"VOTE LINUS: He doesn't have military experience, but he wrote a kernel!"
Actually, Linus did serve in the Finnish army in 1989-1990, as a noncommissioned officer. I don't think it's anything against him that Finland happened not to be at war with anybody at the time.
The funny thing is that I bet the HX-20 had a better keyboard than virtually anything they're turning out today.
I don't expect so.
I had an Epson PX-8 Geneva, a 1984 descendant of the HX-20. Its keyboard was sort of like the keyboard in my ThinkPad 760: Smooth at first, but gradually harder to press unless you're one of those freaks who press keys straight down. Except that it doesn't flex like the ThinkPad's keyboard.
Eventually, several collections of keys stopped working, and the machine in general became really flaky, so I stopped using it.
develop a belief mathematics is useless as a defense mechanism
I'm sorry, but mathematics is useless as a defense mechanism: if someone has a gun on you, you can't put a Riemannian Manifold or anything like that in the way of the bullet, can you?
But you can put a book describing Riemannian manifolds in front of the bullet.
I'm almost thinking I should track down a USB VGA adapter and run a 3rd screen.
That's what Matrox is for. Their DualHead2Go and TripleHead2Go allow you to attach multiple monitors that have the same resolution to the laptop at once. Though, I think there's some software weirdness if you want the benefits of having a window maximized on only one screen.
There's always the rendering issue, especially when you do stuff like multiple character sets, rotations, embedded Word objects... The presentations I run are from other people, so the rendering is really important.
Spanned displays. If you want to specify the secondary monitor for your presentation, you have to use PowerPoint. Apparently, there has been a lot of discussion about this, and rudimentary multi-monitor support might make it to the next release.
Hey, replacing a broken string can lead to complete inoperability. If you mismatch the strings, the tension will be off, and your tone will be loose. If you use too heavy a heavy gauge of wire, you could even break the guitar's neck.
Guitars are a pretty mature product. Do you have any idea how many centuries it took to make the chromatic scale? Then look at the adoption curve for well-tempered tuning. Upgrading that was not as easy as opening a panel and pushing a button. It just seems easy now because we have a couple centuries of practice with this system. Even so, most people play by magic finger positions (chords) dictated by the experts, and can't tune their instruments.
How long is it going to take before we can agree on good metaphors for computer interfaces, like we have for music interfaces? And how much education will we absolutely require?
it's not like you can go grab a recently printed score of Bach, scan it and put it on the web. Although Bach's work may not be covered by copyright, the particular printing you're copying probably is.
My understanding of copyright law is that it requires a certain amount of creativity before something is considered copyrightable.
My understanding is that a lot of music, especially older than Classical period pieces, do involve a bit of creative work. The source materials are highly fragmentary: They were written before copyright was widespread, and the most widely published (and complete) editions are buggy. Even the authors had conflicting editions, as they had to compose for what instruments were available rather than The Vision of the Reasoning Man with a paying popular audience, and they had misprints. Earlier pieces, especially, were often not written until somebody in the 19th Century decided they were worth preserving (especially folk music), or the manuscripts were considered worthless and discarded (like a lot of Bach's work), or the only copies were destroyed in some random church fire (I'm thinking Schutz).
Any reasonably scholarly edition printed today will have an extensive story about which source manuscripts they used, how they adapted them to modern typographic conventions, what did they change because the editor thought it was a typo, where the ossia come from, and any other details. I think they're mostly found as compilations of, like, 12 Sonatas for Solo Flute from Telemann, or Organ Music by Bach including the entire Trio Sonatas and Orgelbuchlein and several more pieces.
Any purely performance edition will have the editor, the composer of the bass realization, the composer who filled in the whatever number of measures that were lost to history or never composed (think Christopher Tolkien but worse source material), and whomever else was involved in the adaptation. Performance editions often deviate in interesting ways from scholarly editions. They're also much easier to find in stores and libraries, often adapted to instruments that the composer never saw or heard.
The obvious solution is to find the piece that you want on paper (or practice enough to compose or improvise credibly), and if necessary adapt it to your instrument. If you read the liner notes on any reasonably scholarly performance recording, they often describe which sources they used, and how they filled in the missing pieces themselves, largely because they're experts showing off, but probably partially because they can't risk the copyrights.
For an image that is naturally limited to opaque 8-bit color, GIF is certainly lossless.... I don't know why some of you photo geeks have such a problem with that.
Because no color image received through a camera lens is naturally limited in such a way. Do you want us all to go back to black-and-white photography?
Ack! Straw man! GIF is used for digital photography only in my nightmares. While the real world is not a 24-bit bitmap, the data from the camera can be pretty nearly represented by it, so GIF is not appropriate. Even for black and white photos, I think the continuous curves don't compress that well with GIF's celebrated LZW.
