What pretty sweet part are we talking about? 2 HDMI ports I can't use without a 2,000$ new TV? 7 player bluetooth, when I rarely (never) have a situation where I go, "damn, I wish my GameCube had 8 ports so everyone could play Mario Party instead of just 4 at a time"? The part where the PS3 is also an Internet router, instead of my current one, with 3 gigabit ports?
Every year, Sony Fan Bois have been going, "OMG CREAM" about it. There is no such thing as a sweet Sony console; it's always a pack of lies. The only reason the PS1 got popular was because everyone hated pompous Nintendo and their "screw everyone" mentality in the mid-1990s. How many 1st party titles (and I don't mean 2nd party, like Polyphony Digital) have there been on the PS1 and PS2 that have been super awesome? How does that number compare to the titles by 3rd party developers?
Sony sells far less PS1s and PS2s than Square Enix, Namco, Capcom, or Konami -- even alone. Hell, the only thing the semi-rational fan bois seem to want on the PS3 is Metal Gear 4 -- a Konami title.
"This is getting out of hand, and you can apologize for your (or my) government's intrusive behavior all you want, but the truth is that everyone will, sometime, somewhere, do something he'd rather other people didn't see. "
Why would someone do something they didn't want other people to see? Well, it could be because they were arranging something (surprise birthday party), or it could be because they were doing something illegal. Now, you can't legistlate right and wrong, but you can draw a legal boundary around certain actions.
So, taking this a bit further, what's so bad about using a system to monitor legal or not actions? If there are extenuating circumstances (a good reason for doing something that bends/breaks the rules), isn't that something you present in a court of law? Innocent until proven guilty?
Within the privacy of my own home, I have complete domain. Outside, I have no domain -- public areas are for everyone. Don't do things there you don't want to possibly have other people know about.
Unlike TCP, where the end-points do all the thinking, multicast requires that the routers are involved in the transactions. They are the ones who have to make decisions like, "does this address get bits, or not?"
The session management protocols of multicast are defined, but there are a few to choose some, and most have some kind of serious drawback associated with them. One of the ones that sticks out in my mind is the one where there's no way to "detect" if a multicast IP is taken, or any more security/authentication than knowing what the address is.
To properly support multicast, we need a session leader, and every router involved in the minimum-cost spanning tree must also know who else is involved. This means the routers have to be able to build the tree, and tear it down as clients join and leave.
Replacing or upgrading routers is hard because a lot of them are fire and forget. They'll place a router in a wall with PoE, and then leave it inside. They'll be on the bottom of the ocean, repeating traffic that goes along a trans-oceanic link. They'll be on top of wireless towers, miles from other people. Most of them were not designed to be remotely upgradable via software, because routers were always meant to be as cheap to produce as possible.
This is also who IPv6 is only really deployed in places where IP space ran out a long time ago (such as Japan). Until it really starts to break, traditional structure will be "good enough" for most people.
Capitalism is a funny system because of a few basic assumptions it needs to function well. One of those assumptions is that users will know things, like what is a better product to buy. Because of this, people who sell shoddy equipment on unreliable gear will not succeed.
Now let's apply this. I have a PowerBook that is very reliable. I also have a desktop that's very reliable (in fact, 3). However, these desktops are component-based machines; they run Linux. How is it that these component-based macchines are as reliable as my end-to-end model PowerBook? I bought components which aren't garbage. AMD CPUs, Kingston lifetime warranty RAM, Enermax power supplies, etc. It's more expensive than what most people probably buy, but I've never had a peap of trouble. I know what components to buy because I take the time to look into it, and because I only buy components that the Linux kernel supports (which, for some reason, happen to me more reliable than random Taiwanese garbage).
