Graham tries to make an argument that some languages are better than others. That's an abstract enough statement on the face of it that it's hard to take real offense. He quickly backs off from saying anything really controversial, though, because "Programmers get very attached to their favorite languages, and I don't want to
hurt anyone's feelings." So, to illustrate his argument, he creates a cute little straw man, sets him up, and knocks him down.
Programmers get very attached to their favorite languages, and I don't want to
hurt anyone's feelings, so to explain this point I'm going to use a hypothetical
language called Blub. Blub.. is not the most powerful language, but it is more powerful than Cobol or machine language.
When we switch to the point of view of a programmer using any of the lan-guages
higher up the power continuum, however, we find that he in turn looks
down upon Blub. How can you get anything done in Blub? It doesn't even have y.
So, what the hell is Y? If you're a Java zealot, you might say that it's strong typing and bytecode portability. Of course, a C nazi would want to tear his hair out because of how "limiting" all that strong typing is, and how infernally slow the produced code is. Meanwhile, Python geeks will sing the praises of syntactically signifigant source-code formatting, ML nuts will talk about how nifty it is to have your whole program look like it's written in EBNF, and Perl monks will spout off huge strings of acronyms which all serve to hilight the Swiss Army knife nature of their language.
In other words, by failing to take a stand on what makes some languages better than others (other than the bland assessment that the addition of lexical closures in Perl 5 was a good thing), he succeeds in avoiding offense, but utterly fails to say anything useful. "Power" becomes a catch-all abstraction, like a D&D stat, and nobody gets to argue about what features they actually want in a hypothetical uber-language, because that might get someone's panties in a bunch.
And now I'm grumpy because I stopped to write this out instead of studying for my Distributed Object Programming final tonight. Feh.
George Lucas, I think I can speak for everyone here on slashdot, and indeed with the entire breath of humanity: you should be ashamed of yourself for what you have wrought.
I'm just re-reading what I wrote and shaking my head. I completely trashed my home Debian system on Tuesday, and because it was all on a single partition, I was forced to erase everything (except what I could SCP off reasonably). When I put the system back together, I carved up my single partition into seperate mounts for/,/usr,/var, and/dev.
In other words, completely disregard the first paragraph in my previous post.:)
Enh. Your point about the importance of having seperate partitions is well taken, but that's more work than I should have to bother with for a home box, especially one that I dual boot anyway.
Besides which, having the separate partitions would have prevented the original poster from having run out of space on / , but it doesn't really help with the chief problem, which is that Medusa spews an assload of data while logging. Pity, dat.
I kinda hate posting this, just because it's such a predictable old saw, but...
If you're not morally opposed to running KDE, you should give serious thought to trying out Konqueror. It runs using the Gecko rendering engine, but has the added benefit of... well, you know. Not crashing constantly. It also runs very quickly, orders of magnitude faster than the last 'zilla build I tried (m18).
The only problems I've had with Konqueror involve javascript-heavy sites, and I really don't feel I can blame that on the browser.
someone told me "these are the best days of your life, jeremy"... i wanted to kill myself then.
It's funny, but when I was in junior high, I remember talking to my Dad for a while about girls. Not the "this is how the plumbing works" talk, but more prosaic "why don't you ask girls out" kinda stuff.
The thing I remember most about that conversation was that he told me, "Don't believe anything anyone else says to you. These are, bar none, the hardest years of your life. It all gets easier from here." And, he was right. By the time I was a high school junior, I was more or less comfortable with my geekiness, and resolved to just have a good time being me. My senior year, the group of geeks I hung out with mysteriously turned into the most popular group of kids in the school. It was nuts. Large numbers of us still hang out together, ten years later, and we even have actual lives.
I was lucky that things got better for me, I know. But I suspect that the improvement in my circumstances stemmed from an understanding that life wasn't all wine and roses, and I didn't have to act like it was all the time.
