Spend... one hour buying some wireless networking gear. Presto! Everybody's happy!
You'd better hope your boss doesn't find out that "ethernet" was originally called that because it was designed to work "over the ether", i.e., it was supposed to be wireless. The wired form was just a stopgap until the wireless transceivers became available (and could legally be used in the US without applying for an FCC broadcast license for every node).
Of course, a boss as clueless as you describe probably wouldn't understand the humor in an explanation like this.
To me, this is an example of catastrophically bad platform design.
Well, as a unix programmer for around 30 years now, my main gripe is people who criticise the conciseness and power of the design, and try to "fix" it by kludging in "safeguards".
One of the main results, from a programmer's viewpoint, is that there's no reliable way to unconditionally remove a file. People keep discovering that files can be removed accidentally, and "protecting" users from such mistakes. This means that my script isn't portable to yet another system. Temp files get left behind ("like little turds all over the disk", as a friend once described it), and code fails because the file that "was just deleted" is still there, and re-uses the old file instead of making a new one. And there's just no way to learn how to correct for this on every new release of every system that comes along.
An especially funny case: On the Mac that I'm using to type this, there's a file inside my.Trash that can't be deleted. It got there because I rsync'd some files over from another computer whose file system was "8-bit clean" and contained file names in German, Swedish, Finnish and a few other languages that use non-English 8-bit letters. These file names got turned to gibberish by the OSX file system. And the software is determined to "protect" me by not letting me delete this file. None of the dozens of tricks I know to defeat "safe" rm commands seem to works. Not even using a hex dump of the directory and typing the file's name in as a char[] array in hex in a C program and calling unlink() works. Queries on Mac newsgroups basically get jeers because I was stupid enough to use rsync with OSX (and I conclude from this that they can't solve the problem either;-).
Life would be a lot easier for us programmers if there were a command that would simply and silently remove a file, and fail only if I don't have write permission on the directory. (Hint: It should not fail if the file isn't there. The file not existing what I want; it's not an error.;-)
Life was much easier when I had unix systems that would do exactly what I told it to do. Now I have to figure out how defeat the "user-friendly" safeguards that are different on every unix system, just to do something as simple as make sure a file isn't there.
OTOH, defeating the growing "user-friendly" versions of unix software was one of the primary motivations for the perl language, so I guess that something good has come out of it all. But my attempts in perl on OSX to remove that file from.Trash have also failed miserably...
No...regular internet companies don't save any of your mail.
Unless, of course, some government agency tells them to. And, in the US, they must keep that fact secret from you.
Also, note that setting up your own mail server (if your ISP doesn't block the email ports, as many do) doesn't protect you. If your IP connection is through a single ISP, they can quite easily collect all packets to or from you with a given TCP port number, and assemble the messages at their leisure. Or the government agency can do so. Or an employee of a competitor who has bribed the right person at your ISP.
The only thing really protecting you is that no ISP wants to buy the storage that it would take to do this for all customers. They'd be bankrupt in a month if they tried it. But under threat from your government, or reward from your competitor, they just might find the resources to save the packets to/from a few select customers.
The only real way to get privacy of course, is end-to-end encryption. But both ends have to have a motive to do this.
I have occasionally wondered how many governments are getting curious about the fact that much traffic to/from unix-type machines now uses ssh/ssl. The main reason, of course, is so that you can send passwords without giving them away to anyone with a network sniffer. Telnet, rlogin and ftp have been mostly supplanted by ssh and scp for this reason. But it turns out that it's usually just as easy to encrypt an entire conversation, rather than just the passwords. And then you don't slip up and accidentally send a password in the clear.
Still, it's rather naive to underestimate the "chance" that your email is being intercepted and cached by someone. Most isn't being saved anywhere. But it's not all that difficult to a motivated party, especially one who can use taxpayer money, to implement this. Unless your email is encrypted, you should always assume that random unknown people are reading it.
I've found some surprising old messages of mine simply by googling for my name or an old email address.
Google is pretty close to having a monopoly on search engine services. Remember, you don't have to be the only provider to have a monopoly, you just have to weild "monopoly power", that is the ability to control the market, and I think Google is getting damn close to that.
Maybe, but as one of several hundred people running a small, highly-specialized search site for a type of technical data (details of which don't matter here), I can say that I have yet to see any signs of google trying to wield such supposed power.
While we all know the value of google, there's still the general problem that keyword search of text just isn't very good for finding a lot of technical data. Suppose, for example, that you're doing DNA research, and want to locate sites that deal with a particular string of DNA. Ask google about "CGA TCC CAT TGG TGC" and see how many responses you get.
There is a fair amount of research going on for other kinds of search, plus of course the research on making computers "understand" text iin order to give matches that are more relevant than is possible with keyword matching. Google is doing some of this, as are the other big search sites. But so far, there doesn't seem to be any pressure on us independent sites to stop our research or sell out to the big guys and do it their way. (Some of us have had feelers out to google to see if they're interested in hiring us; that doesn't seem to be getting much of a response so far.;-)
With google, there are encouraging signs of the opposite. Thus, with google maps, they are actively encouraging people with various kinds of databases to correlate them with google's maps. Some of the sites leveraging google's maps are even commercial sites, and I haven't yet heard that google is trying to discourage them.
Of course, they could be just waiting for a better opportunity, when lots of small sites are dependent on google's maps, and they they'll pounce. Now that google is a public company, as others have pointed out, they just might suddenly decide that a major takeover campaign is in their stockholders' best interests. So we should definitely keep our eyes open. They could suddenly decide to copy Microsoft's move to kill Netscape, and set back search research for years in the process by killing or buying out all the small, specialized search sites.
But if there are signs of this sort of evil from google, I haven't read about them. Anyone have more information? Some of us would be very interested in any symptoms or clues about the future behavior of this 500-pound gorilla of searching.
The skill of using a piece of word processing software is completely trivial compared to the skill of learning how to write and command the language.
Indeed, and I've read any number of comments along this line from professional writers, journalists, and others who make their living producing text. Usually they say something like "I produce words. Then I give them to an editor to format for publication. I have no need to learn to use a word processor; that's the job of the people working for my editor." They often comment that fancy formatting is pointless for them anyway, because the publisher will just throw out all their formatting and redo it. So they just send plain text, with at most a bit of bold or italic emphasis here and there (which will be edited).
