Window managers are extremely customizable, and by comparison to other operating systems, are even easy to create. There's no need for a tool bar that runs along the bottom of the screen, and no limit to icon sizes. Further, while the standard windowing GUI paradigm may not be appropriate, doesn't mean that SOME Windowing paradigm shouldn't be applied.
Lumping X with Mac and Windows is a little unfair, considering it's flexibility and their lack of it. It's like saying that the standard method of programming used by RISC assembly, CISC assembly, and perl are not usable for writing webpages.
That bit about needing corporations and computers being unhackable is a bit untrue.
There is a nearly infinite collection of problems solvable by computers.
There are a finite number of solutions.
Therefore, pushing the envelope means solving new problems. If you don't have any, you're not looking.
If you think you can't solve them, then you're using the wrong operating system.
Most of todays hardware is very willing to tell the OS what it is doing. All you need is an OS that will tell you what the hardware said.
I've got a few questions I'm interested in at the moment, off the top of my head:
1) Computer-based answering machine: how do you do it cheap and effectively? 2) How do you get variable size packet-based cd writing to work? 3) How can you keep track of all possible system events in userland? 4) What's the best way to do network audio?
In truth, there are more like thirty that I'm considering, but these few came to mind right now. It's not like I actually keep a list, after all.
I always tended to think that web was short for "world wide web," which is often defined as the internet.
So...then they'd be synonymous.
I would say that the internet is more than websites, but equivalent to the web. Obviously this terminology is still being solidified.
At one time I was absolutely correct - especially when the older protocols were used more frequently. Perhaps this has changed because the term "website" has been used so frequently to mean "collection of information accessable through http," and website is the only thing with the word "web" in it.
Google shows some dissent among definitions, despite all talking about the same thing: http://www.google.com/search?q=define:Worl d+Wide+W eb
Most silicon dioxide goes by a more common name - sand.
Glass, also, is mostly silicon dioxide, but they dope it with other stuff so that it won't crack to pieces and turn back into sand under environmental conditions.
But will it work in way-shapes, or in forms, like the previous poster asked?
I think the best way thing we could get as far as sleeping goes would be distributed sleep. If I could rest my arms while I was running, my legs while typing, and only do REM during meetings (no thinking involved there), I'd be set.
Or better yet, I'd like to not need it at all without going crazy. Then I could use that time to actually direct my WAKING thoughts.
Sounds like you were an English prof or something like that. I taught a class where none of your solutions would work.
I was TAing a programming course where the assignments changed every year as much as possible, and because of the demands of the course, there were assignments every week, each TA had 90 students, and there were 1500 people in the class.
I caught people handing in the same assignment in the same class with only a few changes, and sometimes not even a few changes. Still, this takes time, especially when we're supposed to do all the work in a mere 6 hours of grading per week (it was a half-time appointment, and the other 14 hours of it are already accounted for).
I'm sure a few got through. It's not like I can exactly remember every paper in its entirity to know who has turned in something similar; a cheat detector would do much better. But I can't possibly see which papers were copied between two different classes.
Of course, you could minimize cheating some by giving each TA's classes the same assignment, but then each TA will have to hold their own office hours each week, since they would be the only ones who would know their specific assignment (instead of sharing hours so that each only has to come in once a month). Remember, there's only 6 hours for grading now. It wouldn't be good to cut that back.
You could also reduce the amount of grading in the course so that it could be graded by an individual. If you don't give homework on it, you might as well not teach it (with programming stuff, where practice is everything), so this is a material-in-the-course question. With big courses like this, however, curriculum decisions are not made by teachers. They're made by departments - usually several of them. They all want material that's important to them in the course, and the result is a lot of material. The politics make this point difficult to do.
So we're left with the only solution that could really work: a cheat detector.