With Apple's model, we skip this step. Apple themselves takes the time to try and get quality components that work reliable with OS X. Since they vend the machine and the hardware, they can't hide behind the "Windows sucks" excuse the way cheaper component suppliers can. However, and this is important to note, they're still interested in shrinking costs as much as possible to maintain their fat margins, and they still like to charge a high markup. Plus, they're not immune from mistakes (note the GOBS of heatsink goo on the heatpipes of the 15" MacBook Pros). This means they don't always do as good a job as someone who knows what I do.
Really, it's just moving the burden of choosing chocolates from shit from the consumer up the chain a bit, but even then it's not perfect. If you want thinks done right, do it yourself -- learn about PC construction, or pay someone you trust (be it Apple or your friend). If you just go buy the cheapest thing you can, you're on a roller-coaster ride to the bottom in terms of quality and consistency -- that's why Wal-mart's stuff is different (they have different product badged the same to cut costs), and also why Wal-mart is not always the best place to shop.
Adam Smith's invisible hand requires you to do research!
These all affect each other. The binaries are different between different arches, duh! The fact is that you invalidate fewer cache lines with smaller x86 instructions, and those Intel-based Macs are way faster than the PowerPC ones.
World of Warcraft on my friend's 15" MacBook Pro blows the doors off my 12" Powerbook G4 of the previous generation. We're talking 20fps average vs. 50!
You start off by saying that what I said was not math, but algebra.
For about most of of humans on this planet, that's all math is. We're very much in the minority. To "normal" people, math and computer programming are very different, much like physics is very different or logic is very different (even though all these things only differ in degree).
I also said that logic, since it expands and describes more in a rational system than pure maths (although the definition of mathematics is also expanding to include many things done in logic), I'd say that logic in all its forms (including programming systems) are a superset of mathematics. You can do more in a formal logical system than you can do in a pure math system -- a Turing machine does more than a push-down automata. "Modern" math is just expanding into areas that were explored by thinkers as far back as Leibniz, who formulated constructs not unlike Turing machines.
What is important is the perspective. Mathematicians are taught in a way different from computer scientists, and thus they see solutions to problems in different ways (there are thousands of mathematician/physics/engineer jokes I can put in here, you get the idea). Computer scientists solve problems of thought, while mathematicians solve problems of relationships. But the thoughts will contain (naturally) thousands of relationships described (implicitly or explicitly) along the way towards the computation of a solution. Computer science is the superset of math the same way my toolbox is the superset of my wrench. Mathematicians tend to see the world as a set of bolts, while computer scientists are free to apply solutions from other logical disciplines.
"Now is your turn, name someone who does a lot of programming and has no mathematical knowledge doing something really significative for computer science. "Web 2.0" reimplementations of 30+ years old concepts are not significative contributions."
There are 2 kinds of people; those that understand things in the larger picture perspective, and can create or discover new ideas. These are people who will, naturally, have some understanding of mathematics. Alternatively, there are those that are like myna birds -- excellent at repeating what they have been told. Much like a pocket calculator is great at adding and dividing, a programmer without math is only good at repeating what is trivial to people who know the bigger picture.
It's impossible for a pocket calculator to discover new things; you ask the impossible. Try to understand why I believe math is a subset, and explain where I have a flaw rather than get distracted with trivia.
"In fact, math is harder, and after doing some higher maths, you will surely be a better coder. Maths expands your mind."
Math is easier than programming. In math, you only express a relationship between things in numerical terms. In the more robust logic systems of a computer, you also have to take into account many more variables (unless, of course, you're doing programming purely on paper).
Math is more formal than computer science thanks to the push towards better proofs in the 19th and 20th centuries; Computer Science is currently around where math was in the 15th and 16th century -- lots of hacks working at it, proofs for sale, "instinctive" and geometrical proofs unbacked up by any kind of algebra. Just because that makes it seem easier doesn't mean it is. I'm sure Dijkstra, an accomplished mathematician, would have harsh words for those that say math is harder.