Teaching a word processor is practical in the same sense that teaching typing has long been practical, and the main value is the same: learning to use the keyboard. But it's worth maybe a couple of months at the high-school level. The rest of those years of "English" classes should be to teach use of the language.
Teach someone proper thinking skills, put him down in front of a computer,...
Teach proper thinking skills? I had a couple of teachers who tried that. All the religious, and half the political folks in the town came down on them hard. Thinking skills are something very dangerous for your typical teacher's professional life. Fat chance of that every happening.
My favorite happened while I was using a sig that said something like:
Unix perversity of the week:
find . '*.o' -exec rm {} ';' Don't try this at $HOME
This got appended to messages that I sent to several mailing lists. Of course, I got several very nasty letters from people who didn't understand what was wrong with the command, and tried it. At $HOME, of course.
You can probably imagine how I learned what this command does. It's one of my favorite examples of bad UI design.
(No; don't try it at $HOME. Try "mkdir foo; cd foo" and link in a few random files from some other directories, to see what it does with them.;-)
Yeah; I've heard there's something new called FTP that lets you download whole files over the internet. Anyone know anything about it? Imagine using this to download, say, music files. Then, instead of carrying around a case of CDs, you could just have the music on your disk. Imagine if someone made a pocket-size music player with a disk that holds a few gigabytes of music, and you could get the music from the internet...
Just as long as they don't claim something like Tango or Polka or Waltz. I'd hate to be dragged off the dance floor and charged with copyright infringement, right there in front of my partner.
I learned Mambo in a class 25 years ago, but I've never had an opportunity to use it siince then, so it's OK if they lay claim to that name.
My favorite cosmology has long been the one in which every particle in this universe is a data structure inside a computer in the real universe. That computer is running the simulation that is our universe.
In this model, the basic unit of our reality is a bit of memory in the real universe. Elementary particles are a second-order concept, a data structure made of a collection of bits. Time itself is quantized, and the quantum is the time it takes the real computer to calculate the "next" state of all the particles in our universe.
It can be fun to argue this cosmology. But it has gotten somewhat less fun since the Matrix movies came out. It's no longer such a radical concept.
In such a universe, miracles are easy to explain. Something has gone wrong, so the simulation is stopped and restored from backup. A bit of editing is done, and the simulation is restarted.
Maybe this is what the Intelligent Design people are really talking about...
Hell, American pronunciations probably sound funny to all Commonwealth states no matter what regional dialect is used.
Not really. Due to American dominance in the entertainment industry, English-speaking people everywhere are very familiar with the major American dialects. They don't sound funny, just American.
Similarly, few Americans would consider British accents (at least RP and others commonly heard) as funny. Cute, quaint, charming perhaps, but not funny. We've heard a lot of Tony Blair lately, and he doesn't sound funny; he just sounds English.
Except for Monty Python, of course; they're funny. Or Tracey Ullman. But she's fluent in so many accents that you have to keep reminding yourself that she's actually a Brit.
They propagate at the speed of light in all reference frames,...
Yup. But not in all materials. TFA blew it for me with the comment:
Light signals race down the information superhighway at about 186,000 miles per second.
Uh, no it doesn't. In a glass fibre, light travels quite a bit slower than that. It only travels at that speed in a vacuum. In glass, light travels significantly slower than c. And parts of the information superhighway is made up of electrons ambling down a wire at speeds very much below c. (Then there's the speeds of the bongo-drum and avian-carrier links.;-)
This comment pretty much made it pointless to accept anything else in the article as fact.
All very true, and this illustrates why such a gadget shouldn't deal with a voice system like E911. Its comm stuff should do wireless IP, the software should connect directly to your hospital's computers, and the computers can notify whoever needs to be notified.
We've had a lot of discussion about how hopeless the wi-fi system is for giving coverage. But my pocket "smartphone" now has full-time, always-on IP connectivity, and the cell-phone system has much better coverage than wi-fi ever can.
Perhaps such medical monitors can be the excuse we need to implement universal wireless net access. And maybe force the recalcitrant comm companies to implement deals that will make everyone's monitor work regardless of which comm company provides service at each remote location, without first requiring minutes to negotiate yet another $10/day "contract" because the service is via a different company than the relay in the previous block.
And, once we've got that, we can include a tiny microphone and speaker, and use VoIP to add "phone calls" as a feature. And we can look forward to the day when young people have no clue what a "phone" used to be or why we use such a word with our pocket/wrist comm/audio/video gadget, whatever it's called by then.
some device that you wore on your wrist that told you the time. More advanced applications could include the date.
Heh. I had a device like that several years ago.
One day, during a spell of hot weather, I noticed that I'd developed a minor rash under it, so I didn't wear it for a couple of weeks, until the rash went away.
Then I noticed that I hadn't missed that wrist display. It seems that in the places I hang out, it's nearly impossible to be out of sight of a time display. If I'm out walking in the woods, well, I have a cell phone in my pocket, and it has a time display. So why carry one around on my wrist?
I haven't missed it.
I've also read that the sale of wrist watches has been declining in the US and Europe for the past 5-10 years. It seems that a lot of people have been reasoning the same way.
OTOH, a compact Dick-Tracy-style wrist communicator could revive the market. Especially if it included a basic health monitor that could display some exercise-related data like heart rate. It's easy enough for such a gadget to display time and date by default.
A couple of years back, I read of a compact wrist cell phone that was being sold in Japan. I wonder what ever happened to it?
Yeah; you're right. The basic problem is that English is the only European language that hasn't had a major spelling reform in the past century or so. Our spelling is a mess, with no reliable connection between spelling and pronunciation.
As a result, there is no good way to transliterate words and names from another alphabet into English "phonetics". English doesn't have a phonetic spelling system. Most English sounds have more than one spelling, and most of our letters have more than one pronunciation.
In addition, Russian sounds are not the same as English. No matter how you spell a Russian word for English-speaking people, they can't get the sounds right. People try, but the only result is several different systems for transliterating Russian to English. This just confuses things more.
It's a mess, and it will probably never be fixed. There is no language authority for English. Other languages have official organizations, usually government departments, to standardize the language and writing systems, but this has never happened with English, and probably never will. I've always felt sorry for people who have to learn English. Maybe our writing system isn't as bad as Chinese, but it's still one of the world's worst.
It's utterly hopeless. You'll just have to accept bad transliterations of Russian words into English, and even worse mispronunciations.
