Yeah. Hint to people considering CS as a major: once you get to college, you better be able to figure out how to program on your own. Even if you take a course on Java or C, your professor will only be able to cover so much in class. You'll basically be figuring everything out on your own, and coming to class to clear up questions you had on personal study time.
Or, to put it differently, putting effort into a long term educational goal while those around him/her are playing. (Probably has more to do with interests than planning, but you have to admit that it takes a lot more effort to learn programming than it does to sit around the house or watch tv.)
I would leave BASIC out and just go Pascal-->C. Pascal teaches you to think in procdures, if statements, and loops. It isn't any harder to learn to think in terms of if/while as your control statements than it is to learn if/GOTO. Yet GOTO must be subsequently unlearned. So why teach GOTO in the first place?
One of the problems I have with C++ is that there are 24 different ways to do something, all of which end you up with approximately the same code efficiency. Maybe I don't like a feature of a language, but if the programmer who designed the library I'm using did like the feature, I have to use it anyway. When there are three different ways to achieve a certain effect (for example, the "Interface" can be implemented with multiple inheritance, interfaces in Visual C++, or owning a translator class), it gets confusing (well, this library exports an interface, and this other library exports a translator class, and this other library expects my classes to inherit from their classes, so in the classes that use all three libraries, I'm toast!).
I liked Modula II better. It wasn't truly object oriented, but you could achieve the same effect, and you could usually only do things one way: the right way. It sure made debugging and interfacing with other people's code much easier.
I'm not advocating Modula II. I understand its deficiencies. But it was nice working with a simple language where there was an obvious RIGHT way to do things (and then you let the compiler do the optimization), thus allowing your code to actually be READABLE!
I like Scheme. But I have a lot of trouble imagining myself teaching it as a first computer language.
Scheme has too much punctuation (in the (form of (nested (parentheses (that end) up looking) like this))). Functional languages like Scheme and Lisp are nice and simple, but they take a lot of getting used to. From my experience, Scheme is a great language for exploring some advanced CS concepts, and it is a nice clean simple language, but putting everything in parentheses and hiding function declarations as
(define xplus1 (lambda (x) (+ x 1)))
is more than simple coding. It's clean, but not simple. The idea doesn't seem to be "teach a clean, ideal functional language," but "teach a simple programming language that students can relate to."
I haven't seen any discussion of LOW bitrate encoding. From what I can tell, all 3 codecs do a fairly good job at 128kbps. Good. Whoopidy do. I could have told you that. It's nice to know that MP3 is a little bit better, but it was just a little bit. All three will produce output from my computer's speakers that sounds identical to the CD if encoded at 128. So what?
I'm interested in how low the different codecs can go and still keep that characteristic. Which codec can encode at the lowest bitrate and still sound like a CD on my computer's speakers? And which codec sounds the best when streamed over a 56k or 28.8k modem? This is probably more important for the overall computer industry. Audiophiles will probably almost always keep the audio on true CD's, and may now occasionally be willing to do some encoding with 192kbps MP3's for greater convenience. But the rest of us want to know which codec to pick when we want to check out a song over the internet and don't have T3's going into our home computers. At 32kbps (56k modem speeds), who sounds best? And which codecs can take a hit from a late/lost packet (graceful degradation)?
The test was valuable. 128k is great for my personal collection. But most of the Internet music industry is running at a much lower bandwidth. Which codec is right for them?
A lot of the top medical schools are more willing to accept chemistry/physics/CS/math/engineering majors than pre-med majors. They want to 1) be able to start from scratch, without having to "unlearn" the junk they got in pre-med, and 2) have well-rounded students who will later be able to do top-notch research. Who do you think invents the robots that help doctors do their surgeries? A pre-med major who became a doctor? Where did he learn the robotics? It is more likely that an engineering major who later became a doctor would be a pioneer in something like that.
Lets see... CS majors, I'd guess around 100/year from my university. Students in CS 142 who plan on being CS majors, I'd estimate about 700. No "weed out." No "You weren't accepted for the major." Just people say, "oh, this isn't the easy money I thought it would be."
Well, I don't think you'll be able to use CVS and make world with MacOS. I would imagine that they've customized it, and that it has branched from the BSD tree. But they can still port bug fixes and patches, if they pay attention.
