That will put you into the proper cynical perspective, which you'll later remember when it turns out you were right to ask.
Thats crazy - go change ISP.
Good advice, for those who can. But here in the US, most people have exactly one ISP to choose from. If you don't like it, you can move somewhere else where there's a different ISP.
Some people do have two ISPs. But if you investigate, you discover that they both have almost exactly the same policies and restrictions (and prices), so it won't do you a bit of good to change.
Isn't capitalism wonderful?
Define 'server'.
;-)
A couple of years back, while discussing such things with our then ISP (RCN), the support guy made it clear that to them, "Internet Service" covers web browsing and email. End of story. Nothing else was supported, and by supported they meant permitted. Oh, and by email they meant access via their own email servers; no other email was supported. All incoming and outgoing email must be stored on their own servers, and no, they won't tell you why they want this. We also have Comcast as an alternative, but I suppose most people here know what sort of alternative that is.
When speakeasy became available in our neighborhood, we switched. It was a bit slower than the "blinding fast" service advertised by RCN, but they allowed (and promise to always allow) unblocked Internet access. They actively encourage things like servers. In fact, they even encourage redistributing your service to neighbors, and they'll even handle the billing for you, so you effectively become a local distributor.
Too bad they aren't allowed access to most potential customers. Maybe we need to be pushing for regulations that only permit real IP service, i.e., just deliver the packets at whatever rate I've paid for.
(The really funny part of the story is that we looked into having Verizon, our phone monopoly, supply DSL service. They couldn't do it; we're too far away. So we got speakeasy DSL instead, over the same Verizon line. We've been happy with them.