(But then, I've heard some rather weird mispronunciations of American place names from Russians, not to mention Germans and French. We just think it's cute and charming.;-)
Jeez; why are the people here such sloppy spellers?
(Actually, I had to look it up at wikipedia, because I didn't remember whether the '' was hard or soft. Hardly anyone ever gets this right in the English transliterations.;-)
Now to see if/. can actually handle simple UTF8 encodings. It looks fine on my screen, and cut-and-paste to several other browser windows here on my Mac PB seems to work ok. But there is still software lurking out there that can't handle anything but ASCII.
The quoted string in the first paragraph above should contain his name spelled correctly; i.e., in the Russian version of the Cyrillic alphabet. No spelling using the Roman alphabet can really be correct, of course.
Yet another vote for overly-picky corect spelings...
Well, yes. But an important side effect is generating and testing explanations of those facts. With an emphasis on testing, which usually means you have to go out and collect more facts (usually called observations, or just data). So as a scientist, most of your life will always be collecting the facts that you need.
One of the more pointed explanations that Stephen Jay Gould made about evolution was to point out that Darwin didn't show that evolution had happened. By the time Darwin was born, this was an accepted fact among scientists. All these fossils had been dug up, and they showed a clear set of changes with time. Geologists got involved, and concurred with the whole thing. Nobody who actually studied the fossil data questioned this. But the observed evolution was very non-random, and a good explanation was lacking.
What Darwin did was to present a theory that explained why the fossil record showed certain kinds of evolutionary change and not others. And, most important, his theory was testable. Incidentally, it offended religious people, because it didn't need an intelligent guiding hand. Scientists immediately jumped all over it, of course, and managed to collect a great deal more data that kept coming up consistent with Darwin's theory. Religious people also jumped all over it, but they didn't understand scientific testing methods, so they couldn't disprove anything, or even understand why they were expected to do so.
And, of course, lots of philosopher types have pointed out that none of this ever dealt with proof or truth. Rather, people had simply failed to find data that disproved Darwin's theory. This sort of double negative is standard scientific method, and is where the term valid comes in. That just means a theory that can successfully explain all the observed data despite many attempts to shoot it down. It doesn't mean truth, because we might have several valid theories competing at once, and new facts might pop up at any time that would shoot down any of them. A valid theory is only tentativily accepted, because it has passed a number of tests and hasn't (yet) failed any. See Karl Popper for lots more words on this topic.
Similarly, Einstein made some rather outrageous predictions about the universe's behavior just a century ago. This was in an attempt to find a theory that explained some rather outrageous observations (i.e., facts) by other scientists in previous decades. Since then, physicists have repeatedly found new ways to collect data that could disprove some of Einstein's equations. They have repeatedly failed; his equations always predict results that are within the error bars of the observations. Maybe next month someone will find an exception, but for now, we have to accept Einstein's theories as valid descriptions of our universe.
Now, scientists often carelessly use true for valid, when true should really only be used for facts. A fact can be true or false; i.e. it does or doesn't describe an actual observation; a theory can only be valid or invalid. It's true that evolution has happened on our planet, but Darwin's theory isn't true; it's valid (so far).
Of course, all of this is above the mental capacity of most of the media or the political system (or the religious communities). So we have an ongoing bogus "debate" on such topics.
(Hmmm... Maybe I should preview this one, to make sure that none of my editing has garbled anything.;-)
Unicorn Theory can also be framed in such a way that it is not in conflict with the known facts,...
Indeed, and the followers of the Invisible Pink Unicorn are following the ID story with a great deal of interest. If the ID people succeed at getting their "theory" imposed on schence teachers is some US states, we will see a followup demand that IPU theory also be included in science classes.
There are a number of other such competing theories waiting in the wings...
What is a color if not something that we perceive to be as such?
We might point out that there's a critical difference between "The sky is blue" and "The sky appears blue". The first isn't quite correct, because "blue" isn't actually an accurate description of the sky's spectrum. The second is correct, because it acknowledges that the color depends on the observer's optical equipment.
It would be even better to say "The sky appears blue to the human eye". It has different colors to other animals. Thus, birds have four visual pigments. Three are like ours; the fourth has peak sensitivity around the violet end of human vision. To birds, the sky would stimulate the blue and violet pigments about equally, and the sky would appear a complex blue/violet/UV blend. An avian interior decorator would have words for those colors. Something similar would happen with a lot of insects, who typically have a V/UV-sensitive pigment but often no red-sensitive pigment.
To get even pickier, we might note that in mammals, birds and insects, there is significant variation in the actual frequency response of the visual pigments. There is also intra-species variation in many species. Humans are one such. When I was in high school back in the 60's, a physics teacher did a lab demo of this. He set up a prism to give a solar spectrum, and it was good enough that he could use a couple of absorbtion lines to calibrate it and label the frequencies. Then he had us go up to the paper and put marks at where we saw the ends of the rainbow-colored bar. The marks were approximately normally-distributed around the 400-nm and 700-nm points. The students were duly impressed by this demo of the variations in their color vision.
He went on to explain that any of us into photography should appreciate this. Different people have different opinions about how good various films and printers reproduce colors. This is partly because their pigments can't match the visual pigments for all of us exactly. How good a picture's colors look is partly determined by how closely the pigments match your visual pigments. Because human eyes vary so much, no printing system using only a few pigments can be accurate for all of us. (In my case, the red ended somewhat past the average for the class, but my violet mark was right at the average.)
But imagine how our pictures (and computer screens) must look to birds. The violets and ultraviolets are missing. It would be like us looking at a picture that has all its blues zeroed out.
A lot of birds and insects have ultraviolet markings, as do the flowers that they pollinate and the fruit that they eat. These markings are invisible to human eyes. Biologists have only recently learned to appreciate this, and a lot of mysterious behavior has become clearer as a result.
In particular, it turns out that a lot of avian navigation can only be understood if you realize that they see ultraviolet and they see polarization. There are situations in which birds are using polarized ultraviolet/violet sky light. If you can't see this yourself, you have difficulty explaining their behavior, because the sky just looks plain "blue" to you.
Info on the topic is easily available online. For avian color vision, google for "avian color vision". Use some photographic or printers' terms to find discussions in those subject areas. What you see through your eyes is only the start of undstanding what color the sky is.
(And I can hear the shouts of "Too much information!";-)
The FIRST RULE while taking to anyone in "tech support" is to swallow all your pride and admit you know nothing.
