True, I wouldn't shove vi on any newbie at first. But lynx does allow arrow keys as well as the vi-style keys (which he could work up to), and it also allows command-line urls. So all he'd have to remember is "lynx [name of web site]" and 'q' for 'quit' at first. I only say that would be easier because I have sat with my grandma and AOL. "Now move your hand to the mouse and double-click on 'NEWMSG'. No, the other button. Click with the other button. You have to click twice. Click twice. Click twice faster." It's a headache regardless of the interface; at least text-only interfaces pose fewer physical challenges.
"So, what do you want to hack for, Bickle?" "I can't sleep nights." "They got porno theaters for that."
The simplest computer to use, bar none, was the Apple ][. The GUI stuff only seems easy after you "get it." Just because something is graphical doesn't make it easier -- after having used a mac for a while, windows 3.1 was confusing as hell. (ah, memories.) There are a ton of different "metaphors" in use for various desktop environments. (What happens when I right-click on an icon in Windows? How about on a Mac? KDE? Gnome? RH5.2-style AfterStep? It's all different.) Plus, if he has any arthritis, moving his hands from the keyboard to the mouse will be excruciating. It was for my grandma, after my uncle set her up to use AOL.
Give him a text-based email client, teach him IRC and usenet, and teach him lynx and screen (both of which can use vi-style home row keys, which means he doesn't have to move his hands around the keyboard as much). Put netscape on there if he finds himself *really needing* to see graphical stuff, which turns out to be less often than one might think.
"So, what do you want to hack for, Bickle?" "I can't sleep nights." "They got porno theaters for that."
Keep in mind that I was trying to defend the anti-Linux Sun people.
Back during the benchmarking circus, Mindcraft published hate mail from Linux people. It was deplorable of people to send this hate mail, don't get me wrong. What Mindcraft did was to publish this hate mail under the guise of: "Here? See what these Linux people are like? Do you want to base your company on stuff written by these folks?" In some slashdot discussions it was pointed out that, by publishing the hate mail, they were spreading FUD.
Similarly, I dislike the fact that Sun advocates sent hate mail to upside. But, just as most Linux folks do not want to be seen as fanatics just because some loonies like the OS, all of Sun should not be demonized because some wackos called an upside journalist an "idiot."
...
Re-reading your post, I think this answers your question. What exactly was your criticism?
"So, what do you want to hack for, Bickle?" "I can't sleep nights." "They got porno theaters for that."
Remember following the Mindcraft follies, when Mindcraft published all its hate mail as a way to misrepresent Linux/OSS advocates? That was a shame, both that some zealots would write hate mail and that Mindcraft would exploit it. While it's too bad that Sun advocates would send hate mail to upside, I'm disappointed in upside for publishing it. Not all Sun employees (I speculate) are short-sighted assholes. They shouldn't be represented as such.
"So, what do you want to hack for, Bickle?" "I can't sleep nights." "They got porno theaters for that."
I'll buy one for my office. My boss and co-workers will need to be able to install and run Debian, although their linux experience is slight. There's no way I'm going to trust our severs to some distro that tries to be 'idiot-proof', but I need some way for them to get up to speed. They would probably not love having to read all their docs online, and they'd probably prefer a CD to an internet-based or even lan-based install. This book is a good answer, as I haven't seen that Debian Installation and Usage book I've heard about in stores anywhere around here.
"So, what do you want to hack for, Bickle?" "I can't sleep nights." "They got porno theaters for that."
Good lord! "You dish it out, you've got to take it too." So because proprietary licenses are mean-spirited, which give RMS the excuse to be hostile and abrasive, which gives ESR the excuse to be derisive of RMS, you are given the excuse to rail on ESR?
It's easy to inadvertently be an asshole; myopia is not your friend. Remember when WRS died, and those posts showed up saying "Stevens didn't like Linux; I jizz on his corpse"? That poster(s?) was so short-sighted, so focused on advocacy that s/he stooped to the level of...well, the level s/he stooped to.
The philosophy "you dish it out, you've got to take it" has not done much good for the world. You're too clueful a person to follow it.