I am inclined to believe that this is a Good Thing(tm). Of course they did it for a marketing objective. Everything every company does is for a marketing objective, to make money in the long/short run. That isn't wrong. What is wrong is to not play fair, to force everyone else's hand, to damage the overall industry to get ahead (and if you have any idea who I'm talking about, good for you). I don't see anything wrong with what Apple has done.
Potential for bad: some hackers waste their time.
Potential for good: We're actually going to know what is going on behind a commercialized OS. We've all heard the rumors about the crazy stuff in Win98 code:
// There is a memory leak here, but I can't figure out what causes it...
or
// Crashes here. I would fix it, but release is next week, so no time. We'll get it in the service pak.
and of course
if(appID=APPID_NETSCAPE_NAVIGATOR){ SetTimer(random,CrashIt) } if(dosVer&DOSVER_DRDOS){ MessageBox("Your machine is running an unstable operating system. Upgrade to MSDOS 5.0."); }
So obviously if they are willing to let the source out, they are comfortable with it. They are confident about it. They have nothing to hide, and can prove it (at least for the part released, which appears to be the most critical part of the OS). No more secrets. This is a Good Thing.
Ok, so it's license sucks rocks. So? I wouldn't expect a whole lot of big user development on their kernel. I don't think that is the point. I'm sure they'll accept and take a look at all submitted bug fixes, but that isn't the point either.
The point is that everyone can study the OS and know everything about it that they need for programming the Mac. I'm reminded of the source code to the original PC BIOS that appeared in the appendix of the original IBM Technical Reference manual (still good reading, by the way, if you do low-level PC programming). I learned a lot about what was going on. When I had a question about "what exactly does this function do," and the docs were sketchy, I could jump to the assembly code in the appendix and figure it out. For some things, the BIOS source was my only doc - like how to program the timer chips, etc.
With MS's OS's you can't do that. And believe me, the docs are sketchy sometimes. With no source to look at when things get confusing, you're always hoping that there is someone on some newsgroup with the answer. And the MS Word team gets privy info on the undocumented API's. None of that with an OS OS:P.
I don't think I'll ever look at the code. I'm stuck in a rut programming my PC under FreeBSD, Linux and Windows. But for those who do need to do a lot of programming on Mac boxes, I have a feeling that this will help a bunch.
I dunno about everyone else... I got plenty of history in my CS classes. I even had to memorize a specific implementation of how to simulate a Turing machine ON a Turing machine (supposedly a final proof of its computing power). Theoretically, I guess, a Turing machine could emulate a Pentium II and run Windows NT or Linux. Any takers?).
I imagine those who learn computer programming on their own or in technical schools probably don't usually get as much of the theory, discipline, and history as those who get a 4 year degree or go on to a Master's degree. Or maybe I've just had good teachers. In any case, I feel like my understanding and appreciation of computing history is a valuable asset that improves the quality of my work. It is valuable to know what they used to do, so that you can understand why we don't do it that way anymore. It is important to know several languages, so you can understand the strengths of the one you use most often. etc.
FWIW: in a warped sort of way, ASF _IS_ MP3. ASF is a wrapper around almost ANY video or audio streaming compression method. One of the encoding options for ASF is MP3. It probably won't play in an MP3 player because of the wrapper, but the sound quality is exactly the same. The wrapper makes it easier for the system to stream it over the Net.
You know, a lot of people get really nervous when they have to speak in public. Granted, those people shouldn't get into politics. Granted, as well, that some of the things Al Gore said are outrageously funny and off the wall. But give the guy a little slack! I know I've said some dumb things under pressure, and I'm not even nervous in public. I sympathize with the poor guy who can't do anything right (sometimes it seems that I'm that way, too!).
I LOVE SomeOS! Nobody has ever broken into my SomeOS Advanced Systems Server. It picks a random port for each service every 48 hours, so that we can pick up snoopers. (Of course, the users then have to scan the ports, too, but that's ok, we trust them, and so we just spend a little bit of time looking over the tripwire logs and filter out all of the normal sniffing.) Our Web Enterprise Access Kit web server is the exception. We leave it on a pre-specified nonstandard port that nobody knows about, except for all of the people who write the web pages that link to it. Other than that, the only way anyone would be able to tell what port the WEAK server is on is to look at the URL that some user has navigated to. How likely is that?
Out of myself and the 2 other SomeOS users in the world, none of us have ever experienced a malicious attack on our server. 100% security!