Yeah; that's my standard approach. You learn more that way. After all, you've called Support because you don't know enough to make something work right. They're there to help you, right? And, hopefully, to educate you a bit in the process. So no matter how much you know, you're ignorant in this situation, and it's best to be honest (to yourself at least) about this.
Why the hell were they even informed you were running Linux on your network anyways? Were you sitting there on the phone and thought it'd be impressive to the support person to mention you were running Linux?
Nope. I didn't volunteer the information until the guy started asking questions about the Airport's Internet connection. I really thought this was irrelevant to diagnosing a Powerbok->airport->printer problem, but I thought I'd let the Support guy make suggestions. Maybe I'd learn something. I answered his questions honestly, which is probably a useful thing in such situations. I didn't know why he was trying to solve the problem by working on the Internet connection, but the obvious thing was to follow his suggestions.
This ended when he first demanded that I reboot the linux firewall. This, IMHO, was so far beyond the pale that I simple answered "No." He got all bent out of shape by my refusal to follow orders, damage my local network, and kill other tasks that were running. When he changed to demanding that the linux boxes be removed from the network, I again simply said "No", and pointed out that I hadn't called about a LAN or Internet problem (and we'd verified that the PB, Airport and linux gateway all had fine Internet access). I suggested getting back to the PB->Airport->printer problem, by isolating them from the LAN. His response was that he couldn't help me as long as I had a linux box on the LAN.
Part of what I was doing was testing Apple's support for a network with mixed equipment. I was doing it on my home network, and I wanted it to work there. But I was also working for people who were interested in various networking equipment on their corporate network. Since I had two PBs, an Airport Extreme, and couple of linux boxes and a Windows box on my home LAN, they wanted data on how well these all worked together. I never mentioned this to any Apple people, of course. But they gave me some very useful information on the subject. I could report that Apple's tech support people were actively hostile to a mixed network like this, and refused to help solve a simple Airport connectivity problem because there was non-Apple equipment on the LAN. For a software-development company that needs a mixed-vendor setup for testing, this is something you want to know, because it's a total deal killer.
Using some of the net-testing stuff on my linux box, I did eventually diagnose the problem myself. In fact, it took less time than I'd wasted talking to Apple Support. The problem was LAN-related. Something I didn't yet know was that the Airport Extreme had a DHCP server enabled by default, and it was handing out addresses that overlapped with the DHCP server on my linux gateway. (I can feel net admins cringing in horror, knowing what's coming.;-)
The printer's IP address was in fact the same as the Windows box's. When I tracked this down, it was just a matter of time until I found the Airport's DHCP stuff, and reconfigured it to not overlap with the gateway's DHCP stuff. (I've looked for Apple tools to diagnose this problem, but haven't yet found them. Log messages on our two PBs weren't helpful, but log messages on the linux box were.)
I don't know whether the Apple Support guy would have found this. His plan was to install the Airport as the Internet gateway, with no LAN other than the wi-fi. The printer would have worked under those conditions. I'd guess that
Depends on your field of study. I've seen a number of geological discussions. That 4-continent definition is included in the wikipedia definition, but few if any geologists would agree. Thus, Africa really is a separate continental plate, which just happens to be bumping up against Eurasia.
Similarly, there are distinct North American and South American plates. The North America plate's southern edge is a bit fuzzy, but it actually ends roughly at the southern edge of Mexico. The isthmus between Mexico and Columbia, "Central America", is really a recent (10 million years) volcanic development, like Cuba and the Antilles. It was produced by plate tectonics, but isn't part of a continental plate.
But this is just for geology. For other fields of study, other definitions make sense. Though it has always been a bit odd to classify Europe and Asia as different continents. This is partly historic confusion caused by the fact that "Asia" used to mean what we now call "Turkey", and what we now call "Asia" wasn't known to Europeans. Somehow "Asia" got extended to all the land to the east of its original definition, and since the fastest way there from Europe was by boat, people decided it must be a separate continent.
It can be hopeless to try to straighten out terminology when it's as confused as this. Especially when people in different fields are using a definition that makes some sort of sense to them, but not to anyone else.
Funny how this confusion keeps coming up, and how few people see the obvious answer.
Similarly, we refer to the Estados Unidos Mexicanos as "Mexico", and the República Federativa do Brasil as "Bra[sz]il". I wonder why people don't object to those shorthands? Maybe because they realize that there is no other country with "Mexico" or "Brazil" in its name. You'd think people would be smart enough to see the same explanation for the "America" shorthand.
Then there's the case of the Bundesrepublik Deutschland. It takes a bit of convoluted linguistic history to explain why we call it "Germany" in English. Most people would fall asleep some time during anything like a detailed explanation.
Anyway, "USA" is also a city in Japan, so using it alone is ambiguous.;--)
I think it's very similar to the way we view roads. Most of them everywhere are "free" in the obvious sense that you don't have to pay a toll to use them.
This doesn't mean that it doesn't cost money to use on a road, of course. Vehicles cost money; fuel costs money; insurance costs money. Some places, you even have to pay extra for a parking spot. Part of your taxes, especially on fuel, go to maintenance of the roads.
Still, it makes sense to say that the roads are "free". You don't hand out money when you use most roads, except for a very few toll roads.
It wouldn't be surprising if the Internet ends up "free" in the same sense. You'll have to buy your comms equipment, and/or pay a monthly rent. But, except in some special spots with special controls, you'll be able to communicate from anywhere without paying a toll to a local ISP. Like the roads, the Net will be accessible if you have equipment able to use it.
Perhaps your monthly rent will be for bandwidth. You pay $N per month, and get B bytes/second of access. Pay $2N and get 2B bytes/sec. Or something like that.
We're nowhere near that yet, of course. Most of our comms are like a road system where you have to stop and pay a toll every few blocks, and each road owner can decide whether your vehicle is allowed on the road, what sort of freight you can carry, etc. for his little stretch of road. But there are signs of change
Actually, a number of science fiction writers have already used this idea as a plot gimmick. Several have considered the effect on society of a guaranteed Right of Communication. The ones I've read have been centuries in the future. Maybe we'll live to see the start of it.
Slashdot Mirror
Spend ... one hour buying some wireless networking gear. Presto! Everybody's happy!
You'd better hope your boss doesn't find out that "ethernet" was originally called that because it was designed to work "over the ether", i.e., it was supposed to be wireless. The wired form was just a stopgap until the wireless transceivers became available (and could legally be used in the US without applying for an FCC broadcast license for every node).