Math is a subset of programming. People who can't see that the logical expression in terms up pure numerical values is a subset of higher-order logic are like Leopold Kronecker, who opposed set theory as having any relationship to mathematics in the 19th century. Nowadays, most mathematicians agree that set theory and mathematics are correct.
There are proofs that are very hard to do with traditional mathematical techniques that end up being only a few lines when you apply Turing's approaches.
Apparently they're charging roughly 35 to 40$ USD a month, and 40$ CDN a month for their "North America Unlimited" plans.
Currently, VOIP outgoing providers sell time for about a cent per minute in North America, or as high as 1.5-2.0 cents for an incoming number. My roomate and I have a 1-888# which comes in to the house for a few cents a minute, with outgoing numbers going through 3 different VOIP providers of varying cost (cheapest first, of course). I think it costs us a grand total of 8$/month-2 months depending on usage.
Basically, Vonage is making so much money because traditional POTS is a cash cow, and VOIP makes it ridiculously cheap since you don't even need to run fibre -- BYOC!
"You can do simple things easily - and in particular, you can do things where the information only passes in one direction quite easily, but anythign else is much much harder, because there is no "shared state" (by design). And in the absense of shared state, you have a hell of a lot of problems trying to make any decision that spans more than one entity in the system."
I think you're looking at this the wrong way around.
There has been a lot of research into this over the past 40 years, ever since Dijkstra first talked about coordination on a really big scale in the THE operating system. Any decent CS program has a class on distributed programming. Any decent SW architect can break down these different parts of the OS into weakly-connected pieces that communicate via a message passing interface (check out this comment by a guy talking about how Dragonfly BSD does this).
It's obvious that breaking something like your process dispatcher into a set of processes or threads is silly, but that can be easily separated from the core context switcher. Most device driver bottom halves live fine as a userland process (each with a message-passing interface to their top-halves).
If you're compiling for an embedded system, I'm sure you could even entirely remove the interface via some #define magic; only debug designs could actually have things in separate address spaces.
The point I'm trying to make is: yes, you can access these fancy data structures inside the same address space, but you still have to serialize the access, otherwise your kernel could get into a strange state. If you mapped out the state diagram of your kernel, you'd want the transistions to be explicit and synchronized.
Once you introduce the abstraction that does this, how much harder is it to make that work between processes as well as between threads in the kernel? How much of a benefit do you gain by not having random poorly-written chunks pissing over memory?
How about security benefits from state-machine breakdowns being controlled and sectioned off from the rest of the machine? A buffer overflow is just a clever way of breaking a state diagram and adding your own state where you have control over the IP; by being in a separate address space, that poorly written module can't interact with the rest of the system to give elevated privileges for the attacker (unless, of course, they find flaws in more of the state machines and can chain them all together, which is highly unlikely!).
Clearly there is a security benefit as much as there is a consistency benefit. Provably correct systems will always be better.
"Fact: OSX is sooooo slow that the only thing it is faster than is OpenBSD. And you cant even blame its slowness on it being a microkernel. How pathetic... Wow, that says it all in my book:)"
Actually, OS X was within a few percentage points of Linux on all hardware tested; actually outperforming it on memory throughput on PowerPC and some other tests. It's also faster than NT.
"But there are ALWAYS situations where it is going to be desirable for seperate parts of an OS to directly touch the same memory in a cooperative manner, and when this is the case a microkernel just gets in your damn way..."
I fail to see how L4's shared-memory is somehow magically different from Linux's shared-memory. Once you let more than 1 process have access to a page via a TLB mapping, it's all the same.
Slashdot may be news for nerds, but it has a serious drawback when it comes to things such as this. The drawback is that what is accepted as "fact" by most people is never questioned.
"Fact": Micorkernel systems perform poorly due to message passing overhead.
Fact: Mach performs poorly due to message passing overhead. L3, L4, hybridized kernels (NT executive, XNU), K42, etc, do not.
"Fact": Micorkernel systems perform poorly in general.