Isn't there an O(1) way to check the solutions? Given enough filter paper, you can do your DNA weight analysis all at once to find the strands that are the right weight, do some other chemical processes with the whole mess at once to get the optimal solution. At least that is the way I understand it. (Of course, again we get into the linear time vs. exponential equipment/resources.)
Well, if MS were to distribute Toc, it would say AOL all over it. That wouldn't look as cool as, oh, say, MICROSOFT... And then there is that sticky issue of legality, although it never stopped MS before.
I've heard rumors (unconfirmed) that a lot of people have had trouble with Toc clients lately. I dunno. I ICQ anyway.
Programming JavaScript for Netscape is a nightmare. IE isn't great, but it is a lot better. That and reliability aside, the two browsers are pretty 12 for one, a dozen for the other.
As far as reliability, IE seems much more stable under Windows 98 and NT 4.0 (YMMV), although Netscape under Unix seems much more stable than Netscape under Windows (go figure).
Hangs randomly? Of course. Unfortunately, IE hanging randomly every 6 hours isn't as bad as Netscape hanging predictably every time you load a certain page or do a certain task + hanging randomly every hour or so.
I'm not saying Netscape is worse. Just that in a Windows environment, Netscape is on enemy territory and it shows. (Did you expect MS to just lie down and let Netscape run reliably under Windows? You're a bigger fool than I thought you were!)
There is a setup called PicoBSD. It is a set of minimal distributions of FreeBSD that fit in 1.44 Megs. There is a router distro, firewall, dialup, among others.
If you dial up with a modem, none. But if you use the ISDN/DSL/cable modem, you'll need 1 to connect to the special modem, and the other to connect to the rest of the network.
Let's see, the (67 watt) power supply finally died last year on my mom's IBM/PC (4.77 Mhz 8088 power!), so I swapped the supply out of the Extension Unit and it ran great. Even so, we finally gave it to Goodwill. (Probably the deciding factor was that the WordPerfect 5.1 install disks developed a bad sector, so I couldn't put it on.) We got her a 286 and found a 486 motherboard to put in it, 20 megs RAM and she's happily running Win95 and getting everything done that she needs to.
She also had a souped-up PC/XT (with an Orchid 286-8 card in it) that she used at the school she teaches in for wordprocessing and "Where in the USA is Carmen Sandiego." The school finally got a grant and put iMacs in all the classrooms, so it finally got sent to Goodwill last spring.
And we still get good use out of the Epson FX-80 (prints company checks better than anything else around!)
That's two computers and a printer with a useful lifetime of 15 years. This 18 month lifetime crap ticks me off. Is it that they don't make them like they used to, or that we are just so darned rich that we can't stand to be a week behind our neighbors?
Re:New computers - who needs 'em?
on
High Tech Junk
·
· Score: 1
Yeah, no kidding. Intel is really pushing research into faster hard disks (99% of the time we spend waiting for computers nowadays is waiting for the hard disk) because who is going to buy a faster processor if they are already just waiting on other parts of the computer?
Set up a network and make this the network router. Give it two network cards. Put Linux/FreeBSD on the machine, run NATD and SAMBA. Hook this computer up to the net via modem/cable modem/DSL/ISDN, and hook the rest of the network to the other card. Now, instead of fighting over who gets to use the Internet (only one phone line), everyone can share the connection, and the server will even autodial when necessary. Also, you can set up shared directories with Samba, so that it doesn't matter which computer you are using - you can get your files on any of them. If you use Windows, you may have a little more trouble getting the thing to be a printer server (although I'm sure it is possible - what do the experts say?). But it is cake to set up one of the Windows computers to be the printer server in that case.
And of course, the biggest advantage is that you get a Linux/FreeBSD box to play around with!
I really don't know how it is implemented or enforced, so don't ask me. However, a friend of mine recently went on a trip to Israel. She sent a few emails. Even though the computer and connection were provided for free by the group she was traveling with, she had to pay 1 shekel per email for email tax. I was stunned.
Slashdot Mirror
Yeah. Hint to people considering CS as a major: once you get to college, you better be able to figure out how to program on your own. Even if you take a course on Java or C, your professor will only be able to cover so much in class. You'll basically be figuring everything out on your own, and coming to class to clear up questions you had on personal study time.