Of course, a boss as clueless as you describe probably wouldn't understand the humor in an explanation like this.
To me, this is an example of catastrophically bad platform design.
.Trash that can't be deleted. It got there because I rsync'd some files over from another computer whose file system was "8-bit clean" and contained file names in German, Swedish, Finnish and a few other languages that use non-English 8-bit letters. These file names got turned to gibberish by the OSX file system. And the software is determined to "protect" me by not letting me delete this file. None of the dozens of tricks I know to defeat "safe" rm commands seem to works. Not even using a hex dump of the directory and typing the file's name in as a char[] array in hex in a C program and calling unlink() works. Queries on Mac newsgroups basically get jeers because I was stupid enough to use rsync with OSX (and I conclude from this that they can't solve the problem either ;-).
;-)
.Trash have also failed miserably ...
Well, as a unix programmer for around 30 years now, my main gripe is people who criticise the conciseness and power of the design, and try to "fix" it by kludging in "safeguards".
One of the main results, from a programmer's viewpoint, is that there's no reliable way to unconditionally remove a file. People keep discovering that files can be removed accidentally, and "protecting" users from such mistakes. This means that my script isn't portable to yet another system. Temp files get left behind ("like little turds all over the disk", as a friend once described it), and code fails because the file that "was just deleted" is still there, and re-uses the old file instead of making a new one. And there's just no way to learn how to correct for this on every new release of every system that comes along.
An especially funny case: On the Mac that I'm using to type this, there's a file inside my
Life would be a lot easier for us programmers if there were a command that would simply and silently remove a file, and fail only if I don't have write permission on the directory. (Hint: It should not fail if the file isn't there. The file not existing what I want; it's not an error.
Life was much easier when I had unix systems that would do exactly what I told it to do. Now I have to figure out how defeat the "user-friendly" safeguards that are different on every unix system, just to do something as simple as make sure a file isn't there.
OTOH, defeating the growing "user-friendly" versions of unix software was one of the primary motivations for the perl language, so I guess that something good has come out of it all. But my attempts in perl on OSX to remove that file from
No...regular internet companies don't save any of your mail.
Unless, of course, some government agency tells them to. And, in the US, they must keep that fact secret from you.
Also, note that setting up your own mail server (if your ISP doesn't block the email ports, as many do) doesn't protect you. If your IP connection is through a single ISP, they can quite easily collect all packets to or from you with a given TCP port number, and assemble the messages at their leisure. Or the government agency can do so. Or an employee of a competitor who has bribed the right person at your ISP.
The only thing really protecting you is that no ISP wants to buy the storage that it would take to do this for all customers. They'd be bankrupt in a month if they tried it. But under threat from your government, or reward from your competitor, they just might find the resources to save the packets to/from a few select customers.
The only real way to get privacy of course, is end-to-end encryption. But both ends have to have a motive to do this.
I have occasionally wondered how many governments are getting curious about the fact that much traffic to/from unix-type machines now uses ssh/ssl. The main reason, of course, is so that you can send passwords without giving them away to anyone with a network sniffer. Telnet, rlogin and ftp have been mostly supplanted by ssh and scp for this reason. But it turns out that it's usually just as easy to encrypt an entire conversation, rather than just the passwords. And then you don't slip up and accidentally send a password in the clear.
Still, it's rather naive to underestimate the "chance" that your email is being intercepted and cached by someone. Most isn't being saved anywhere. But it's not all that difficult to a motivated party, especially one who can use taxpayer money, to implement this. Unless your email is encrypted, you should always assume that random unknown people are reading it.
I've found some surprising old messages of mine simply by googling for my name or an old email address.
Google is pretty close to having a monopoly on search engine services. Remember, you don't have to be the only provider to have a monopoly, you just have to weild "monopoly power", that is the ability to control the market, and I think Google is getting damn close to that.
;-)
Maybe, but as one of several hundred people running a small, highly-specialized search site for a type of technical data (details of which don't matter here), I can say that I have yet to see any signs of google trying to wield such supposed power.
While we all know the value of google, there's still the general problem that keyword search of text just isn't very good for finding a lot of technical data. Suppose, for example, that you're doing DNA research, and want to locate sites that deal with a particular string of DNA. Ask google about "CGA TCC CAT TGG TGC" and see how many responses you get.
There is a fair amount of research going on for other kinds of search, plus of course the research on making computers "understand" text iin order to give matches that are more relevant than is possible with keyword matching. Google is doing some of this, as are the other big search sites. But so far, there doesn't seem to be any pressure on us independent sites to stop our research or sell out to the big guys and do it their way. (Some of us have had feelers out to google to see if they're interested in hiring us; that doesn't seem to be getting much of a response so far.
With google, there are encouraging signs of the opposite. Thus, with google maps, they are actively encouraging people with various kinds of databases to correlate them with google's maps. Some of the sites leveraging google's maps are even commercial sites, and I haven't yet heard that google is trying to discourage them.
Of course, they could be just waiting for a better opportunity, when lots of small sites are dependent on google's maps, and they they'll pounce. Now that google is a public company, as others have pointed out, they just might suddenly decide that a major takeover campaign is in their stockholders' best interests. So we should definitely keep our eyes open. They could suddenly decide to copy Microsoft's move to kill Netscape, and set back search research for years in the process by killing or buying out all the small, specialized search sites.
But if there are signs of this sort of evil from google, I haven't read about them. Anyone have more information? Some of us would be very interested in any symptoms or clues about the future behavior of this 500-pound gorilla of searching.
The skill of using a piece of word processing software is completely trivial compared to the skill of learning how to write and command the language.
...
Indeed, and I've read any number of comments along this line from professional writers, journalists, and others who make their living producing text. Usually they say something like "I produce words. Then I give them to an editor to format for publication. I have no need to learn to use a word processor; that's the job of the people working for my editor." They often comment that fancy formatting is pointless for them anyway, because the publisher will just throw out all their formatting and redo it. So they just send plain text, with at most a bit of bold or italic emphasis here and there (which will be edited).
Teaching a word processor is practical in the same sense that teaching typing has long been practical, and the main value is the same: learning to use the keyboard. But it's worth maybe a couple of months at the high-school level. The rest of those years of "English" classes should be to teach use of the language.