If John Deere wants to advertise it's mowers and stuff, what they can do (and probably should!) is to host and design gardening, landscaping, and home-maintainance websites!
There are plenty of examples of this type of site already on the web. One of my personal favorites (WARNING: shameless plug) is the home page for Weber (as in the manufacturer of outdoor grills). I found a fantastic salmon recipe there which I use whenever I have the chance.
The only problem here is that you're never going to see any information critical of Weber on that site. These sites can generate customers, but not unless they come there looking specifically for you. For example, I never bothered to even look at Weber's site until after I'd already selected my grill, a decision made with a little help from Consumer Reports. From an advertising perspective, other than creating a little bit more in the way of "brand awareness" for me, this had precisely zero value. And for information on non-Weber products, I'm going to look somewhere considerably less professional looking, but with a good deal of info.
More importantly, this creates a huge burden on would-be advertisers. It used to be that to sell a product, you got a few (possibly fake) testimonials, whipped up a contest, and generally tried to convince everyone that your product was the best thing since sliced bread. Using the mini-portal method, suddenly every manufacturer of every type of product has to become an information clearinghouse on their own website. Here's a news flash: I'm not looking for everybody who wants to sell to me to impress me with their knowledge of whatever. I'm looking for a good deal. And that's the kind of thing that good old ads - the kind that take a second or two to read - are good for.
does this mean we will have to expect Intel, AMD, and/or Transmeta to adopt this?
Yes. However, prohibitive licensing costs aside, it doesn't sound like this is a particularly expensive process, so I don't see why any chip manufacturer would shy away from this. Of course, I'm no engineer.
And if you read the very next sentence, you would see the part where it said "To be commercially useful, these devices will have to perform several times better than this; this should be possible with further improvements, the researchers estimate."
You're right: not only is the header not a standard HTTP header (standard compliance good! embrace and extend bad!), it's not even easily accessed by the user.
I didn't mean to suggest that this particular header was a good idea: I meant that content negotiation based on bandwidth constraints isn't a bad one.
Yeah, exactly. Content negotiation is a good thing - when you do it the right way. I couldn't tell you why Google has made the decision to use IP address rather than the Accept-Language header to determine what language to serve up files, but obviously, it has a pretty stupid result.
I'll tell you what I'd really like to see (now that it's just occurred to me) - a "Reject-MIME" heading. That way, if I get sick and tired of watching some hack's Flash movies, I could tell the server not to send 'em to me. Or a "Max-Content-Length" heading, so sites wouldn't shoot 5 meg files at me without asking.
I disagree about as strongly as it's possible to disagree. Content negotiation is a Good Thing(tm).
Here's an example: when I go to a web site, I expect (hope?) that the content of the site will be rendered in English. For large web sites with a multi-lingual user base, that's not always a safe assumption. Fortunately, content negotiation makes that possible.
3.10 Language Tags
A language tag identifies a natural language spoken, written, or otherwise conveyed by human beings for communication of information to other human beings. Computer languages are explicitly excluded. HTTP uses language tags within the Accept-Language and Content-Language fields.
- from RFC 2616
Apache makes on-the-fly decisions about what content to send based on this.
Does that mean that webmasters need to be careful about how they set up their sites if they're using this technique? Sure. But it also opens up a wide range of options.
1. It's more expensive to design 2 sets of pages. That money should be spent on more content.
Speaking on behalf of webmasters everywhere: thanks for telling me how to spend my money. Allow me to suggest that doing two versions of the same image - one at a high bit-depth, and another at a lower quality - isn't too much of a strain on my budget.
2. Sometimes people with slow modems don't mind waiting - maybe they let your site load in the background while they do something else. It's not polite to make these choices for your users.
Content negotiation doesn't have to be like making the choice for the user. Instead, it can work as a reasonable best-guess. Besides which, I've seen plenty of sites which simply assume high bandwidth (or pathetic bandwidth) and make all the design decisions based on that information. In what way is that giving the user a choice, other than to vote with his feet?