What would be appropriate is something intentionally simple, most likely an icon for a button or a logo. Drawings naturally have only the colors that you put in them. Sometimes, shading and multilevel transparency are just extra work for the artist, the display device, and the storage device, and sometimes LZ77 inflate takes too much power. As those times become rarer, GIF becomes less relevant.
False. GIF is lossy, as it is limited to 255 colors per frame. Conversion from a 24-bit image to GIF involves an operation called dithering, which adds noise to the image.
"False" back atcha. With that criterion, you might as well say that PNG is lossy because it doesn't accurately reproduce image depth or iridescence.
For an image that is naturally limited to opaque 8-bit color, GIF is certainly lossless. If you adjust the palette correctly, you can represent any appropriate color without dithering. I don't know why some of you photo geeks have such a problem with that.
You can't drive a car without training; why should a computer be different?
Wow. Just wow. I'm ashamed to ever have used Linux. If our developers and/or users really think with their heads this far up their asses, the platform is dead.
You forget that you were trained on the WIMP GUI many years ago, and that familiarity continues to make Windows or MacOS easy to use. One thing I find to be fun is to find one of those old introduction to Macintosh books. (The last one I saw was in the back of a cabinet full of manuals in my undergraduate Physics lab.) After using the GUI for such a long time, it's enlightening to see just how many pages are devoted to making sure you understand the WIMP metaphor.
Granted, the WIMP GUI with a good mapping to the system software is much more intuitive than the man/info/HOWTO text-based system for more people. The main difference, then, is how much training you have to do to use a system.
Im sick of this (and will probably get modded down) but this isnt the sole fault of the vendors, now is it. For whatever reason, they will not release their driver set under opensource licenses, and thats agreeable because its their code and their decision. On the other hand, the linux kernel devs wont supply a stable module API, because they dont like binary modules, which is also agreeable because its also their code and their decision.
Linux is not an x86-only program. While that's not a problem with a chipset, there's that handful of Mac users running Linux that can't get speedy graphics. Apparently, nVidia can't produce a PPC driver anytime soon, as they have let Apple do the driver development all this time. The only real option is to switch to ATI.
Even if nVidia produces a PPC driver, that wouldn't be any help to the (admittedly small) groups of people using SPARC or Alpha or MIPS as their desktops.
We also use Perfigo at our University and have had almost zero problems past the first week.
I have to wonder whether you're actually using your system from the students' perspective, because people are used to taking a lot of abuse without complaint. See Bill Gates' comment some years back about Microsoft software not having any significant bugs. See also your term, "almost zero problems".
A friend of mine administers the network at another university, where they use Perfigo's system, on hardware that Perfigo recommends, only for the wireless segment. He says that it's really slow there, and its interface is very clunky, and it just doesn't work well. Furthermore, the company doesn't respond very well to requests for customer support.
However, they did respond with legal threats when he voiced his complaints in a public forum, and he's now too scared to say anything publicly against them. I'm sure that that gives a particular bias to comments in other public forums like this one.
At UCSC, we use Perfigo's SecureSmart servers for making it safe to plug the students' computers into the school network. It's bad.
The server is constantly going down. Get this: It checks every 6 hours to make sure that it's currently registered. Frequently, it forgets that it's registered or Perfigo's registration server scrambles its licenses or something, because the dorm network then goes down. This happens about once per week.
The system is based on a router running Red Hat 7 on commodity x86 machines. Last I heard, it was still using Linux 2.4.9. The upgrade procedure is a drive reimage. The actual routing goes on in a proprietary routing program with fairly low performance. The scanning is done with a customized Nessus. The administration is some custom PHP (IIRC) code, with no security roles and complete control via a single password.
Furthermore, the source to the free software they use, they refuse to send to the customer. Somebody really should see if they can sue Perfigo for violation of the GPL.
Ignoring the above, the Resnet administrator has set up the SecureSmart server to scan PCs for the usual Windows problems. If it finds one, he has it set up to let the user see only antivirus pages and Windows Update. Then it's supposed to scan the user's computer again after 24 hours. What usually happens is that the user's computer doesn't get an IP address anymore, ever, and the administrator has to unblock the specific MAC manually (using his single password).
I'm guessing that we're still using it because the administrator feels that he has invested too much effort into it already. I don't know exactly what Cisco was thinking. Perfigo is just a bad investment.
If you're also going to UCSC, you should check out https://api.alkaid.org/ It's currently a bit out of date, but it shows that the administrator should have known not to use Perfigo.
From what I hear, that person just took names from the forums and added @spymac.net. As noted by other people, you could do the same with Yahoo accounts. And, just like Yahoo, you could choose not to use the email, nor to enter any implicating information.