Teach someone proper thinking skills, put him down in front of a computer,
Teach proper thinking skills? I had a couple of teachers who tried that. All the religious, and half the political folks in the town came down on them hard. Thinking skills are something very dangerous for your typical teacher's professional life. Fat chance of that every happening.
My favorite happened while I was using a sig that said something like:
;-)
Unix perversity of the week:
find . '*.o' -exec rm {} ';'
Don't try this at $HOME
This got appended to messages that I sent to several mailing lists. Of course, I got several very nasty letters from people who didn't understand what was wrong with the command, and tried it. At $HOME, of course.
You can probably imagine how I learned what this command does. It's one of my favorite examples of bad UI design.
(No; don't try it at $HOME. Try "mkdir foo; cd foo" and link in a few random files from some other directories, to see what it does with them.
Yeah; I've heard there's something new called FTP that lets you download whole files over the internet. Anyone know anything about it? Imagine using this to download, say, music files. Then, instead of carrying around a case of CDs, you could just have the music on your disk. Imagine if someone made a pocket-size music player with a disk that holds a few gigabytes of music, and you could get the music from the internet ...
They are coming up with a new name, ...
Just as long as they don't claim something like Tango or Polka or Waltz. I'd hate to be dragged off the dance floor and charged with copyright infringement, right there in front of my partner.
I learned Mambo in a class 25 years ago, but I've never had an opportunity to use it siince then, so it's OK if they lay claim to that name.
My favorite cosmology has long been the one in which every particle in this universe is a data structure inside a computer in the real universe. That computer is running the simulation that is our universe.
...
In this model, the basic unit of our reality is a bit of memory in the real universe. Elementary particles are a second-order concept, a data structure made of a collection of bits. Time itself is quantized, and the quantum is the time it takes the real computer to calculate the "next" state of all the particles in our universe.
It can be fun to argue this cosmology. But it has gotten somewhat less fun since the Matrix movies came out. It's no longer such a radical concept.
In such a universe, miracles are easy to explain. Something has gone wrong, so the simulation is stopped and restored from backup. A bit of editing is done, and the simulation is restarted.
Maybe this is what the Intelligent Design people are really talking about
Hell, American pronunciations probably sound funny to all Commonwealth states no matter what regional dialect is used.
Not really. Due to American dominance in the entertainment industry, English-speaking people everywhere are very familiar with the major American dialects. They don't sound funny, just American.
Similarly, few Americans would consider British accents (at least RP and others commonly heard) as funny. Cute, quaint, charming perhaps, but not funny. We've heard a lot of Tony Blair lately, and he doesn't sound funny; he just sounds English.
Except for Monty Python, of course; they're funny. Or Tracey Ullman. But she's fluent in so many accents that you have to keep reminding yourself that she's actually a Brit.
They propagate at the speed of light in all reference frames, ...
;-)
Yup. But not in all materials. TFA blew it for me with the comment:
Light signals race down the information superhighway at about 186,000 miles per second.
Uh, no it doesn't. In a glass fibre, light travels quite a bit slower than that. It only travels at that speed in a vacuum. In glass, light travels significantly slower than c. And parts of the information superhighway is made up of electrons ambling down a wire at speeds very much below c. (Then there's the speeds of the bongo-drum and avian-carrier links.
This comment pretty much made it pointless to accept anything else in the article as fact.
All very true, and this illustrates why such a gadget shouldn't deal with a voice system like E911. Its comm stuff should do wireless IP, the software should connect directly to your hospital's computers, and the computers can notify whoever needs to be notified.
We've had a lot of discussion about how hopeless the wi-fi system is for giving coverage. But my pocket "smartphone" now has full-time, always-on IP connectivity, and the cell-phone system has much better coverage than wi-fi ever can.
Perhaps such medical monitors can be the excuse we need to implement universal wireless net access. And maybe force the recalcitrant comm companies to implement deals that will make everyone's monitor work regardless of which comm company provides service at each remote location, without first requiring minutes to negotiate yet another $10/day "contract" because the service is via a different company than the relay in the previous block.
And, once we've got that, we can include a tiny microphone and speaker, and use VoIP to add "phone calls" as a feature. And we can look forward to the day when young people have no clue what a "phone" used to be or why we use such a word with our pocket/wrist comm/audio/video gadget, whatever it's called by then.
We could be doing all this now, y'know.
some device that you wore on your wrist that told you the time. More advanced applications could include the date.
Heh. I had a device like that several years ago.
One day, during a spell of hot weather, I noticed that I'd developed a minor rash under it, so I didn't wear it for a couple of weeks, until the rash went away.
Then I noticed that I hadn't missed that wrist display. It seems that in the places I hang out, it's nearly impossible to be out of sight of a time display. If I'm out walking in the woods, well, I have a cell phone in my pocket, and it has a time display. So why carry one around on my wrist?
I haven't missed it.
I've also read that the sale of wrist watches has been declining in the US and Europe for the past 5-10 years. It seems that a lot of people have been reasoning the same way.
OTOH, a compact Dick-Tracy-style wrist communicator could revive the market. Especially if it included a basic health monitor that could display some exercise-related data like heart rate. It's easy enough for such a gadget to display time and date by default.
A couple of years back, I read of a compact wrist cell phone that was being sold in Japan. I wonder what ever happened to it?
I would start by suggesting that you contact a Professional Financial Advisor ...
Financial Advisor (n). Someone who invests your money until it's all gone.
Yeah; you're right. The basic problem is that English is the only European language that hasn't had a major spelling reform in the past century or so. Our spelling is a mess, with no reliable connection between spelling and pronunciation.
;-)
As a result, there is no good way to transliterate words and names from another alphabet into English "phonetics". English doesn't have a phonetic spelling system. Most English sounds have more than one spelling, and most of our letters have more than one pronunciation.
In addition, Russian sounds are not the same as English. No matter how you spell a Russian word for English-speaking people, they can't get the sounds right. People try, but the only result is several different systems for transliterating Russian to English. This just confuses things more.
It's a mess, and it will probably never be fixed. There is no language authority for English. Other languages have official organizations, usually government departments, to standardize the language and writing systems, but this has never happened with English, and probably never will. I've always felt sorry for people who have to learn English. Maybe our writing system isn't as bad as Chinese, but it's still one of the world's worst.
It's utterly hopeless. You'll just have to accept bad transliterations of Russian words into English, and even worse mispronunciations.