Well, you do have to pay for the driver development kit. It's $200 for an MSDN subscription, which is required to download it. And just because it's documented doesn't mean it's easy enough to actually use. Neither Windows kernel was programmed as modularly as the Linux kernel was. It's hacks on top of hacks, and what you write may not work. Besides, you'd have to reverse engineer the hardware most of the time to do it, so it'll take a long time, just as linux has taken with, for instance, Winmodems.
You don't seem to get what I meant by priority. Locations (customers) that have downloaded a lot would have a lower priority than locations that have not.
The high-bandwidth users would see temporary dips in their usage as low-usage customers did stuff, but if they were constantly using bandwidth (such, as for instance people with file servers), the overall usage would still most likely remain high, but with no harm done.
Note that this MAY limit bandwidth available to people based upon usage, but it might not, and the added bonus is that you get more satisified customers than you would have had (and this is a big deal because I know that most techies are high-bandwidth users whose opinion is highly valued to their less tech-savvy friends, so this is a good way to get good word-of-mouth advertising).
So the biggest bandwidth users always get the smallest allotment of bandwidth, and therefore generate lots of extra network traffic with ping requests?
Doesn't sound like a good idea to me. Wouldn't it be a better idea to prioritize packets based on location, rather than just cutting the bandwidth back?
That way, when no one else is using the huge connection, the bandwidth hogs can have it, but as soon as someone else needs it, they get it.
We could have an authority that you pick a username and password for, and a list of e-mail addresses, and then allow you to make records with three data items: 1) Key itself 2) Company 3) The e-mail address used
If there is only one such authority, and each e-mail address can only be registered once, then spammers would be forced to illegal action. Companies wouldn't be allowed to sell e-mail addresses, because only they would have the right to use them, NOT whoever they would sell them to.
Of course, spammers could register and then opt in other people's addresses, but that would obviously be equally illegal and actually easy to prove.
Actually, they're not at the moment in most places.
But the suggestion is that they're not marriages - between two gays, it's not God-sanctioned union between a man and a woman, which has pretty much been the definition of it.
It's only a state (i.e. civic) sanctioned union. Civic union sounds like a pretty good description of it to me.
My experience with poetry has led me to believe that the purpose of poetry is to muddle things that one understands by explaining them a different way in order to see them in a new light.
Coding, on the other hand, appears to be trying to express complicated things in a simplified (universally structured and explained using a small set of concepts) way in order to have them understood even by automatons.
Clarity is sought in coding while confusion is sought with poetry. I would almost say that coding and poetry are opposites.
Re:How will H usage affect this?
on
Global Dimming
·
· Score: 4, Insightful
You're right about the storage mechanism thing, but wrong about the emissions thing.
Automobiles are one of the more dirty ways of converting fossil fuel energy into usable energy, specifically because really good filters, and very high temperature combustion are not desirable (for both portability and usefulness reasons).
However, if this is done at a plant, these issues go away. The burning process will be much cleaner.
Yes. When looking at an infringing piece of code that has been claimed by SCO, and which says at the top "written by John Smith," they'll have to ask Linus who John Smith is and where he lives, and then call him up for a statement instead of reading it on a weblog.
Are blogs admissible in court? I'm not sure this actually changes anything.
1,000,000 rewrite cycles? I think they're lying, or being very gentle during the writes. That sounds more like the number of writes available to hard drives (no they don't have infinite rewritability).
You might get 500,000 writes of all zeros followed by 500,000 writes of all ones, but I doubt you'll go over 100,000 intermittently switching between them (and probably not over 10,000) without a few of those bits getting stuck. There's just not a standard that says what a write is.
A gaussian distributed changing of bits? A complete toggle of all bits? A change of a fixed percentage and random location of bits? Also, they don't really say how it stands up to heat, which shortens the lifespan of flash chips and is an issue in many places.
Flash chips haven't changed much materials-wise since they came out, so why would they get more writes all of a sudden?
While I agree that this might be a good learning step, and a good thing to do with one's life (in fact, this is how I started out my career as a Computer Scientist), your argument is a logical fallacy known as "tu quoque," one of the many red herring fallacies.