Or, to put it differently, putting effort into a long term educational goal while those around him/her are playing. (Probably has more to do with interests than planning, but you have to admit that it takes a lot more effort to learn programming than it does to sit around the house or watch tv.)
I would leave BASIC out and just go Pascal-->C. Pascal teaches you to think in procdures, if statements, and loops. It isn't any harder to learn to think in terms of if/while as your control statements than it is to learn if/GOTO. Yet GOTO must be subsequently unlearned. So why teach GOTO in the first place?
One of the problems I have with C++ is that there are 24 different ways to do something, all of which end you up with approximately the same code efficiency. Maybe I don't like a feature of a language, but if the programmer who designed the library I'm using did like the feature, I have to use it anyway. When there are three different ways to achieve a certain effect (for example, the "Interface" can be implemented with multiple inheritance, interfaces in Visual C++, or owning a translator class), it gets confusing (well, this library exports an interface, and this other library exports a translator class, and this other library expects my classes to inherit from their classes, so in the classes that use all three libraries, I'm toast!).
I liked Modula II better. It wasn't truly object oriented, but you could achieve the same effect, and you could usually only do things one way: the right way. It sure made debugging and interfacing with other people's code much easier.
I'm not advocating Modula II. I understand its deficiencies. But it was nice working with a simple language where there was an obvious RIGHT way to do things (and then you let the compiler do the optimization), thus allowing your code to actually be READABLE!
Ok, I'm getting off the soap box now.
I like Scheme. But I have a lot of trouble imagining myself teaching it as a first computer language.
Scheme has too much punctuation (in the (form of (nested (parentheses (that end) up looking) like this))). Functional languages like Scheme and Lisp are nice and simple, but they take a lot of getting used to. From my experience, Scheme is a great language for exploring some advanced CS concepts, and it is a nice clean simple language, but putting everything in parentheses and hiding function declarations as
(define xplus1 (lambda (x) (+ x 1)))
is more than simple coding. It's clean, but not simple. The idea doesn't seem to be "teach a clean, ideal functional language," but "teach a simple programming language that students can relate to."
My 2c.
I haven't seen any discussion of LOW bitrate encoding. From what I can tell, all 3 codecs do a fairly good job at 128kbps. Good. Whoopidy do. I could have told you that. It's nice to know that MP3 is a little bit better, but it was just a little bit. All three will produce output from my computer's speakers that sounds identical to the CD if encoded at 128. So what?
I'm interested in how low the different codecs can go and still keep that characteristic. Which codec can encode at the lowest bitrate and still sound like a CD on my computer's speakers? And which codec sounds the best when streamed over a 56k or 28.8k modem? This is probably more important for the overall computer industry. Audiophiles will probably almost always keep the audio on true CD's, and may now occasionally be willing to do some encoding with 192kbps MP3's for greater convenience. But the rest of us want to know which codec to pick when we want to check out a song over the internet and don't have T3's going into our home computers. At 32kbps (56k modem speeds), who sounds best? And which codecs can take a hit from a late/lost packet (graceful degradation)?
The test was valuable. 128k is great for my personal collection. But most of the Internet music industry is running at a much lower bandwidth. Which codec is right for them?
A lot of the top medical schools are more willing to accept chemistry/physics/CS/math/engineering majors than pre-med majors. They want to 1) be able to start from scratch, without having to "unlearn" the junk they got in pre-med, and 2) have well-rounded students who will later be able to do top-notch research. Who do you think invents the robots that help doctors do their surgeries? A pre-med major who became a doctor? Where did he learn the robotics? It is more likely that an engineering major who later became a doctor would be a pioneer in something like that.
Lets see... CS majors, I'd guess around 100/year from my university. Students in CS 142 who plan on being CS majors, I'd estimate about 700. No "weed out." No "You weren't accepted for the major." Just people say, "oh, this isn't the easy money I thought it would be."
EASY MONEY????
Well, I don't think you'll be able to use CVS and make world with MacOS. I would imagine that they've customized it, and that it has branched from the BSD tree. But they can still port bug fixes and patches, if they pay attention.
I am inclined to believe that this is a Good Thing(tm). Of course they did it for a marketing objective. Everything every company does is for a marketing objective, to make money in the long/short run. That isn't wrong. What is wrong is to not play fair, to force everyone else's hand, to damage the overall industry to get ahead (and if you have any idea who I'm talking about, good for you). I don't see anything wrong with what Apple has done.
