I agree with not liking Flash as a navigational tool. I don't like sites designed around flash, and find it irritating that I hear beeping noises just by mouse-overing things, or that I can't bookmark specific pages.
However, I think many anti-Flash people go too far and dislike it entirely, while I do think it has its uses. It's currently the best format for vector-graphics animation with any degree of widespread support. If you want to make web animated comics, you use Flash. (You certainly shouldn't use MPEG or some other raster format.) It also has some uses for interactive manipulatable graphs and so on.
The whole point of SVMs is that they can be used to model a linear decision boundary. They were developed to find a maximum-margin hyperplane separating positive and negative training instances, and the kernel methods to allow them to work on non-linear boundaries were a later addition.
That's only really true of Scandinavia and some other parts of northern Europe. Greece, for example, has been largely Christian since Christianity was first started. It's true that the ancient Greek religions were pagan, but they're hardly pagan in a way that's closely related to northern European paganism (and certainly not to neopaganism), and the development of modern Greek culture has been far more influenced by 2000 years of Christianity, particularly the Byzantine Empire.
Similar things might be said of Italy, Spain, and Portugal.
[Note that I'm personally not a member of any religion.]
Sure, the firmware isn't Free, but neither is the firmware actually loaded in your motherboard's EEPROM chips. You don't see people raising a ruckus about how they refuse to purchase motherboards on which the firmware is not Free Software, so why are they worried about this? This firmware is pretty tightly coupled to the hardware in a similar way as the EEPROM firmware is.
Now maybe if people were going to an 100% Free system in which every single piece of their computer was Free, then I'd see the point, but if you're not going to do that anyway, I don't really see the advantage of causing a huge hassle over this relatively minor issue.
This is just going to make people care less about free software. If Debian is lumping documentation from the Free Software Foundation and licensed under the GNU Free Documentation License under 'non-free', that tends to look a little bit pedantic to most of the outside world, which already sees GNU as being pretty hardline---but apparently GNU are the reasonable moderates here!
All this is going to do is make lots more people use 'non-free', so Debian will be officially Free Software, but everyone will use everything else anyway. That's counter-productive, because for the purpose of avoiding some Invariant Sections in GFDL'd documents, you've pushed a bunch of people towards using actually non-free software, since once you move to 'non-free' to get around their stupid GFDL rules, it's wide open from there on out.
Don't get me wrong, I love Debian, and use it myself, but the installer is downright crappy, typically requiring a bunch of manual editing of kernel module configurations and whatnot to get a system to install (usually with the aid of some HOWTOs). Knoppix is Debian-based and Just Works, auto-detecting everything fine---and it's Free Software. Why doesn't Debian just borrow their installer or something?
Well, I'll admit it's no longer fashionable, but many people still at least try to keep up the pretenses of neutrality rather than blatantly admitting that they're trying to indoctrinate students with a particular opinion.
You are correct though that less overtly it happens a lot. Seems to happen in both "conservative" and "liberal" areas, though the "liberal" ones seem to be worse: lots of classes in Women's Studies departments on abortion, for example, start with the assumption that abortion is fully moral and a right. No reading of material from both sides or any of that old-fashioned nonsense.
Students in philosophy classes might be asked, for example, to "write comparing Kantian ethics with a utilitarian system of your choice." They are rarely asked to "write an essay on why Kantian ethics is superior to alternate systems of ethics." Even in areas where everyone agrees this isn't done: A history class doesn't ask the students to write a paper on why Adolf Hitler was a bad man. The entire point of education is to critically think through ideas, not to have a conclusion assigned before you've even started.
MP3s from 1997 may well be non-transparent, but it's very difficult in double-blind listening tests for people to distinguish a good VBR mp3 made using a perceptually-tuned preset (say, LAME --preset standard, and especially LAME --preset extreme) from the original CD. There are a few isolated codec-killer cases that are distinguishable by people who have trained to listen for specific artifacts (mostly cases of pre-echo), but they're not that common.
I buy the CDs I like quite a bit, and expect to listen to many, many times. That's worth $16 to me. That's about the price of two movie tickets, and a typical CD over the dozens of times I listen to it gives me a lot more enjoyment than seeing two movies does.
the first one makes it difficult
on
A Taste of Qt 4
·
· Score: 1
No GPL version for Windows makes it impractical for a lot of cross-platform applications. Gaim, for example, uses GTK for both its Linux and Windows versions, but this would be impossible to do with Qt.
A ton of people use AOL for this: There are dozens of chatrooms with automated bots that list warez and mp3s and whatnot and will auto-forward you emails with them attached. This is very efficient for the warez distributors because forwarding an email takes no bandwidth on their part, since it's already on the server end, so they can serve dozens of people off a dial-up modem once the files have been uploaded once.
AOL doesn't seem to care all that much. If anything it's another sort of clandestine plus of AOL: easy warez! Sort of how they were "anti-porn" in the early days, but everyone knew porn-trading was a big part of their userbase.
If you have a huge disk quota and webmail, emailing files to yourself is the most accessible way of moving files around, especially to/from kiosk computers that may not have anything useful installed besides a webbrowser. I do it myself even with my relatively small space quota.