They seemed to be flinging those "check whether the code will cause error or exception"s around pretty liberally. I wonder how deep that will go (and how much computation time it would take!); and wouldn't sandboxing, not "code checking", really be what's necessary (if one wants security) when swapping in new processor states?
This is pretty damned funny, but I question its appropriateness. Support technicians should not be revealing specific details of support calls. While it's no big deal in this case, it has been a big deal in other cases (such as the Harvard divinity prof who was fired for computer porn). Doctors make sure not to reveal non-relevant specifics in public discourse; if computer professionals are to be respected, they must act in a similar way.
True, the specific name of the person was changed, but, amusing as this is, it's still inappropriate. The job of a support staff member is to help users, not judge them.
> In order to get the program into shape, I email > it to a friend to test and report back with a bug > list. I admonish him not to release it until I > finish it. Am I breaking the law?
If it is already under the GPL, and you try to add restrictions onto that previously-GPLed code, then yes, I speculate that you would be breaking the law. But 1) the code isn't necessarily GPLed yet; and 2) you're not changing the licensing restrictions, you're just asking your friend to do you a favor and not redistribute yet. Your friend is (I assume) under no legal compulsion to not redistribute the code, and (I assume) he could legally do so if he really wanted to (even though doing so would be evil).
I agree with your post, except for your comments on virus prevention. Virus scanners are a fundamentally incomplete solution. Choice of an operating system does matter, but not for obscurity reasons. OpenBSD is secure because it has been designed (and verified) to be that way; it is not secure because it is "unusual". And if one has a secure platform, one should be able to run untrusted binaries (provided they are not run by a privileged user such as root).
Obviously you don't want anything that's actually turing-complete
True, we'd probably want an upper limit on the time and space resources of the programs. (Traipsing offtopic: Given that it only has finite space resources, is my desktop machine really turing-complete?) There would be other practical concerns too, like the granularity with which programs could query the database of slashdot articles. (Raw text? Keywords?)
However, without a certain amount of problem-specific hacking, people's filter scripts won't have any useful data to work with. Without such primitives as "moderated up", "moderated down", and "adjective:funny", what will your script look for? "Show me all the posts from people that have never replied to posts with the words 'hemos sux' in them"?
A sufficently general scripting language would have the immediate benefit of dramatically simplifying the current preferences page. That page largely has to do with filtering and customization. (Strip sigs, block Jon Katz, block posts with "hemos sux", etc.)
I agree that we'd need a set of primitives -- indeed, a really large set of primitives would probably be a good idea. But much of the debate about moderation comes from things like karma, sorting, granularity, etc. -- not the primitives themselves. These issues would be resolved with a general enough scripting system.
(Maybe it's the crack talking, but such a scripting language might be able to put some moderation administration into user-space as well. Those users who think there should be more points circulating around could voluntarily assign each other "unofficial" moderator points in an alternative [complementary] system.)
The only real solution to all of this is a turing-complete filtering/sorting/query language. That's what all of this is expanding into. And writing the most general solution once can be easier than coding up a bunch of idiosyncratic, incomplete solutions. Code a solution to the problem domain, Malda, and solving a specific problem will be (relatively) easy.
The last time I read an explanation of this that was posted by Malda, it said the number of unused moderations points is always constant. Once a moderator spends a point, a new point is given to another moderator. So in theory, no points are ever "wasted" (except from that moderator's perspective -- J. Random Moderator probably wants to save up h{is|er} points for exceptional posts). If all moderators were to spend their points fast and furiously, in theory more points would end up being used. (But then, I could be totally misunderstanding things. Perhaps in the year 2525 when Rob releases the source code again, we'll find out for sure.)
Because when you moderate, you're supposed to be regulating the quality, not the content, of the discussion. If you have issues about content (which is why you would post), it's less likely that you will be an impartial judge of quality.
This is interesting -- it sounds like MetaModeration, in practice, will only give negative feedback. MetaModerators exist to keep evil moderators in line; thus mm's will likely spend their time searching for badly-moderated posts. I'd be surprised if any moderators end up with positive moderation karma. (And BTW, Rob does mean that there will be a seperate "moderation" karma, right? If MetaModeration affects regular karma, I think a lot of people will opt out of moderation.)