(But then, I've heard some rather weird mispronunciations of American place names from Russians, not to mention Germans and French. We just think it's cute and charming.
Boris Yeltsin (sp?)
;-)
/. can actually handle simple UTF8 encodings. It looks fine on my screen, and cut-and-paste to several other browser windows here on my Mac PB seems to work ok. But there is still software lurking out there that can't handle anything but ASCII.
...
Hey, the correct spelling is " ".
Jeez; why are the people here such sloppy spellers?
(Actually, I had to look it up at wikipedia, because I didn't remember whether the '' was hard or soft. Hardly anyone ever gets this right in the English transliterations.
Now to see if
The quoted string in the first paragraph above should contain his name spelled correctly; i.e., in the Russian version of the Cyrillic alphabet. No spelling using the Roman alphabet can really be correct, of course.
Yet another vote for overly-picky corect spelings
(Just thought I'd test your theory.)
Hey, who let that Buddhist into the discussion? ;-)
The purpose of science is the search for fact.
... Maybe I should preview this one, to make sure that none of my editing has garbled anything. ;-)
Well, yes. But an important side effect is generating and testing explanations of those facts. With an emphasis on testing, which usually means you have to go out and collect more facts (usually called observations, or just data). So as a scientist, most of your life will always be collecting the facts that you need.
One of the more pointed explanations that Stephen Jay Gould made about evolution was to point out that Darwin didn't show that evolution had happened. By the time Darwin was born, this was an accepted fact among scientists. All these fossils had been dug up, and they showed a clear set of changes with time. Geologists got involved, and concurred with the whole thing. Nobody who actually studied the fossil data questioned this. But the observed evolution was very non-random, and a good explanation was lacking.
What Darwin did was to present a theory that explained why the fossil record showed certain kinds of evolutionary change and not others. And, most important, his theory was testable. Incidentally, it offended religious people, because it didn't need an intelligent guiding hand. Scientists immediately jumped all over it, of course, and managed to collect a great deal more data that kept coming up consistent with Darwin's theory. Religious people also jumped all over it, but they didn't understand scientific testing methods, so they couldn't disprove anything, or even understand why they were expected to do so.
And, of course, lots of philosopher types have pointed out that none of this ever dealt with proof or truth. Rather, people had simply failed to find data that disproved Darwin's theory. This sort of double negative is standard scientific method, and is where the term valid comes in. That just means a theory that can successfully explain all the observed data despite many attempts to shoot it down. It doesn't mean truth, because we might have several valid theories competing at once, and new facts might pop up at any time that would shoot down any of them. A valid theory is only tentativily accepted, because it has passed a number of tests and hasn't (yet) failed any. See Karl Popper for lots more words on this topic.
Similarly, Einstein made some rather outrageous predictions about the universe's behavior just a century ago. This was in an attempt to find a theory that explained some rather outrageous observations (i.e., facts) by other scientists in previous decades. Since then, physicists have repeatedly found new ways to collect data that could disprove some of Einstein's equations. They have repeatedly failed; his equations always predict results that are within the error bars of the observations. Maybe next month someone will find an exception, but for now, we have to accept Einstein's theories as valid descriptions of our universe.
Now, scientists often carelessly use true for valid, when true should really only be used for facts. A fact can be true or false; i.e. it does or doesn't describe an actual observation; a theory can only be valid or invalid. It's true that evolution has happened on our planet, but Darwin's theory isn't true; it's valid (so far).
Of course, all of this is above the mental capacity of most of the media or the political system (or the religious communities). So we have an ongoing bogus "debate" on such topics.
(Hmmm
Unicorn Theory can also be framed in such a way that it is not in conflict with the known facts, ...
...
Indeed, and the followers of the Invisible Pink Unicorn are following the ID story with a great deal of interest. If the ID people succeed at getting their "theory" imposed on schence teachers is some US states, we will see a followup demand that IPU theory also be included in science classes.
There are a number of other such competing theories waiting in the wings
What is a color if not something that we perceive to be as such?
;-)
We might point out that there's a critical difference between "The sky is blue" and "The sky appears blue". The first isn't quite correct, because "blue" isn't actually an accurate description of the sky's spectrum. The second is correct, because it acknowledges that the color depends on the observer's optical equipment.
It would be even better to say "The sky appears blue to the human eye". It has different colors to other animals. Thus, birds have four visual pigments. Three are like ours; the fourth has peak sensitivity around the violet end of human vision. To birds, the sky would stimulate the blue and violet pigments about equally, and the sky would appear a complex blue/violet/UV blend. An avian interior decorator would have words for those colors. Something similar would happen with a lot of insects, who typically have a V/UV-sensitive pigment but often no red-sensitive pigment.
To get even pickier, we might note that in mammals, birds and insects, there is significant variation in the actual frequency response of the visual pigments. There is also intra-species variation in many species. Humans are one such. When I was in high school back in the 60's, a physics teacher did a lab demo of this. He set up a prism to give a solar spectrum, and it was good enough that he could use a couple of absorbtion lines to calibrate it and label the frequencies. Then he had us go up to the paper and put marks at where we saw the ends of the rainbow-colored bar. The marks were approximately normally-distributed around the 400-nm and 700-nm points. The students were duly impressed by this demo of the variations in their color vision.
He went on to explain that any of us into photography should appreciate this. Different people have different opinions about how good various films and printers reproduce colors. This is partly because their pigments can't match the visual pigments for all of us exactly. How good a picture's colors look is partly determined by how closely the pigments match your visual pigments. Because human eyes vary so much, no printing system using only a few pigments can be accurate for all of us. (In my case, the red ended somewhat past the average for the class, but my violet mark was right at the average.)
But imagine how our pictures (and computer screens) must look to birds. The violets and ultraviolets are missing. It would be like us looking at a picture that has all its blues zeroed out.
A lot of birds and insects have ultraviolet markings, as do the flowers that they pollinate and the fruit that they eat. These markings are invisible to human eyes. Biologists have only recently learned to appreciate this, and a lot of mysterious behavior has become clearer as a result.
In particular, it turns out that a lot of avian navigation can only be understood if you realize that they see ultraviolet and they see polarization. There are situations in which birds are using polarized ultraviolet/violet sky light. If you can't see this yourself, you have difficulty explaining their behavior, because the sky just looks plain "blue" to you.
Info on the topic is easily available online. For avian color vision, google for "avian color vision". Use some photographic or printers' terms to find discussions in those subject areas. What you see through your eyes is only the start of undstanding what color the sky is.