There have been some pretty bad remote-root Linux holes. If 90% of the world's desktops had been running Linux, you can bet there would be worms exploiting them. In fact, back when the internet was mostly Unix, this very thing happened with the Morris worm.
Several universities I'm familiar with used to use Suns for all their web- and mail-servers, but they're for the most part switching to x86, because it's simply a lot cheaper---and faster. Replacing a 4-year-old 300 MHz CPU Sun server for an $800 dual-2GHz CPU box from Dell works wonders for responsiveness of the IMAP server. The newer Sun hardware that can compete at that level is far more expensive.
If the database only included people who have, say, filed two claims that were both rejected by the courts, then I'd be okay with it. There are some people out there who will sue for everything that is completely baseless, and these people shouldn't be allowed to keep wasting doctors' time and money.
If the lawsuit was successful on the other hand, then this isn't the way to go. Either it was a valid claim, in which case there is no fault with the person who sued, or it was an invalid claim that was somehow upheld by the courts, in which case the proper remedy is legislation to change that.
If you're looking for really earthshattering ideas that every computer scientist would know, like, say, the idea of reducibility to NP-complete problems, I'm not aware of any recent ones. But if you're willing to go for more narrowly-applicable ones, there's a bunch. There's been a few new transforms in the last five or so years that have been applied to audio compression, though many are just variants on older ones, so maybe aren't "really" new. Machine learning has a bunch of stuff as well: support vector machines were invented in 1998 I think, to pick one example.
Just out of curiosity, since I know and use SML but have only heard of Ocaml: what're the relative strengths? Is there a reason for me to learn Ocaml as well? I gather that they're both derived from the original ML, and share a lot of features, but I was wondering whether I ought to care about Ocaml or just stick with SML (since I already know it).
People are pretty disillusioned, but most predictions are that of those who bothered to vote at all, most would've voted for a reformist over one of the conservatives. Without the 2000 disqualifications, the reformists would probably have had 200 or so of the 290 seats, and some predictions were as high as 240. Remember, it's a first-past-the-post system, so a conservative would've had to win 50% of the votes in an area to get a seat, and the conservatives only have that level of support in the less-populated rural areas.
Slashdot Mirror
I agree with not liking Flash as a navigational tool. I don't like sites designed around flash, and find it irritating that I hear beeping noises just by mouse-overing things, or that I can't bookmark specific pages.
However, I think many anti-Flash people go too far and dislike it entirely, while I do think it has its uses. It's currently the best format for vector-graphics animation with any degree of widespread support. If you want to make web animated comics, you use Flash. (You certainly shouldn't use MPEG or some other raster format.) It also has some uses for interactive manipulatable graphs and so on.
The whole point of SVMs is that they can be used to model a linear decision boundary. They were developed to find a maximum-margin hyperplane separating positive and negative training instances, and the kernel methods to allow them to work on non-linear boundaries were a later addition.
I think Christianity should be mentioned since its role in Europe's history was indeed crucial.
What, crucial in terms of causing loads of wars and strife?
Yep---exactly like the pagans. You don't think the Norse warrior epics were stories about peace-loving warriors, do you?
That's only really true of Scandinavia and some other parts of northern Europe. Greece, for example, has been largely Christian since Christianity was first started. It's true that the ancient Greek religions were pagan, but they're hardly pagan in a way that's closely related to northern European paganism (and certainly not to neopaganism), and the development of modern Greek culture has been far more influenced by 2000 years of Christianity, particularly the Byzantine Empire.
Similar things might be said of Italy, Spain, and Portugal.
[Note that I'm personally not a member of any religion.]
Well, they could search ebay.
Sure, the firmware isn't Free, but neither is the firmware actually loaded in your motherboard's EEPROM chips. You don't see people raising a ruckus about how they refuse to purchase motherboards on which the firmware is not Free Software, so why are they worried about this? This firmware is pretty tightly coupled to the hardware in a similar way as the EEPROM firmware is.
Now maybe if people were going to an 100% Free system in which every single piece of their computer was Free, then I'd see the point, but if you're not going to do that anyway, I don't really see the advantage of causing a huge hassle over this relatively minor issue.
It's not hard to have it set up so it only tries to auto-detect i386 hardware if it's installing on an i386 system. Don't see the problem at all.
This is just going to make people care less about free software. If Debian is lumping documentation from the Free Software Foundation and licensed under the GNU Free Documentation License under 'non-free', that tends to look a little bit pedantic to most of the outside world, which already sees GNU as being pretty hardline---but apparently GNU are the reasonable moderates here!
All this is going to do is make lots more people use 'non-free', so Debian will be officially Free Software, but everyone will use everything else anyway. That's counter-productive, because for the purpose of avoiding some Invariant Sections in GFDL'd documents, you've pushed a bunch of people towards using actually non-free software, since once you move to 'non-free' to get around their stupid GFDL rules, it's wide open from there on out.