In a way, moderators regulate other moderators already. If a post is marked down unfairly, other moderators will come along and mark it back up. (This assumes that there are an adequate number of moderator points floating around, which IMHO is not currently the case.) This is already a form of "meta-moderation".
1) If I just click "post anonymously" instead of logging out, and my post gets moderated down, does my karma go down? That's the whole point of posting anonymously -- to be able to say potentially contreversial things without reprimand (beyond flames).
2) To this end, could the "no score +1 bonus" allow one to decrease one's score to 0 or even -1? Sometimes posts are offtopic but still valid, and so should be posted but pre-moderated down.
3) (This is a little offtopic, but at least regards improving slashdot.) Remember how Alan Cox said that (paraphrased) "the problem with slashdot is that everyone tries to get 'First Post' instead of 'First Patch'"? Well, why not post bug tracking info on slashdot? Registered users could filter it out. All Rob would have to do would be to give story-posting privs to list maintainers. Then we'd start getting the slashdot effect on bugs.
MS will probably want to cash in on the Linux hype, and want to put out some competition for it in the OS market. But it would make a lot more sense for them to take a BSD and proprietize it, rather than going with a GPLed kernel. (The re-licensing freedom of the BSD is probably a reason why Apple chose it for Darwin; Woz's ties to Berkeley might be another.) They'd then either need to rewrite any BSD components under the GPL (I can't recall which components of, say, FreeBSD those are) or slap a Win98/NT/etc layer on top of BSD and run user stuff there (leaving admin tasks in the BSD layer). The resulting "Winux" would have a UNIX administrative interface (which would make a lot of [not all] sysadmins happy), the MS brand name and possibly the Win98/NT/etc interface (which would make a lot of [not all] PHBs and technophobe users happy) and would be proprietary (which would make MS happy all over).
DISCLAIMER: This is not a comment on the relative technical merits of Linux, the BSDs, or the Windowses. I'm just saying that FROM MS'S PERSPECTIVE, repackaging a BSD would be a more likely choice from the licensing angle.
Foogle> This new rule hadn't been implemented then
You've misidentified the new rule. For several months, the rule was that a poster with karma of -10 or below is given a default score of zero. The new rule (which leaves the old rule intact) is that a poster with karma of -20 or below is given a default posting score of -1.
If the US powers-that-be really wanted to increase voter turnout, they'd eliminate the prerequisite of voter registration. Registering is annoying and decreases the likelihood that young people will vote. The actual process of voting isn't all that bad -- just find the building and do it. Making registration easier (by eliminating it, or allowing voters to register at the voting site on the day of the vote) is the key.
It appears from a post elsewhere in this discussion that Ellis-D is the other default -1 poster. That sucks: I've seen good posts from Ellis-D before. And DaveO got a lot of -1's when he first came on/., mainly because he posted in an abrasive, Meept-like way (all caps, third person).
But everyone, at some time, has posted stuff that's badly written. DaveO has at least caught on that all-caps posts are bad, and one of these days he'll realize the same about the third-person voice.
I agree with Malda that something ought to be changed to prevent a repeat of the "I jizz on WRS's corpse" posts. However, scapegoating DaveO and Ellis-D will solve nothing.
1) So say someone on my ISP posts a bunch of trolls and logs off, and that when I come home from work and go online, I get assigned that IP. Will I be blocked from posting? (Rob's essay said that if your account *or* IP had a sufficently bad record, you'd be blocked. Why isn't that an "and"?)
2) I've marvelled that there's no ad-spam on slashdot. Do you guys have a team of grunts that deletes those, or do the folks who flood USENET with "Pamela suck off Tommy Lee" posts just not peruse slashdot? If there is a troupe of squirells to delete ads, maybe they can be used to delete the really bad trolls.
3) What were the offending posts that inspired this?
> That has to be the most oddball essay on a > programming language I have seen recently.