You have not shown that the previous poster has not right to criticize. I would postulate that, this being a free web forum, he does have this right, barring any good argument against it.
Um...there's already a C toolchain for it. There was before this (for almost as long as there has been TI92, which was out before 89 and uses the same processor) because there's one for M68K. All they had to do was add a few macros that added stuff so that it was up to par with the file specification for TI89/92.
Yes. And we should blame knife makers because they made a weapon that was used in a brutal stabbing, or a gun-maker maker for a brutal shooting, and the DMCA is a perfectly justified law.
Of course that's proposterous. The tool is not the crime. Sociopathy and lack of social responsibility knows no limits or bounds, and self-justification for such behavior is limited only by the imagination. Little old ladies who go to church and feed the homeless can have areas of social irresponsibility as well.
I know that one of my grandmothers, who is one of those little old ladies who goes to church and feeds the homeless, just happens to be racist. Does that make racism justifiable?
Jewish literature isn't the only place to have golems, just one of the few to call them that. It was a word they used because they didn't have robots, but basically they're the same thing.
The idea of the genre (folk tales the world over have this genre) 1) Person/people are discontent for some reason 2) Person/people create an automaton, or one is created for them. 3) Person/people do not take responsibility for created automaton in some way, and it gets out of control (and a mind of it's own), and bad things happen/almost happen. 4) The moral of the story is that people have to take care of their problems themselves because that involves thought, and it can't be done by automatons; they create problems. Alternatively, the moral is that creation of life is a taboo, meant only for God.
I've read a couple of South American stories on the subject, and then there's the golem one, and of course, the most well known: Frankenstien.
I'm not sure that this applies here. Frodo pities gollum and accepts him because he has faith that one so fallen can return to a state of goodness. If gollum can't can Frodo, whose thoughts verge into gollum-logic themselves? It is, I think, Tolkien's attempt to show the concept of forgiveness (Aren't we but a few steps away from being just like those we could forgive? How are we then worthy to judge?).
We called them "teachers." They were given some subroutines for face recognition during the first few years of their construction in order to recognize individual students and reject those who didn't actually go to our school. Apparently these had some other function as well, usually, but I forget what it was. Something about information transfer, I believe.
The advanced model of these, "administrators" also had some programming for student retrieval (of outlier students with difficient programming, leading them to go to well-traveled entertainment locations rather than going to the school). Administrators were also programmed for information retrieval, augmenting their face-recognition and reasoning skills - allowing them to run intrusion-detection hiring subroutines with heuristics designed to limit the presence of malicious entities at the school.
Is this a new model of administrator? How does it stack up to previous versions?
Look, he doesn't mean that you should randomly go out and search google. If you're a newbie and don't even know the question, you can read the guides.
If you have a straightforward problem that has been solved MANY, MANY times, like how to set up a mail,web, or ftp server, you can check the howtos.
In my experience (and I do the IRC thing a lot), you can't expect to have esoteric/advanced questions answered on IRC. If people don't know the answer immediately, they don't want to put in the work to figuring it out. What they can help you with is understanding the things in TLDP if you don't - like perhaps telling you what question you should be asking, or where you should be reading.
Okay. I'll spend a six or so months learning the UDF specification and how the drives work. Then I'll spend a year or two (probably two) learning how the kernel works (because it would be necessary to understand most of the kernel for this process), and another year making the largescale modification that would be necessary to solve this problem. Then I'll spend two more years supporting the project so that it'll make it into the kernel.
At this point, I've taken about as long as it will take for the people who already know how to do it to decide that making the change is a worthwhile endeavor, even from this point.
This comes to a fundamental nature of monolithic development: it takes much, much longer for anyone who didn't write it to write things for it than it does for it's developers to do something on it.
Slashdot Mirror
Not a Linux user, I take it?