It doesn't really sound like vigilatism. I'd classify that message as spam, actually. It tells what should be obvious. At worst, it's awfully neglectful of the Spymac operators not to have a large privacy policy that explains such things. Sending email to everybody in the forums isn't a solution, and is likely to cause more confusion.
One serious objection that I have to Spymac, which can be checked out, is that it doesn't use SSL. Even for the paid webhosting and webmail. And all of the services are prone to failing without warning; it's been 7 months or so since the new services came out, and I'm still hearing complaints about their reliability.
Actually, there are some 640x480 B&W LCD panels, but without backlight.
One place that sells them, that I mention only because my brother was looking at them recently, is All Electronics http://www.allelectronics.com
The 640x480 LCD panel is $25, not counting material and labor needed to connect it.
Slashdot Mirror
Actually, Linus did serve in the Finnish army in 1989-1990, as a noncommissioned officer. I don't think it's anything against him that Finland happened not to be at war with anybody at the time.
I don't expect so.
I had an Epson PX-8 Geneva, a 1984 descendant of the HX-20. Its keyboard was sort of like the keyboard in my ThinkPad 760: Smooth at first, but gradually harder to press unless you're one of those freaks who press keys straight down. Except that it doesn't flex like the ThinkPad's keyboard.
Eventually, several collections of keys stopped working, and the machine in general became really flaky, so I stopped using it.
But you can put a book describing Riemannian manifolds in front of the bullet.
That's what Matrox is for. Their DualHead2Go and TripleHead2Go allow you to attach multiple monitors that have the same resolution to the laptop at once. Though, I think there's some software weirdness if you want the benefits of having a window maximized on only one screen.
There's always the rendering issue, especially when you do stuff like multiple character sets, rotations, embedded Word objects... The presentations I run are from other people, so the rendering is really important.
Spanned displays. If you want to specify the secondary monitor for your presentation, you have to use PowerPoint. Apparently, there has been a lot of discussion about this, and rudimentary multi-monitor support might make it to the next release.
It's also unbelievably slow and bloated.
Engelbart invented the GUI, not Xerox.
Hey, replacing a broken string can lead to complete inoperability. If you mismatch the strings, the tension will be off, and your tone will be loose. If you use too heavy a heavy gauge of wire, you could even break the guitar's neck.
Guitars are a pretty mature product. Do you have any idea how many centuries it took to make the chromatic scale? Then look at the adoption curve for well-tempered tuning. Upgrading that was not as easy as opening a panel and pushing a button. It just seems easy now because we have a couple centuries of practice with this system. Even so, most people play by magic finger positions (chords) dictated by the experts, and can't tune their instruments.
How long is it going to take before we can agree on good metaphors for computer interfaces, like we have for music interfaces? And how much education will we absolutely require?
My understanding is that a lot of music, especially older than Classical period pieces, do involve a bit of creative work. The source materials are highly fragmentary: They were written before copyright was widespread, and the most widely published (and complete) editions are buggy. Even the authors had conflicting editions, as they had to compose for what instruments were available rather than The Vision of the Reasoning Man with a paying popular audience, and they had misprints. Earlier pieces, especially, were often not written until somebody in the 19th Century decided they were worth preserving (especially folk music), or the manuscripts were considered worthless and discarded (like a lot of Bach's work), or the only copies were destroyed in some random church fire (I'm thinking Schutz).
Any reasonably scholarly edition printed today will have an extensive story about which source manuscripts they used, how they adapted them to modern typographic conventions, what did they change because the editor thought it was a typo, where the ossia come from, and any other details. I think they're mostly found as compilations of, like, 12 Sonatas for Solo Flute from Telemann, or Organ Music by Bach including the entire Trio Sonatas and Orgelbuchlein and several more pieces.
Any purely performance edition will have the editor, the composer of the bass realization, the composer who filled in the whatever number of measures that were lost to history or never composed (think Christopher Tolkien but worse source material), and whomever else was involved in the adaptation. Performance editions often deviate in interesting ways from scholarly editions. They're also much easier to find in stores and libraries, often adapted to instruments that the composer never saw or heard.
The obvious solution is to find the piece that you want on paper (or practice enough to compose or improvise credibly), and if necessary adapt it to your instrument. If you read the liner notes on any reasonably scholarly performance recording, they often describe which sources they used, and how they filled in the missing pieces themselves, largely because they're experts showing off, but probably partially because they can't risk the copyrights.
Ack! Straw man! GIF is used for digital photography only in my nightmares. While the real world is not a 24-bit bitmap, the data from the camera can be pretty nearly represented by it, so GIF is not appropriate. Even for black and white photos, I think the continuous curves don't compress that well with GIF's celebrated LZW.