Fact: OpenBSD (monolithic kernel) performs worse than MacOS X (microkernel) on comparable hardware! Go download lmbench and do some testing of the VFS layer.
Within the size of L1 cache, your speed is determined by how quickly your cache will fill. Within L2, it's how effecient your algorithm is (do you invalidate too many cache lines?) -- smaller sections of kernel code are a win here, as much as good algorithms are a win here. Outside of L2 (anything over 512k on my Athlon64), throughput of common operations is limited by how fast the RAM is -- not IPC throughput. Most microkernel overhead is a constant value -- if your Linux kernel us O(n) or O(1), then it's possible to tune the microkernel to be O(n+k) or O(1+k) for the equivalent operations. The faster your hardware, the smaller this value of k since it's a constant value. L4Linux was 4-5% slower than "pure" Linux in 1997 (See L4Linux site for the PDF of the paper).
But none of this is something the average slashdotter will do. No, I see lots of comments such as "micorkernels suck!" already at +4 and +5. Just because Mach set back microkernel research by about 20 years, doesn't mean that all micorkernels suck.
You can do X as a reverse of a K in that alphabet; U and V were a bit different (V is easier to do right-to-left for the machine to recognize the stroke, but you can make the shape the same as a "real" v). I actually did some of my paper notes in Graffiti (in University) since they tend to be mono-strokes (rather than the polystrokes to make up them more complicated letters when we write out traditional english), making it easier for shorthand note taking (which has a strict stroke form/order dictated by speed and readability!).
There you go: simple strokes, easy to read by eye, easy to read by my 68000-based Palm from 1998 as well as my current StrongARM Palm of 2004!
"Instead of cut and try, cut and try, cut and try, like the hot rod guys do, you have to do a whole bunch of computer analysis before you build it," he said. "We did (computerized) structural analysis and we did stability analysis. And by God, you know what happens? It works! Duh."
I have to agree with him regarding hot-rodders. A lot of people seem to think the way to solve a problem is to frob at it until you get something that works. All the Motorola phone hacking kids, Xbox homebrewers, and PSP kiddies seem to think that the spackle approach (throw things at the wall until something sticks) is the best way to solve problems. You know, rather than solving them by understanding them:)
Straight from the innovations in software page, we have: "As patentability has increased, there's good evidence that the number of software innovations has decreased. Bessen and Maskin also demonstrated a statistical correlation between the spread of patentability in the United States and a decline in innovation in software. In particular, between 1987 and 1994 , software patents issuance rose 195%, yet real company funded R&D fell by 21% in these (software) industries while rising by 25% in industries in general. This paper gives additional evidence that software patents are inversely related to innovation; it's hard to not notice that as patenting become more common (e.g., 1987 and later) that the number of major innovations slowed down and are almost always not patented anyway."
The link supplied is to this PDF about patents. It's worth your time to read about this research.
But when you can make your whole house wired with 1Gbps ethernet in 2006 for half the cost what it cost your home to be wired with 100Mbps ethernet in 2000, isn't it reasonable to expect that ISPs aren't paying the same peering fees they used to?
Isn't it reasonable to expect that the interconnections are faster, and can handle more? Isn't there a near zero cost of adding more users, as long as the pipe isn't oversubscribed?
I think a lot of it is that business models for ISPs are based around a much longer cycle than technology will permit them to have. Anytime a disruptive technology comes along, must we have a debate about how the old, no longer valid assumptions still hold true?
Distributed content replication is one of the key features of the Internet, which is why the MPAA/RIAA hate it.
"It is not the same. The PSP is OFF while the DS remains on. Consuming less power to be sure but still draining the battery. The really big difference? If power fails because of an empty battery the PSP can resume where it left off as if nothing has happened. The DS loses its progress."