Window managers are extremely customizable, and by comparison to other operating systems, are even easy to create. There's no need for a tool bar that runs along the bottom of the screen, and no limit to icon sizes. Further, while the standard windowing GUI paradigm may not be appropriate, doesn't mean that SOME Windowing paradigm shouldn't be applied.
Lumping X with Mac and Windows is a little unfair, considering it's flexibility and their lack of it. It's like saying that the standard method of programming used by RISC assembly, CISC assembly, and perl are not usable for writing webpages.
That bit about needing corporations and computers being unhackable is a bit untrue.
There is a nearly infinite collection of problems solvable by computers.
There are a finite number of solutions.
Therefore, pushing the envelope means solving new problems. If you don't have any, you're not looking.
If you think you can't solve them, then you're using the wrong operating system.
Most of todays hardware is very willing to tell the OS what it is doing. All you need is an OS that will tell you what the hardware said.
I've got a few questions I'm interested in at the moment, off the top of my head:
1) Computer-based answering machine: how do you do it cheap and effectively?
2) How do you get variable size packet-based cd writing to work?
3) How can you keep track of all possible system events in userland?
4) What's the best way to do network audio?
In truth, there are more like thirty that I'm considering, but these few came to mind right now. It's not like I actually keep a list, after all.
I always tended to think that web was short for "world wide web," which is often defined as the internet.
l d+Wide+W eb
So...then they'd be synonymous.
I would say that the internet is more than websites, but equivalent to the web. Obviously this terminology is still being solidified.
At one time I was absolutely correct - especially when the older protocols were used more frequently. Perhaps this has changed because the term "website" has been used so frequently to mean "collection of information accessable through http," and website is the only thing with the word "web" in it.
Google shows some dissent among definitions, despite all talking about the same thing:
http://www.google.com/search?q=define:Wor
Most silicon dioxide goes by a more common name - sand.
Glass, also, is mostly silicon dioxide, but they dope it with other stuff so that it won't crack to pieces and turn back into sand under environmental conditions.
But will it work in way-shapes, or in forms, like the previous poster asked?
I think the best way thing we could get as far as sleeping goes would be distributed sleep. If I could rest my arms while I was running, my legs while typing, and only do REM during meetings (no thinking involved there), I'd be set.
Or better yet, I'd like to not need it at all without going crazy. Then I could use that time to actually direct my WAKING thoughts.
Sounds like you were an English prof or something like that. I taught a class where none of your solutions would work.
I was TAing a programming course where the assignments changed every year as much as possible, and because of the demands of the course, there were assignments every week, each TA had 90 students, and there were 1500 people in the class.
I caught people handing in the same assignment in the same class with only a few changes, and sometimes not even a few changes. Still, this takes time, especially when we're supposed to do all the work in a mere 6 hours of grading per week (it was a half-time appointment, and the other 14 hours of it are already accounted for).
I'm sure a few got through. It's not like I can exactly remember every paper in its entirity to know who has turned in something similar; a cheat detector would do much better. But I can't possibly see which papers were copied between two different classes.
Of course, you could minimize cheating some by giving each TA's classes the same assignment, but then each TA will have to hold their own office hours each week, since they would be the only ones who would know their specific assignment (instead of sharing hours so that each only has to come in once a month). Remember, there's only 6 hours for grading now. It wouldn't be good to cut that back.
You could also reduce the amount of grading in the course so that it could be graded by an individual. If you don't give homework on it, you might as well not teach it (with programming stuff, where practice is everything), so this is a material-in-the-course question. With big courses like this, however, curriculum decisions are not made by teachers. They're made by departments - usually several of them. They all want material that's important to them in the course, and the result is a lot of material. The politics make this point difficult to do.
So we're left with the only solution that could really work: a cheat detector.
Well, you do have to pay for the driver development kit. It's $200 for an MSDN subscription, which is required to download it.
And just because it's documented doesn't mean it's easy enough to actually use. Neither Windows kernel was programmed as modularly as the Linux kernel was. It's hacks on top of hacks, and what you write may not work. Besides, you'd have to reverse engineer the hardware most of the time to do it, so it'll take a long time, just as linux has taken with, for instance, Winmodems.