:P.
Potential for bad: some hackers waste their time.
Potential for good: We're actually going to know what is going on behind a commercialized OS. We've all heard the rumors about the crazy stuff in Win98 code:
// There is a memory leak here, but I can't figure out what causes it...
or
// Crashes here. I would fix it, but release is next week, so no time. We'll get it in the service pak.
and of course
if(appID=APPID_NETSCAPE_NAVIGATOR){
SetTimer(random,CrashIt)
}
if(dosVer&DOSVER_DRDOS){
MessageBox("Your machine is running an unstable operating system. Upgrade to MSDOS 5.0.");
}
So obviously if they are willing to let the source out, they are comfortable with it. They are confident about it. They have nothing to hide, and can prove it (at least for the part released, which appears to be the most critical part of the OS). No more secrets. This is a Good Thing.
Ok, so it's license sucks rocks. So? I wouldn't expect a whole lot of big user development on their kernel. I don't think that is the point. I'm sure they'll accept and take a look at all submitted bug fixes, but that isn't the point either.
The point is that everyone can study the OS and know everything about it that they need for programming the Mac. I'm reminded of the source code to the original PC BIOS that appeared in the appendix of the original IBM Technical Reference manual (still good reading, by the way, if you do low-level PC programming). I learned a lot about what was going on. When I had a question about "what exactly does this function do," and the docs were sketchy, I could jump to the assembly code in the appendix and figure it out. For some things, the BIOS source was my only doc - like how to program the timer chips, etc.
With MS's OS's you can't do that. And believe me, the docs are sketchy sometimes. With no source to look at when things get confusing, you're always hoping that there is someone on some newsgroup with the answer. And the MS Word team gets privy info on the undocumented API's. None of that with an OS OS
I don't think I'll ever look at the code. I'm stuck in a rut programming my PC under FreeBSD, Linux and Windows. But for those who do need to do a lot of programming on Mac boxes, I have a feeling that this will help a bunch.
Hope that made sense. Happy hacking!
I dunno about everyone else... I got plenty of history in my CS classes. I even had to memorize a specific implementation of how to simulate a Turing machine ON a Turing machine (supposedly a final proof of its computing power). Theoretically, I guess, a Turing machine could emulate a Pentium II and run Windows NT or Linux. Any takers?).
I imagine those who learn computer programming on their own or in technical schools probably don't usually get as much of the theory, discipline, and history as those who get a 4 year degree or go on to a Master's degree. Or maybe I've just had good teachers. In any case, I feel like my understanding and appreciation of computing history is a valuable asset that improves the quality of my work. It is valuable to know what they used to do, so that you can understand why we don't do it that way anymore. It is important to know several languages, so you can understand the strengths of the one you use most often. etc.
FWIW: in a warped sort of way, ASF _IS_ MP3. ASF is a wrapper around almost ANY video or audio streaming compression method. One of the encoding options for ASF is MP3. It probably won't play in an MP3 player because of the wrapper, but the sound quality is exactly the same. The wrapper makes it easier for the system to stream it over the Net.
I think it's 2038 or something that the 32-bit date field is good to under Linux isn't it? .. nice lot of time to move to a 64-bit architecture.
.. nice lot of time to move to a 16-bit architecture.
I think it's 2000 or something that the 8-bit BCD date field is good to under COBOL isn't it?
:)
You know, a lot of people get really nervous when they have to speak in public. Granted, those people shouldn't get into politics. Granted, as well, that some of the things Al Gore said are outrageously funny and off the wall. But give the guy a little slack! I know I've said some dumb things under pressure, and I'm not even nervous in public. I sympathize with the poor guy who can't do anything right (sometimes it seems that I'm that way, too!).
:P
And besides, I invented SlashDot.
I LOVE SomeOS! Nobody has ever broken into my SomeOS Advanced Systems Server. It picks a random port for each service every 48 hours, so that we can pick up snoopers. (Of course, the users then have to scan the ports, too, but that's ok, we trust them, and so we just spend a little bit of time looking over the tripwire logs and filter out all of the normal sniffing.) Our Web Enterprise Access Kit web server is the exception. We leave it on a pre-specified nonstandard port that nobody knows about, except for all of the people who write the web pages that link to it. Other than that, the only way anyone would be able to tell what port the WEAK server is on is to look at the URL that some user has navigated to. How likely is that?