Not true. The PSP, if left in sleep mode, will die out. It still periodically refreshes its RAM, even if the cores are turned off entirely. I own(ed) a PSP and left it sleeping. When it was totally dead (would not resume from the button), it did the fully power on Sony logo the next time it was plugged in, and had lost all progress in the game. The PSP does not sleep to flash.
"As for speed of writing and reading. No competition. Nintendo has chosen not to have onboard save. That no doubt saves them money meaning that Nintendo makes a profit on a cheaper console but now the save file has to be in the game catridge and it is not going to be top of the line because of again the cost issue."
What!? I'm not sure what you're trying to say here. It's true the DS doesn't have a save slot in the device, because the carts can have a little flash spot in them.
"AC:WW takes about 30 secs to load and 30 secs to save. Since I love the DS lite for its ease of use in those lost 5 minutes that is a bit too long. The PSP is a bit to large to carry in my pocket but it allows me to play faster and quit faster."
I'm not sure you're quite aware of the time factor required to seek around on those PSP UMDs. They are no where near the speed of ROM or flash. I can fold my DS up instantly and have it sleep for about 4 days of standby time (equivalent to the PSP sleep mode), but startup is way faster. Getting into Brain Age is a matter of tapping the screen a few times. With Lumines, "Now Loading" pops up, and then a whole lot of seeking would happen. Some games have upwards of 45 second load times!
I load a lot more than I save, and it seems to me that the DS is better optimized for this usage.
claiming things like breakage sounds a lot more like the anus defence in terms of "getting off" with the artists' money.
Subverting legal agreements seems to be too acceptable to these types.
Yes, this is one of the drawbacks of North America
on
EA Spouse Outed
·
· Score: 1
There's as fairly informative essay on work patterns I read recently.
"On the other hand, the "market" for free time hardly even exists in America. With few exceptions, employers (the sellers) don't offer the chance to trade off income gains for a shorter work day or the occasional sabbatical. They just pass on income, in the form of annual pay raises or bonuses, or, if granting increased vacation or personal days, usually do so unilaterally. Employees rarely have the chance to exercise an actual choice about how they will spend their productivity dividend. The closest substitute for a "market in leisure" is the travel and other leisure industries that advertise products to occupy, our free time. But this indirect effect has been weak, as consumers crowd increasingly expensive leisure spending into smaller periods of time."
I think it summarizes the weird dichotomy I see here. Some yanks just can't get enough of that 9 to 5, like it's their way of buying a stairway to heaven or something. I'd rather enjoy my time off, but if I were to live in the US, I'd have no real way of enforcing this since employers discourage less pay for more time off.
"We don't want to be like the Europeans. Generally speaking, the US produces far more and creates much more wealth than most European countries."
Since when was my main goal in life declared to be wealth creation, instead of something a bit more hedonistic and less puritan?
I mean, if you want no vacation time a year, go ahead. I'd much rather have a couple months of paid vacation to work on my hobbies and ideas. Google seems to understand, with their hobby Friday model.
Seems pretty redundant to me that I should have to pay the DVD playback royalty (we past costs on the the consumer, we're a business) to Nintendo regardless of whether or not I have 0 or 5 things that already play DVDs.
If the PS2 didn't play DVD out of the box, it'd be cheaper to make. The Xbox didn't play DVD out of the box because it was cheaper to do it that way.
Seriously, I can buy a progressive scan DVD player at Wal-mart better than the Nintendo Wii's will probably be for less than the price of a universal remote. Including it would be stupid.
My user CSS ensures that links are underlined. This is forced on. Otherwise, I get to play magic wand. Ditto for image links: if it's an image, it has a bright, blue border around it. Then I don't have to play magic cursor.
I have seen too many people, "is that a link?" (mouses over it, clicks at it), "I guess not."
I like taking the guessing out of the web. You can even set your user CSS so that you have to click on a flash before it can play! It's great.
Slashdot Mirror
"what was once a pretty sweet console."