You don't seem to get what I meant by priority.
Locations (customers) that have downloaded a lot would have a lower priority than locations that have not.
The high-bandwidth users would see temporary dips in their usage as low-usage customers did stuff, but if they were constantly using bandwidth (such, as for instance people with file servers), the overall usage would still most likely remain high, but with no harm done.
Note that this MAY limit bandwidth available to people based upon usage, but it might not, and the added bonus is that you get more satisified customers than you would have had (and this is a big deal because I know that most techies are high-bandwidth users whose opinion is highly valued to their less tech-savvy friends, so this is a good way to get good word-of-mouth advertising).
So the biggest bandwidth users always get the smallest allotment of bandwidth, and therefore generate lots of extra network traffic with ping requests?
Doesn't sound like a good idea to me. Wouldn't it be a better idea to prioritize packets based on location, rather than just cutting the bandwidth back?
That way, when no one else is using the huge connection, the bandwidth hogs can have it, but as soon as someone else needs it, they get it.
That's why I'm using it.
The fastest way to run Windows programs under Linux is with Win4lin.
Win4lin doesn't support W2K, so I use Win98 for it.
I'll have to agree with the other poster on this. Your router is dying on ten users?
That should be a piece of cake if they're all browsing.
Or we could make opt-in harder.
We could have an authority that you pick a username and password for, and a list of e-mail addresses, and then allow you to make records with three data items:
1) Key itself
2) Company
3) The e-mail address used
If there is only one such authority, and each e-mail address can only be registered once, then spammers would be forced to illegal action. Companies wouldn't be allowed to sell e-mail addresses, because only they would have the right to use them, NOT whoever they would sell them to.
Of course, spammers could register and then opt in other people's addresses, but that would obviously be equally illegal and actually easy to prove.
Actually, they're not at the moment in most places.
But the suggestion is that they're not marriages - between two gays, it's not God-sanctioned union between a man and a woman, which has pretty much been the definition of it.
It's only a state (i.e. civic) sanctioned union. Civic union sounds like a pretty good description of it to me.
My experience with poetry has led me to believe that the purpose of poetry is to muddle things that one understands by explaining them a different way in order to see them in a new light.
Coding, on the other hand, appears to be trying to express complicated things in a simplified (universally structured and explained using a small set of concepts) way in order to have them understood even by automatons.
Clarity is sought in coding while confusion is sought with poetry. I would almost say that coding and poetry are opposites.
You're right about the storage mechanism thing, but wrong about the emissions thing.
Automobiles are one of the more dirty ways of converting fossil fuel energy into usable energy, specifically because really good filters, and very high temperature combustion are not desirable (for both portability and usefulness reasons).
However, if this is done at a plant, these issues go away. The burning process will be much cleaner.
Yes. When looking at an infringing piece of code that has been claimed by SCO, and which says at the top "written by John Smith," they'll have to ask Linus who John Smith is and where he lives, and then call him up for a statement instead of reading it on a weblog.
Are blogs admissible in court? I'm not sure this actually changes anything.
1,000,000 rewrite cycles? I think they're lying, or being very gentle during the writes. That sounds more like the number of writes available to hard drives (no they don't have infinite rewritability).
You might get 500,000 writes of all zeros followed by 500,000 writes of all ones, but I doubt you'll go over 100,000 intermittently switching between them (and probably not over 10,000) without a few of those bits getting stuck. There's just not a standard that says what a write is.
A gaussian distributed changing of bits? A complete toggle of all bits? A change of a fixed percentage and random location of bits? Also, they don't really say how it stands up to heat, which shortens the lifespan of flash chips and is an issue in many places.
Flash chips haven't changed much materials-wise since they came out, so why would they get more writes all of a sudden?