What would be appropriate is something intentionally simple, most likely an icon for a button or a logo. Drawings naturally have only the colors that you put in them. Sometimes, shading and multilevel transparency are just extra work for the artist, the display device, and the storage device, and sometimes LZ77 inflate takes too much power. As those times become rarer, GIF becomes less relevant.
Of course, Internet Explorer is the spoiler.
"False" back atcha. With that criterion, you might as well say that PNG is lossy because it doesn't accurately reproduce image depth or iridescence.
For an image that is naturally limited to opaque 8-bit color, GIF is certainly lossless. If you adjust the palette correctly, you can represent any appropriate color without dithering. I don't know why some of you photo geeks have such a problem with that.
You forget that you were trained on the WIMP GUI many years ago, and that familiarity continues to make Windows or MacOS easy to use. One thing I find to be fun is to find one of those old introduction to Macintosh books. (The last one I saw was in the back of a cabinet full of manuals in my undergraduate Physics lab.) After using the GUI for such a long time, it's enlightening to see just how many pages are devoted to making sure you understand the WIMP metaphor.
Granted, the WIMP GUI with a good mapping to the system software is much more intuitive than the man/info/HOWTO text-based system for more people. The main difference, then, is how much training you have to do to use a system.
Linux is not an x86-only program. While that's not a problem with a chipset, there's that handful of Mac users running Linux that can't get speedy graphics. Apparently, nVidia can't produce a PPC driver anytime soon, as they have let Apple do the driver development all this time. The only real option is to switch to ATI.
Even if nVidia produces a PPC driver, that wouldn't be any help to the (admittedly small) groups of people using SPARC or Alpha or MIPS as their desktops.
I have to wonder whether you're actually using your system from the students' perspective, because people are used to taking a lot of abuse without complaint. See Bill Gates' comment some years back about Microsoft software not having any significant bugs. See also your term, "almost zero problems".
A friend of mine administers the network at another university, where they use Perfigo's system, on hardware that Perfigo recommends, only for the wireless segment. He says that it's really slow there, and its interface is very clunky, and it just doesn't work well. Furthermore, the company doesn't respond very well to requests for customer support.
However, they did respond with legal threats when he voiced his complaints in a public forum, and he's now too scared to say anything publicly against them. I'm sure that that gives a particular bias to comments in other public forums like this one.
At UCSC, we use Perfigo's SecureSmart servers for making it safe to plug the students' computers into the school network. It's bad.
The server is constantly going down. Get this: It checks every 6 hours to make sure that it's currently registered. Frequently, it forgets that it's registered or Perfigo's registration server scrambles its licenses or something, because the dorm network then goes down. This happens about once per week.
The system is based on a router running Red Hat 7 on commodity x86 machines. Last I heard, it was still using Linux 2.4.9. The upgrade procedure is a drive reimage. The actual routing goes on in a proprietary routing program with fairly low performance. The scanning is done with a customized Nessus. The administration is some custom PHP (IIRC) code, with no security roles and complete control via a single password.
Furthermore, the source to the free software they use, they refuse to send to the customer. Somebody really should see if they can sue Perfigo for violation of the GPL.
Ignoring the above, the Resnet administrator has set up the SecureSmart server to scan PCs for the usual Windows problems. If it finds one, he has it set up to let the user see only antivirus pages and Windows Update. Then it's supposed to scan the user's computer again after 24 hours. What usually happens is that the user's computer doesn't get an IP address anymore, ever, and the administrator has to unblock the specific MAC manually (using his single password).
I'm guessing that we're still using it because the administrator feels that he has invested too much effort into it already. I don't know exactly what Cisco was thinking. Perfigo is just a bad investment.
If you're also going to UCSC, you should check out https://api.alkaid.org/ It's currently a bit out of date, but it shows that the administrator should have known not to use Perfigo.
From what I hear, that person just took names from the forums and added @spymac.net. As noted by other people, you could do the same with Yahoo accounts. And, just like Yahoo, you could choose not to use the email, nor to enter any implicating information.
It doesn't really sound like vigilatism. I'd classify that message as spam, actually. It tells what should be obvious. At worst, it's awfully neglectful of the Spymac operators not to have a large privacy policy that explains such things. Sending email to everybody in the forums isn't a solution, and is likely to cause more confusion.
One serious objection that I have to Spymac, which can be checked out, is that it doesn't use SSL. Even for the paid webhosting and webmail. And all of the services are prone to failing without warning; it's been 7 months or so since the new services came out, and I'm still hearing complaints about their reliability.
Actually, there are some 640x480 B&W LCD panels, but without backlight. One place that sells them, that I mention only because my brother was looking at them recently, is All Electronics http://www.allelectronics.com The 640x480 LCD panel is $25, not counting material and labor needed to connect it.