What pretty sweet part are we talking about? 2 HDMI ports I can't use without a 2,000$ new TV? 7 player bluetooth, when I rarely (never) have a situation where I go, "damn, I wish my GameCube had 8 ports so everyone could play Mario Party instead of just 4 at a time"? The part where the PS3 is also an Internet router, instead of my current one, with 3 gigabit ports?
Sony went and said, "everything those guys have, plus EXTRA!" for the past 3 years. Like the online service, next generation graphics, or any of that other shit I mentioned.
Every year, Sony Fan Bois have been going, "OMG CREAM" about it. There is no such thing as a sweet Sony console; it's always a pack of lies. The only reason the PS1 got popular was because everyone hated pompous Nintendo and their "screw everyone" mentality in the mid-1990s. How many 1st party titles (and I don't mean 2nd party, like Polyphony Digital) have there been on the PS1 and PS2 that have been super awesome? How does that number compare to the titles by 3rd party developers?
Sony sells far less PS1s and PS2s than Square Enix, Namco, Capcom, or Konami -- even alone. Hell, the only thing the semi-rational fan bois seem to want on the PS3 is Metal Gear 4 -- a Konami title.
"This is getting out of hand, and you can apologize for your (or my) government's intrusive behavior all you want, but the truth is that everyone will, sometime, somewhere, do something he'd rather other people didn't see. "
Why would someone do something they didn't want other people to see? Well, it could be because they were arranging something (surprise birthday party), or it could be because they were doing something illegal. Now, you can't legistlate right and wrong, but you can draw a legal boundary around certain actions.
So, taking this a bit further, what's so bad about using a system to monitor legal or not actions? If there are extenuating circumstances (a good reason for doing something that bends/breaks the rules), isn't that something you present in a court of law? Innocent until proven guilty?
Within the privacy of my own home, I have complete domain. Outside, I have no domain -- public areas are for everyone. Don't do things there you don't want to possibly have other people know about.
Unlike TCP, where the end-points do all the thinking, multicast requires that the routers are involved in the transactions. They are the ones who have to make decisions like, "does this address get bits, or not?"
The session management protocols of multicast are defined, but there are a few to choose some, and most have some kind of serious drawback associated with them. One of the ones that sticks out in my mind is the one where there's no way to "detect" if a multicast IP is taken, or any more security/authentication than knowing what the address is.
To properly support multicast, we need a session leader, and every router involved in the minimum-cost spanning tree must also know who else is involved. This means the routers have to be able to build the tree, and tear it down as clients join and leave.
Replacing or upgrading routers is hard because a lot of them are fire and forget. They'll place a router in a wall with PoE, and then leave it inside. They'll be on the bottom of the ocean, repeating traffic that goes along a trans-oceanic link. They'll be on top of wireless towers, miles from other people. Most of them were not designed to be remotely upgradable via software, because routers were always meant to be as cheap to produce as possible.
This is also who IPv6 is only really deployed in places where IP space ran out a long time ago (such as Japan). Until it really starts to break, traditional structure will be "good enough" for most people.
"Under the current regime, millions of households a day are breaking the law when they tape a show and watch it at another time."
Imagine that, an entire nation composed of criminals!
I guess it's true; history repeats itself.
Ken Kutaragi will not get a cent from me.
The PS2 was supposed to support Linux. How many homebrewers bought the PS2 dev kit and actually had success?
How about the amazing no-show on the PSP?
And now the PS3 is supposed to be friendly? It'd be cheaper to buy a MacMini and howebrew up something for Ubuntu or OS X than the Sony PS3.
Capitalism is a funny system because of a few basic assumptions it needs to function well. One of those assumptions is that users will know things, like what is a better product to buy. Because of this, people who sell shoddy equipment on unreliable gear will not succeed.