While I agree that this might be a good learning step, and a good thing to do with one's life (in fact, this is how I started out my career as a Computer Scientist), your argument is a logical fallacy known as "tu quoque," one of the many red herring fallacies.
You have not shown that the previous poster has not right to criticize. I would postulate that, this being a free web forum, he does have this right, barring any good argument against it.
Um...there's already a C toolchain for it. There was before this (for almost as long as there has been TI92, which was out before 89 and uses the same processor) because there's one for M68K. All they had to do was add a few macros that added stuff so that it was up to par with the file specification for TI89/92.
Yes. And we should blame knife makers because they made a weapon that was used in a brutal stabbing, or a gun-maker maker for a brutal shooting, and the DMCA is a perfectly justified law.
Of course that's proposterous. The tool is not the crime. Sociopathy and lack of social responsibility knows no limits or bounds, and self-justification for such behavior is limited only by the imagination. Little old ladies who go to church and feed the homeless can have areas of social irresponsibility as well.
I know that one of my grandmothers, who is one of those little old ladies who goes to church and feeds the homeless, just happens to be racist. Does that make racism justifiable?
Jewish literature isn't the only place to have golems, just one of the few to call them that. It was a word they used because they didn't have robots, but basically they're the same thing.
The idea of the genre (folk tales the world over have this genre)
1) Person/people are discontent for some reason
2) Person/people create an automaton, or one is created for them.
3) Person/people do not take responsibility for created automaton in some way, and it gets out of control (and a mind of it's own), and bad things happen/almost happen.
4) The moral of the story is that people have to take care of their problems themselves because that involves thought, and it can't be done by automatons; they create problems. Alternatively, the moral is that creation of life is a taboo, meant only for God.
I've read a couple of South American stories on the subject, and then there's the golem one, and of course, the most well known: Frankenstien.
I'm not sure that this applies here. Frodo pities gollum and accepts him because he has faith that one so fallen can return to a state of goodness. If gollum can't can Frodo, whose thoughts verge into gollum-logic themselves? It is, I think, Tolkien's attempt to show the concept of forgiveness (Aren't we but a few steps away from being just like those we could forgive? How are we then worthy to judge?).
We called them "teachers." They were given some subroutines for face recognition during the first few years of their construction in order to recognize individual students and reject those who didn't actually go to our school. Apparently these had some other function as well, usually, but I forget what it was. Something about information transfer, I believe.
The advanced model of these, "administrators" also had some programming for student retrieval (of outlier students with difficient programming, leading them to go to well-traveled entertainment locations rather than going to the school). Administrators were also programmed for information retrieval, augmenting their face-recognition and reasoning skills - allowing them to run intrusion-detection hiring subroutines with heuristics designed to limit the presence of malicious entities at the school.
Is this a new model of administrator? How does it stack up to previous versions?
But those are books - the guides are.
They're just online.
You can of course, just print them out.
Look, he doesn't mean that you should randomly go out and search google. If you're a newbie and don't even know the question, you can read the guides.
If you have a straightforward problem that has been solved MANY, MANY times, like how to set up a mail,web, or ftp server, you can check the howtos.
In my experience (and I do the IRC thing a lot), you can't expect to have esoteric/advanced questions answered on IRC. If people don't know the answer immediately, they don't want to put in the work to figuring it out. What they can help you with is understanding the things in TLDP if you don't - like perhaps telling you what question you should be asking, or where you should be reading.
Okay. I'll spend a six or so months learning the UDF specification and how the drives work. Then I'll spend a year or two (probably two) learning how the kernel works (because it would be necessary to understand most of the kernel for this process), and another year making the largescale modification that would be necessary to solve this problem. Then I'll spend two more years supporting the project so that it'll make it into the kernel.
At this point, I've taken about as long as it will take for the people who already know how to do it to decide that making the change is a worthwhile endeavor, even from this point.
This comes to a fundamental nature of monolithic development: it takes much, much longer for anyone who didn't write it to write things for it than it does for it's developers to do something on it.