But we've said it before, Reps don't understand bits and bytes. If you don't send them dead trees, they don't think you vote.
Oh, stop being such a crybaby, Taco. If you received over 1000 emails a day, you wouldn't read them all either. If you were in charge of representing several hundred thousand people to Congress, and kept getting e-mail from the millions you weren't representing, it would probably sour you on the whole thing as well.
Don't forget, these are people whose lives don't revolve around sitting in front of a computer all day. E-mail makes sending a thought to another user nearly effortless, the product of mere seconds of work. In such a position, I'd probably treat the average e-mail with the same amount of gravity I usually reserve for Post-it notes. Less, since the majority of post-its I get are from people I actually know, about things which are actually important.
It'a a neat distinction. Of course, you can say it in English too - "gratis and liberated" - but liberated doesn't have the same mouthfeel as Free-with-a-capital-F
but I doubt that MS will make it fully Linux xompatible (sic)...
I suspect you're right. On the other hand (and speaking as a guy with very precious knowledge of the NET platform), I wouldn't be surprised if there are elements of the system that simply don't make sense for Linux, or that would require a herculean effort in order to implement. And even if that's not the case, it would be a convenient excuse for MS to not even try a full implementation.
This is why I'd rather see a decent API set released for public consumption, so we could get folks interesting in building their own implementations busy. However, that's never going to happen. According to Ballmer, "...our overall strategy is not to get [non-Windows based] Web sites over to Windows, but we will provide a way for those Linux servers to use.NET." In other words, they can't allow competing implementations to pop up, or they run the risk of being hoisted by their own collective petard.
This isn't as trivial a decision as it may sound. A system which, in theory, can interrupt the user every five minutes to deliver a security patch, is gonna get disabled. Excessively onerous "warnings" are almost as much a problem in software design as the absence of warning signs.
For a shocking example, I refer you to "An Investigation of the Therac-25 Accidents." Basically, an X-ray device malfunctioned and killed a whole bunch of people in part because it popped up warning messages as a matter of course. The operators got so desensitized to them that they lost their effectiveness, and people got hurt as a result.
The moral of the story: it's important to warn the user when he's doing something dangerous. It's as important to leave him alone and let him get some work done the rest of the time.
Slashdot Mirror
See, if you'd just said PIASACNASAK, instead of writing the whole thing out, you'd have illustrated my point perfectly. Thank you.
So, what the hell is Y? If you're a Java zealot, you might say that it's strong typing and bytecode portability. Of course, a C nazi would want to tear his hair out because of how "limiting" all that strong typing is, and how infernally slow the produced code is. Meanwhile, Python geeks will sing the praises of syntactically signifigant source-code formatting, ML nuts will talk about how nifty it is to have your whole program look like it's written in EBNF, and Perl monks will spout off huge strings of acronyms which all serve to hilight the Swiss Army knife nature of their language.
In other words, by failing to take a stand on what makes some languages better than others (other than the bland assessment that the addition of lexical closures in Perl 5 was a good thing), he succeeds in avoiding offense, but utterly fails to say anything useful. "Power" becomes a catch-all abstraction, like a D&D stat, and nobody gets to argue about what features they actually want in a hypothetical uber-language, because that might get someone's panties in a bunch.
And now I'm grumpy because I stopped to write this out instead of studying for my Distributed Object Programming final tonight. Feh.
Perhaps, but does the room really understand Chinese?
Aha. In which case, the original poster said nothing about his own level of experience.
He has Linux experience: he's just prejudiced against products that claim to obviate the need for it.
Anyway, he'd better have experience, or he has no place trying to set up his own webhosting company and configure a user control panel himself.
I'm just re-reading what I wrote and shaking my head. I completely trashed my home Debian system on Tuesday, and because it was all on a single partition, I was forced to erase everything (except what I could SCP off reasonably). When I put the system back together, I carved up my single partition into seperate mounts for /, /usr, /var, and /dev.