You must not be a member of ACM's SIGPLAN. The weirdest programming language essay I've read was in Volume 33, number 12 of SIGPLAN Notices. An essay therein, "Evolution of the High Level Programming Languages", by Masud Ahmad Malik of Amman, Jordan, derided Ada as being bad for mission critical situations (precisely the domain for which Ada is well-suited). The essay was revealed in Volume 34, number 2 of Notices to be not just innacurate and fourteen years out-of-date, but also plagiarized! It prompted (relatively) angry letters from Ada advocates and an apology by the editors. *That* was a fucked-up comparative programming language essay.
Subnetting under IPv4 is a good idea not fully realized. While it is useful in its present form, it is flawed in that it allows the IP address space to be divided up wastefully. Use of a superset of the current subnetting functions will be needed to remedy address waste.
At least, I think that's what he's trying to say.
(Hey, isn't there an term used in OS-theory circles regarding overly-general division of resources? I recall it from the memory management chapter...can't recall the term....)
> It would make sense that Antimatter Galaxies > may exist, simply because we have found > antimatter.
If I understand the article correctly (IANAP), antiprotons are sometimes found as the result of (matter) particle collisions. They're looking specifically for anti-helium, which is unlikely to come about due to the random collision of matter.
Slashdot Mirror
"So, what do you want to hack for, Bickle?"
"I can't sleep nights."
"They got porno theaters for that."
Give him a text-based email client, teach him IRC and usenet, and teach him lynx and screen (both of which can use vi-style home row keys, which means he doesn't have to move his hands around the keyboard as much). Put netscape on there if he finds himself *really needing* to see graphical stuff, which turns out to be less often than one might think.
"So, what do you want to hack for, Bickle?"
"I can't sleep nights."
"They got porno theaters for that."
Back during the benchmarking circus, Mindcraft published hate mail from Linux people. It was deplorable of people to send this hate mail, don't get me wrong. What Mindcraft did was to publish this hate mail under the guise of: "Here? See what these Linux people are like? Do you want to base your company on stuff written by these folks?" In some slashdot discussions it was pointed out that, by publishing the hate mail, they were spreading FUD.
Similarly, I dislike the fact that Sun advocates sent hate mail to upside. But, just as most Linux folks do not want to be seen as fanatics just because some loonies like the OS, all of Sun should not be demonized because some wackos called an upside journalist an "idiot."
...
Re-reading your post, I think this answers your question. What exactly was your criticism?
"So, what do you want to hack for, Bickle?"
"I can't sleep nights."
"They got porno theaters for that."
"So, what do you want to hack for, Bickle?"
"I can't sleep nights."
"They got porno theaters for that."
"So, what do you want to hack for, Bickle?"
"I can't sleep nights."
"They got porno theaters for that."
It's easy to inadvertently be an asshole; myopia is not your friend. Remember when WRS died, and those posts showed up saying "Stevens didn't like Linux; I jizz on his corpse"? That poster(s?) was so short-sighted, so focused on advocacy that s/he stooped to the level of...well, the level s/he stooped to.
The philosophy "you dish it out, you've got to take it" has not done much good for the world. You're too clueful a person to follow it.
They seemed to be flinging those "check whether the code will cause error or exception"s around pretty liberally. I wonder how deep that will go (and how much computation time it would take!); and wouldn't sandboxing, not "code checking", really be what's necessary (if one wants security) when swapping in new processor states?
True, the specific name of the person was changed, but, amusing as this is, it's still inappropriate. The job of a support staff member is to help users, not judge them.
See SAGE and Computer Professionals for Social Responsibility, and the ACM for more on this subject.
> In order to get the program into shape, I email
> it to a friend to test and report back with a bug
> list. I admonish him not to release it until I
> finish it. Am I breaking the law?
If it is already under the GPL, and you try to add restrictions onto that previously-GPLed code, then yes, I speculate that you would be breaking the law. But 1) the code isn't necessarily GPLed yet; and 2) you're not changing the licensing restrictions, you're just asking your friend to do you a favor and not redistribute yet. Your friend is (I assume) under no legal compulsion to not redistribute the code, and (I assume) he could legally do so if he really wanted to (even though doing so would be evil).