Out of myself and the 2 other SomeOS users in the world, none of us have ever experienced a malicious attack on our server. 100% security!
Isn't there an O(1) way to check the solutions? Given enough filter paper, you can do your DNA weight analysis all at once to find the strands that are the right weight, do some other chemical processes with the whole mess at once to get the optimal solution. At least that is the way I understand it. (Of course, again we get into the linear time vs. exponential equipment/resources.)
Well, if MS were to distribute Toc, it would say AOL all over it. That wouldn't look as cool as, oh, say, MICROSOFT... And then there is that sticky issue of legality, although it never stopped MS before.
I've heard rumors (unconfirmed) that a lot of people have had trouble with Toc clients lately. I dunno. I ICQ anyway.
Programming JavaScript for Netscape is a nightmare. IE isn't great, but it is a lot better. That and reliability aside, the two browsers are pretty 12 for one, a dozen for the other.
As far as reliability, IE seems much more stable under Windows 98 and NT 4.0 (YMMV), although Netscape under Unix seems much more stable than Netscape under Windows (go figure).
Hangs randomly? Of course. Unfortunately, IE hanging randomly every 6 hours isn't as bad as Netscape hanging predictably every time you load a certain page or do a certain task + hanging randomly every hour or so.
I'm not saying Netscape is worse. Just that in a Windows environment, Netscape is on enemy territory and it shows. (Did you expect MS to just lie down and let Netscape run reliably under Windows? You're a bigger fool than I thought you were!)
There is a setup called PicoBSD. It is a set of minimal distributions of FreeBSD that fit in 1.44 Megs. There is a router distro, firewall, dialup, among others.
Something was really wrong with those MFM/RLL 40 meg hard drives. I had two of them kicking around, and they both game me no end of trouble.
On the other hand, both 10 Meg drives (full height - cool!) I had lasted 15 years with not one bad sector. Then we gave them to Goodwill.
If you dial up with a modem, none. But if you use the ISDN/DSL/cable modem, you'll need 1 to connect to the special modem, and the other to connect to the rest of the network.
Let's see, the (67 watt) power supply finally died last year on my mom's IBM/PC (4.77 Mhz 8088 power!), so I swapped the supply out of the Extension Unit and it ran great. Even so, we finally gave it to Goodwill. (Probably the deciding factor was that the WordPerfect 5.1 install disks developed a bad sector, so I couldn't put it on.) We got her a 286 and found a 486 motherboard to put in it, 20 megs RAM and she's happily running Win95 and getting everything done that she needs to.
She also had a souped-up PC/XT (with an Orchid 286-8 card in it) that she used at the school she teaches in for wordprocessing and "Where in the USA is Carmen Sandiego." The school finally got a grant and put iMacs in all the classrooms, so it finally got sent to Goodwill last spring.
And we still get good use out of the Epson FX-80 (prints company checks better than anything else around!)
That's two computers and a printer with a useful lifetime of 15 years. This 18 month lifetime crap ticks me off. Is it that they don't make them like they used to, or that we are just so darned rich that we can't stand to be a week behind our neighbors?
Yeah, no kidding. Intel is really pushing research into faster hard disks (99% of the time we spend waiting for computers nowadays is waiting for the hard disk) because who is going to buy a faster processor if they are already just waiting on other parts of the computer?
Set up a network and make this the network router. Give it two network cards. Put Linux/FreeBSD on the machine, run NATD and SAMBA. Hook this computer up to the net via modem/cable modem/DSL/ISDN, and hook the rest of the network to the other card. Now, instead of fighting over who gets to use the Internet (only one phone line), everyone can share the connection, and the server will even autodial when necessary. Also, you can set up shared directories with Samba, so that it doesn't matter which computer you are using - you can get your files on any of them. If you use Windows, you may have a little more trouble getting the thing to be a printer server (although I'm sure it is possible - what do the experts say?). But it is cake to set up one of the Windows computers to be the printer server in that case.
And of course, the biggest advantage is that you get a Linux/FreeBSD box to play around with!
I really don't know how it is implemented or enforced, so don't ask me. However, a friend of mine recently went on a trip to Israel. She sent a few emails. Even though the computer and connection were provided for free by the group she was traveling with, she had to pay 1 shekel per email for email tax. I was stunned.