:)
In other words, completely disregard the first paragraph in my previous post.
Enh. Your point about the importance of having seperate partitions is well taken, but that's more work than I should have to bother with for a home box, especially one that I dual boot anyway.
Besides which, having the separate partitions would have prevented the original poster from having run out of space on / , but it doesn't really help with the chief problem, which is that Medusa spews an assload of data while logging. Pity, dat.
I kinda hate posting this, just because it's such a predictable old saw, but...
If you're not morally opposed to running KDE, you should give serious thought to trying out Konqueror. It runs using the Gecko rendering engine, but has the added benefit of... well, you know. Not crashing constantly. It also runs very quickly, orders of magnitude faster than the last 'zilla build I tried (m18).
The only problems I've had with Konqueror involve javascript-heavy sites, and I really don't feel I can blame that on the browser.
It's funny, but when I was in junior high, I remember talking to my Dad for a while about girls. Not the "this is how the plumbing works" talk, but more prosaic "why don't you ask girls out" kinda stuff.
The thing I remember most about that conversation was that he told me, "Don't believe anything anyone else says to you. These are, bar none, the hardest years of your life. It all gets easier from here." And, he was right. By the time I was a high school junior, I was more or less comfortable with my geekiness, and resolved to just have a good time being me. My senior year, the group of geeks I hung out with mysteriously turned into the most popular group of kids in the school. It was nuts. Large numbers of us still hang out together, ten years later, and we even have actual lives.
I was lucky that things got better for me, I know. But I suspect that the improvement in my circumstances stemmed from an understanding that life wasn't all wine and roses, and I didn't have to act like it was all the time.
There are plenty of examples of this type of site already on the web. One of my personal favorites (WARNING: shameless plug) is the home page for Weber (as in the manufacturer of outdoor grills). I found a fantastic salmon recipe there which I use whenever I have the chance.
The only problem here is that you're never going to see any information critical of Weber on that site. These sites can generate customers, but not unless they come there looking specifically for you. For example, I never bothered to even look at Weber's site until after I'd already selected my grill, a decision made with a little help from Consumer Reports. From an advertising perspective, other than creating a little bit more in the way of "brand awareness" for me, this had precisely zero value. And for information on non-Weber products, I'm going to look somewhere considerably less professional looking, but with a good deal of info.
More importantly, this creates a huge burden on would-be advertisers. It used to be that to sell a product, you got a few (possibly fake) testimonials, whipped up a contest, and generally tried to convince everyone that your product was the best thing since sliced bread. Using the mini-portal method, suddenly every manufacturer of every type of product has to become an information clearinghouse on their own website. Here's a news flash: I'm not looking for everybody who wants to sell to me to impress me with their knowledge of whatever. I'm looking for a good deal. And that's the kind of thing that good old ads - the kind that take a second or two to read - are good for.
You're not imagining the DiVX site.
Yeah. Unless it's vim. Or vile. Or elvis. Or nvi.
Yes. However, prohibitive licensing costs aside, it doesn't sound like this is a particularly expensive process, so I don't see why any chip manufacturer would shy away from this. Of course, I'm no engineer.
And if you read the very next sentence, you would see the part where it said "To be commercially useful, these devices will have to perform several times better than this; this should be possible with further improvements, the researchers estimate."
You're right: not only is the header not a standard HTTP header (standard compliance good! embrace and extend bad!), it's not even easily accessed by the user.
I didn't mean to suggest that this particular header was a good idea: I meant that content negotiation based on bandwidth constraints isn't a bad one.
-----
"You owe me a case of beer. Sucka'."
Yeah, exactly. Content negotiation is a good thing - when you do it the right way. I couldn't tell you why Google has made the decision to use IP address rather than the Accept-Language header to determine what language to serve up files, but obviously, it has a pretty stupid result.