I agree with your post, except for your comments on virus prevention. Virus scanners are a fundamentally incomplete solution. Choice of an operating system does matter, but not for obscurity reasons. OpenBSD is secure because it has been designed (and verified) to be that way; it is not secure because it is "unusual". And if one has a secure platform, one should be able to run untrusted binaries (provided they are not run by a privileged user such as root).
Obviously you don't want anything that's actually turing-complete
True, we'd probably want an upper limit on the time and space resources of the programs. (Traipsing offtopic: Given that it only has finite space resources, is my desktop machine really turing-complete?) There would be other practical concerns too, like the granularity with which programs could query the database of slashdot articles. (Raw text? Keywords?)
However, without a certain amount of problem-specific hacking, people's filter scripts won't have any useful data to work with. Without such primitives as "moderated up", "moderated down", and "adjective:funny", what will your script look for? "Show me all the posts from people that have never replied to posts with the words 'hemos sux' in them"?
A sufficently general scripting language would have the immediate benefit of dramatically simplifying the current preferences page. That page largely has to do with filtering and customization. (Strip sigs, block Jon Katz, block posts with "hemos sux", etc.)
I agree that we'd need a set of primitives -- indeed, a really large set of primitives would probably be a good idea. But much of the debate about moderation comes from things like karma, sorting, granularity, etc. -- not the primitives themselves. These issues would be resolved with a general enough scripting system.
(Maybe it's the crack talking, but such a scripting language might be able to put some moderation administration into user-space as well. Those users who think there should be more points circulating around could voluntarily assign each other "unofficial" moderator points in an alternative [complementary] system.)
The only real solution to all of this is a turing-complete filtering/sorting/query language. That's what all of this is expanding into. And writing the most general solution once can be easier than coding up a bunch of idiosyncratic, incomplete solutions. Code a solution to the problem domain, Malda, and solving a specific problem will be (relatively) easy.
The last time I read an explanation of this that was posted by Malda, it said the number of unused moderations points is always constant. Once a moderator spends a point, a new point is given to another moderator. So in theory, no points are ever "wasted" (except from that moderator's perspective -- J. Random Moderator probably wants to save up h{is|er} points for exceptional posts). If all moderators were to spend their points fast and furiously, in theory more points would end up being used. (But then, I could be totally misunderstanding things. Perhaps in the year 2525 when Rob releases the source code again, we'll find out for sure.)
Because when you moderate, you're supposed to be regulating the quality, not the content, of the discussion. If you have issues about content (which is why you would post), it's less likely that you will be an impartial judge of quality.
This is interesting -- it sounds like MetaModeration, in practice, will only give negative feedback. MetaModerators exist to keep evil moderators in line; thus mm's will likely spend their time searching for badly-moderated posts. I'd be surprised if any moderators end up with positive moderation karma. (And BTW, Rob does mean that there will be a seperate "moderation" karma, right? If MetaModeration affects regular karma, I think a lot of people will opt out of moderation.)
In a way, moderators regulate other moderators already. If a post is marked down unfairly, other moderators will come along and mark it back up. (This assumes that there are an adequate number of moderator points floating around, which IMHO is not currently the case.) This is already a form of "meta-moderation".
1) If I just click "post anonymously" instead of logging out, and my post gets moderated down, does my karma go down? That's the whole point of posting anonymously -- to be able to say potentially contreversial things without reprimand (beyond flames).
2) To this end, could the "no score +1 bonus" allow one to decrease one's score to 0 or even -1? Sometimes posts are offtopic but still valid, and so should be posted but pre-moderated down.
3) (This is a little offtopic, but at least regards improving slashdot.) Remember how Alan Cox said that (paraphrased) "the problem with slashdot is that everyone tries to get 'First Post' instead of 'First Patch'"? Well, why not post bug tracking info on slashdot? Registered users could filter it out. All Rob would have to do would be to give story-posting privs to list maintainers. Then we'd start getting the slashdot effect on bugs.