Don't get me wrong, I love Debian, and use it myself, but the installer is downright crappy, typically requiring a bunch of manual editing of kernel module configurations and whatnot to get a system to install (usually with the aid of some HOWTOs). Knoppix is Debian-based and Just Works, auto-detecting everything fine---and it's Free Software. Why doesn't Debian just borrow their installer or something?
Well, I'll admit it's no longer fashionable, but many people still at least try to keep up the pretenses of neutrality rather than blatantly admitting that they're trying to indoctrinate students with a particular opinion.
You are correct though that less overtly it happens a lot. Seems to happen in both "conservative" and "liberal" areas, though the "liberal" ones seem to be worse: lots of classes in Women's Studies departments on abortion, for example, start with the assumption that abortion is fully moral and a right. No reading of material from both sides or any of that old-fashioned nonsense.
Students in philosophy classes might be asked, for example, to "write comparing Kantian ethics with a utilitarian system of your choice." They are rarely asked to "write an essay on why Kantian ethics is superior to alternate systems of ethics." Even in areas where everyone agrees this isn't done: A history class doesn't ask the students to write a paper on why Adolf Hitler was a bad man. The entire point of education is to critically think through ideas, not to have a conclusion assigned before you've even started.
MP3s from 1997 may well be non-transparent, but it's very difficult in double-blind listening tests for people to distinguish a good VBR mp3 made using a perceptually-tuned preset (say, LAME --preset standard, and especially LAME --preset extreme) from the original CD. There are a few isolated codec-killer cases that are distinguishable by people who have trained to listen for specific artifacts (mostly cases of pre-echo), but they're not that common.
I buy the CDs I like quite a bit, and expect to listen to many, many times. That's worth $16 to me. That's about the price of two movie tickets, and a typical CD over the dozens of times I listen to it gives me a lot more enjoyment than seeing two movies does.
No GPL version for Windows makes it impractical for a lot of cross-platform applications. Gaim, for example, uses GTK for both its Linux and Windows versions, but this would be impossible to do with Qt.
A ton of people use AOL for this: There are dozens of chatrooms with automated bots that list warez and mp3s and whatnot and will auto-forward you emails with them attached. This is very efficient for the warez distributors because forwarding an email takes no bandwidth on their part, since it's already on the server end, so they can serve dozens of people off a dial-up modem once the files have been uploaded once.
AOL doesn't seem to care all that much. If anything it's another sort of clandestine plus of AOL: easy warez! Sort of how they were "anti-porn" in the early days, but everyone knew porn-trading was a big part of their userbase.
If you have a huge disk quota and webmail, emailing files to yourself is the most accessible way of moving files around, especially to/from kiosk computers that may not have anything useful installed besides a webbrowser. I do it myself even with my relatively small space quota.
There have been some pretty bad remote-root Linux holes. If 90% of the world's desktops had been running Linux, you can bet there would be worms exploiting them. In fact, back when the internet was mostly Unix, this very thing happened with the Morris worm.
Several universities I'm familiar with used to use Suns for all their web- and mail-servers, but they're for the most part switching to x86, because it's simply a lot cheaper---and faster. Replacing a 4-year-old 300 MHz CPU Sun server for an $800 dual-2GHz CPU box from Dell works wonders for responsiveness of the IMAP server. The newer Sun hardware that can compete at that level is far more expensive.
If the database only included people who have, say, filed two claims that were both rejected by the courts, then I'd be okay with it. There are some people out there who will sue for everything that is completely baseless, and these people shouldn't be allowed to keep wasting doctors' time and money.
If the lawsuit was successful on the other hand, then this isn't the way to go. Either it was a valid claim, in which case there is no fault with the person who sued, or it was an invalid claim that was somehow upheld by the courts, in which case the proper remedy is legislation to change that.
wow!!! who would've thought??? and not only that!!! finally an article with enthusiasm!!!
Given that the post you were replying to has a Subject of "McKellen Up For It" ...
If you're looking for really earthshattering ideas that every computer scientist would know, like, say, the idea of reducibility to NP-complete problems, I'm not aware of any recent ones. But if you're willing to go for more narrowly-applicable ones, there's a bunch. There's been a few new transforms in the last five or so years that have been applied to audio compression, though many are just variants on older ones, so maybe aren't "really" new. Machine learning has a bunch of stuff as well: support vector machines were invented in 1998 I think, to pick one example.
Just out of curiosity, since I know and use SML but have only heard of Ocaml: what're the relative strengths? Is there a reason for me to learn Ocaml as well? I gather that they're both derived from the original ML, and share a lot of features, but I was wondering whether I ought to care about Ocaml or just stick with SML (since I already know it).
I don't think it's a first-past-the-post system. Regardless, the 200-240 estimates are ones I've heard.
People are pretty disillusioned, but most predictions are that of those who bothered to vote at all, most would've voted for a reformist over one of the conservatives. Without the 2000 disqualifications, the reformists would probably have had 200 or so of the 290 seats, and some predictions were as high as 240. Remember, it's a first-past-the-post system, so a conservative would've had to win 50% of the votes in an area to get a seat, and the conservatives only have that level of support in the less-populated rural areas.