I'll tell you what I'd really like to see (now that it's just occurred to me) - a "Reject-MIME" heading. That way, if I get sick and tired of watching some hack's Flash movies, I could tell the server not to send 'em to me. Or a "Max-Content-Length" heading, so sites wouldn't shoot 5 meg files at me without asking.
-----
"You owe me a case of beer. Sucka'."
I disagree about as strongly as it's possible to disagree. Content negotiation is a Good Thing(tm).
Here's an example: when I go to a web site, I expect (hope?) that the content of the site will be rendered in English. For large web sites with a multi-lingual user base, that's not always a safe assumption. Fortunately, content negotiation makes that possible.
Apache makes on-the-fly decisions about what content to send based on this.
Does that mean that webmasters need to be careful about how they set up their sites if they're using this technique? Sure. But it also opens up a wide range of options.
Speaking on behalf of webmasters everywhere: thanks for telling me how to spend my money. Allow me to suggest that doing two versions of the same image - one at a high bit-depth, and another at a lower quality - isn't too much of a strain on my budget.
Content negotiation doesn't have to be like making the choice for the user. Instead, it can work as a reasonable best-guess. Besides which, I've seen plenty of sites which simply assume high bandwidth (or pathetic bandwidth) and make all the design decisions based on that information. In what way is that giving the user a choice, other than to vote with his feet?
-----
"You owe me a case of beer. Sucka'."
Oh, stop being such a crybaby, Taco. If you received over 1000 emails a day, you wouldn't read them all either. If you were in charge of representing several hundred thousand people to Congress, and kept getting e-mail from the millions you weren't representing, it would probably sour you on the whole thing as well.
Don't forget, these are people whose lives don't revolve around sitting in front of a computer all day. E-mail makes sending a thought to another user nearly effortless, the product of mere seconds of work. In such a position, I'd probably treat the average e-mail with the same amount of gravity I usually reserve for Post-it notes. Less, since the majority of post-its I get are from people I actually know, about things which are actually important.
-----
"You owe me a case of beer. Sucka'."
-----
"You owe me a case of beer. Sucka'."
And God only knows, it's hard to find porn on the Internet nowadays without a credit card.
-----
"You owe me a case of beer. Sucka'."
I suspect you're right. On the other hand (and speaking as a guy with very precious knowledge of the NET platform), I wouldn't be surprised if there are elements of the system that simply don't make sense for Linux, or that would require a herculean effort in order to implement. And even if that's not the case, it would be a convenient excuse for MS to not even try a full implementation.
This is why I'd rather see a decent API set released for public consumption, so we could get folks interesting in building their own implementations busy. However, that's never going to happen. According to Ballmer, "...our overall strategy is not to get [non-Windows based] Web sites over to Windows, but we will provide a way for those Linux servers to use .NET." In other words, they can't allow competing implementations to pop up, or they run the risk of being hoisted by their own collective petard.
-----
"You owe me a case of beer. Sucka'."
This isn't as trivial a decision as it may sound. A system which, in theory, can interrupt the user every five minutes to deliver a security patch, is gonna get disabled. Excessively onerous "warnings" are almost as much a problem in software design as the absence of warning signs.
For a shocking example, I refer you to "An Investigation of the Therac-25 Accidents." Basically, an X-ray device malfunctioned and killed a whole bunch of people in part because it popped up warning messages as a matter of course. The operators got so desensitized to them that they lost their effectiveness, and people got hurt as a result.
The moral of the story: it's important to warn the user when he's doing something dangerous. It's as important to leave him alone and let him get some work done the rest of the time.
-----
"You owe me a case of beer. Sucka'."
Seriously. And I even like some of Katz's stuff. But quips like that make me wonder if he's ever actually bothered reading this site.
-----
"You owe me a case of beer. Sucka'."
...except, of course, right in the Slashdot headline. So technically, /. is in violation of the DMCA by publishing the algorithim. Heathens.
-----
"You owe me a case of beer. Sucka'."