MS will probably want to cash in on the Linux hype, and want to put out some competition for it in the OS market. But it would make a lot more sense for them to take a BSD and proprietize it, rather than going with a GPLed kernel. (The re-licensing freedom of the BSD is probably a reason why Apple chose it for Darwin; Woz's ties to Berkeley might be another.) They'd then either need to rewrite any BSD components under the GPL (I can't recall which components of, say, FreeBSD those are) or slap a Win98/NT/etc layer on top of BSD and run user stuff there (leaving admin tasks in the BSD layer). The resulting "Winux" would have a UNIX administrative interface (which would make a lot of [not all] sysadmins happy), the MS brand name and possibly the Win98/NT/etc interface (which would make a lot of [not all] PHBs and technophobe users happy) and would be proprietary (which would make MS happy all over).
DISCLAIMER: This is not a comment on the relative technical merits of Linux, the BSDs, or the Windowses. I'm just saying that FROM MS'S PERSPECTIVE, repackaging a BSD would be a more likely choice from the licensing angle.
Foogle> This new rule hadn't been implemented then
You've misidentified the new rule. For several months, the rule was that a poster with karma of -10 or below is given a default score of zero. The new rule (which leaves the old rule intact) is that a poster with karma of -20 or below is given a default posting score of -1.
If the US powers-that-be really wanted to increase voter turnout, they'd eliminate the prerequisite of voter registration. Registering is annoying and decreases the likelihood that young people will vote. The actual process of voting isn't all that bad -- just find the building and do it. Making registration easier (by eliminating it, or allowing voters to register at the voting site on the day of the vote) is the key.
It appears from a post elsewhere in this discussion that Ellis-D is the other default -1 poster. That sucks: I've seen good posts from Ellis-D before. And DaveO got a lot of -1's when he first came on /., mainly because he posted in an abrasive, Meept-like way (all caps, third person).
But everyone, at some time, has posted stuff that's badly written. DaveO has at least caught on that all-caps posts are bad, and one of these days he'll realize the same about the third-person voice.
I agree with Malda that something ought to be changed to prevent a repeat of the "I jizz on WRS's corpse" posts. However, scapegoating DaveO and Ellis-D will solve nothing.
1) So say someone on my ISP posts a bunch of trolls and logs off, and that when I come home from work and go online, I get assigned that IP. Will I be blocked from posting? (Rob's essay said that if your account *or* IP had a sufficently bad record, you'd be blocked. Why isn't that an "and"?)
2) I've marvelled that there's no ad-spam on slashdot. Do you guys have a team of grunts that deletes those, or do the folks who flood USENET with "Pamela suck off Tommy Lee" posts just not peruse slashdot? If there is a troupe of squirells to delete ads, maybe they can be used to delete the really bad trolls.
3) What were the offending posts that inspired this?
> That has to be the most oddball essay on a
> programming language I have seen recently.
You must not be a member of ACM's SIGPLAN. The weirdest programming language essay I've read was in Volume 33, number 12 of SIGPLAN Notices. An essay therein, "Evolution of the High Level Programming Languages", by Masud Ahmad Malik of Amman, Jordan, derided Ada as being bad for mission critical situations (precisely the domain for which Ada is well-suited). The essay was revealed in Volume 34, number 2 of Notices to be not just innacurate and fourteen years out-of-date, but also plagiarized! It prompted (relatively) angry letters from Ada advocates and an apology by the editors. *That* was a fucked-up comparative programming language essay.
Bingo -- internal fragmentation. Thanks; "granularity" was the only thing that came to mind.
Subnetting under IPv4 is a good idea not fully realized. While it is useful in its present form, it is flawed in that it allows the IP address space to be divided up wastefully. Use of a superset of the current subnetting functions will be needed to remedy address waste.
At least, I think that's what he's trying to say.
(Hey, isn't there an term used in OS-theory circles regarding overly-general division of resources? I recall it from the memory management chapter...can't recall the term....)
> It would make sense that Antimatter Galaxies
> may exist, simply because we have found
> antimatter.
If I understand the article correctly (IANAP),
antiprotons are sometimes found as the result
of (matter) particle collisions. They're looking
specifically for anti-helium, which is unlikely
to come about due to the random collision of
matter.