(And I can hear the shouts of "Too much information!"
Are you stupid?
;-)
;-)
Probably.
The FIRST RULE while taking to anyone in "tech support" is to swallow all your pride and admit you know nothing.
Yeah; that's my standard approach. You learn more that way. After all, you've called Support because you don't know enough to make something work right. They're there to help you, right? And, hopefully, to educate you a bit in the process. So no matter how much you know, you're ignorant in this situation, and it's best to be honest (to yourself at least) about this.
Why the hell were they even informed you were running Linux on your network anyways? Were you sitting there on the phone and thought it'd be impressive to the support person to mention you were running Linux?
Nope. I didn't volunteer the information until the guy started asking questions about the Airport's Internet connection. I really thought this was irrelevant to diagnosing a Powerbok->airport->printer problem, but I thought I'd let the Support guy make suggestions. Maybe I'd learn something. I answered his questions honestly, which is probably a useful thing in such situations. I didn't know why he was trying to solve the problem by working on the Internet connection, but the obvious thing was to follow his suggestions.
This ended when he first demanded that I reboot the linux firewall. This, IMHO, was so far beyond the pale that I simple answered "No." He got all bent out of shape by my refusal to follow orders, damage my local network, and kill other tasks that were running. When he changed to demanding that the linux boxes be removed from the network, I again simply said "No", and pointed out that I hadn't called about a LAN or Internet problem (and we'd verified that the PB, Airport and linux gateway all had fine Internet access). I suggested getting back to the PB->Airport->printer problem, by isolating them from the LAN. His response was that he couldn't help me as long as I had a linux box on the LAN.
Part of what I was doing was testing Apple's support for a network with mixed equipment. I was doing it on my home network, and I wanted it to work there. But I was also working for people who were interested in various networking equipment on their corporate network. Since I had two PBs, an Airport Extreme, and couple of linux boxes and a Windows box on my home LAN, they wanted data on how well these all worked together. I never mentioned this to any Apple people, of course. But they gave me some very useful information on the subject. I could report that Apple's tech support people were actively hostile to a mixed network like this, and refused to help solve a simple Airport connectivity problem because there was non-Apple equipment on the LAN. For a software-development company that needs a mixed-vendor setup for testing, this is something you want to know, because it's a total deal killer.
Using some of the net-testing stuff on my linux box, I did eventually diagnose the problem myself. In fact, it took less time than I'd wasted talking to Apple Support. The problem was LAN-related. Something I didn't yet know was that the Airport Extreme had a DHCP server enabled by default, and it was handing out addresses that overlapped with the DHCP server on my linux gateway. (I can feel net admins cringing in horror, knowing what's coming.
The printer's IP address was in fact the same as the Windows box's. When I tracked this down, it was just a matter of time until I found the Airport's DHCP stuff, and reconfigured it to not overlap with the gateway's DHCP stuff. (I've looked for Apple tools to diagnose this problem, but haven't yet found them. Log messages on our two PBs weren't helpful, but log messages on the linux box were.)
I don't know whether the Apple Support guy would have found this. His plan was to install the Airport as the Internet gateway, with no LAN other than the wi-fi. The printer would have worked under those conditions. I'd guess that
Depends on your field of study. I've seen a number of geological discussions. That 4-continent definition is included in the wikipedia definition,
but few if any geologists would agree. Thus, Africa really is a separate continental plate, which just happens to be bumping up against Eurasia.
Similarly, there are distinct North American and South American plates. The North America plate's southern edge is a bit fuzzy, but it actually ends roughly at the southern edge of Mexico. The isthmus between Mexico and Columbia, "Central America", is really a recent (10 million years) volcanic development, like Cuba and the Antilles. It was produced by plate tectonics, but isn't part of a continental plate.
But this is just for geology. For other fields of study, other definitions make sense. Though it has always been a bit odd to classify Europe and Asia as different continents. This is partly historic confusion caused by the fact that "Asia" used to mean what we now call "Turkey", and what we now call "Asia" wasn't known to Europeans. Somehow "Asia" got extended to all the land to the east of its original definition, and since the fastest way there from Europe was by boat, people decided it must be a separate continent.
It can be hopeless to try to straighten out terminology when it's as confused as this. Especially when people in different fields are using a definition that makes some sort of sense to them, but not to anyone else.
Funny how this confusion keeps coming up, and how few people see the obvious answer.
;--)
Similarly, we refer to the Estados Unidos Mexicanos as "Mexico", and the República Federativa do Brasil as "Bra[sz]il". I wonder why people don't object to those shorthands? Maybe because they realize that there is no other country with "Mexico" or "Brazil" in its name. You'd think people would be smart enough to see the same explanation for the "America" shorthand.
Then there's the case of the Bundesrepublik Deutschland. It takes a bit of convoluted linguistic history to explain why we call it "Germany" in English. Most people would fall asleep some time during anything like a detailed explanation.
Anyway, "USA" is also a city in Japan, so using it alone is ambiguous.
I think it's very similar to the way we view roads. Most of them everywhere are "free" in the obvious sense that you don't have to pay a toll to use them.
This doesn't mean that it doesn't cost money to use on a road, of course. Vehicles cost money; fuel costs money; insurance costs money. Some places, you even have to pay extra for a parking spot. Part of your taxes, especially on fuel, go to maintenance of the roads.
Still, it makes sense to say that the roads are "free". You don't hand out money when you use most roads, except for a very few toll roads.
It wouldn't be surprising if the Internet ends up "free" in the same sense. You'll have to buy your comms equipment, and/or pay a monthly rent. But, except in some special spots with special controls, you'll be able to communicate from anywhere without paying a toll to a local ISP. Like the roads, the Net will be accessible if you have equipment able to use it.
Perhaps your monthly rent will be for bandwidth. You pay $N per month, and get B bytes/second of access. Pay $2N and get 2B bytes/sec. Or something like that.
We're nowhere near that yet, of course. Most of our comms are like a road system where you have to stop and pay a toll every few blocks, and each road owner can decide whether your vehicle is allowed on the road, what sort of freight you can carry, etc. for his little stretch of road. But there are signs of change
Actually, a number of science fiction writers have already used this idea as a plot gimmick. Several have considered the effect on society of a guaranteed Right of Communication. The ones I've read have been centuries in the future. Maybe we'll live to see the start of it.