Now let's apply this. I have a PowerBook that is very reliable. I also have a desktop that's very reliable (in fact, 3). However, these desktops are component-based machines; they run Linux. How is it that these component-based macchines are as reliable as my end-to-end model PowerBook? I bought components which aren't garbage. AMD CPUs, Kingston lifetime warranty RAM, Enermax power supplies, etc. It's more expensive than what most people probably buy, but I've never had a peap of trouble. I know what components to buy because I take the time to look into it, and because I only buy components that the Linux kernel supports (which, for some reason, happen to me more reliable than random Taiwanese garbage).
With Apple's model, we skip this step. Apple themselves takes the time to try and get quality components that work reliable with OS X. Since they vend the machine and the hardware, they can't hide behind the "Windows sucks" excuse the way cheaper component suppliers can. However, and this is important to note, they're still interested in shrinking costs as much as possible to maintain their fat margins, and they still like to charge a high markup. Plus, they're not immune from mistakes (note the GOBS of heatsink goo on the heatpipes of the 15" MacBook Pros). This means they don't always do as good a job as someone who knows what I do.
Really, it's just moving the burden of choosing chocolates from shit from the consumer up the chain a bit, but even then it's not perfect. If you want thinks done right, do it yourself -- learn about PC construction, or pay someone you trust (be it Apple or your friend). If you just go buy the cheapest thing you can, you're on a roller-coaster ride to the bottom in terms of quality and consistency -- that's why Wal-mart's stuff is different (they have different product badged the same to cut costs), and also why Wal-mart is not always the best place to shop.
Adam Smith's invisible hand requires you to do research!
Code size? Memory usage? Execution speed?
These all affect each other. The binaries are different between different arches, duh! The fact is that you invalidate fewer cache lines with smaller x86 instructions, and those Intel-based Macs are way faster than the PowerPC ones.
World of Warcraft on my friend's 15" MacBook Pro blows the doors off my 12" Powerbook G4 of the previous generation. We're talking 20fps average vs. 50!
You start off by saying that what I said was not math, but algebra.
For about most of of humans on this planet, that's all math is. We're very much in the minority. To "normal" people, math and computer programming are very different, much like physics is very different or logic is very different (even though all these things only differ in degree).
I also said that logic, since it expands and describes more in a rational system than pure maths (although the definition of mathematics is also expanding to include many things done in logic), I'd say that logic in all its forms (including programming systems) are a superset of mathematics. You can do more in a formal logical system than you can do in a pure math system -- a Turing machine does more than a push-down automata. "Modern" math is just expanding into areas that were explored by thinkers as far back as Leibniz, who formulated constructs not unlike Turing machines.
What is important is the perspective. Mathematicians are taught in a way different from computer scientists, and thus they see solutions to problems in different ways (there are thousands of mathematician/physics/engineer jokes I can put in here, you get the idea). Computer scientists solve problems of thought, while mathematicians solve problems of relationships. But the thoughts will contain (naturally) thousands of relationships described (implicitly or explicitly) along the way towards the computation of a solution. Computer science is the superset of math the same way my toolbox is the superset of my wrench. Mathematicians tend to see the world as a set of bolts, while computer scientists are free to apply solutions from other logical disciplines.
"Now is your turn, name someone who does a lot of programming and has no mathematical knowledge doing something really significative for computer science. "Web 2.0" reimplementations of 30+ years old concepts are not significative contributions."
There are 2 kinds of people; those that understand things in the larger picture perspective, and can create or discover new ideas. These are people who will, naturally, have some understanding of mathematics. Alternatively, there are those that are like myna birds -- excellent at repeating what they have been told. Much like a pocket calculator is great at adding and dividing, a programmer without math is only good at repeating what is trivial to people who know the bigger picture.
It's impossible for a pocket calculator to discover new things; you ask the impossible. Try to understand why I believe math is a subset, and explain where I have a flaw rather than get distracted with trivia.
"In fact, math is harder, and after doing some higher maths, you will surely be a better coder. Maths expands your mind."
Math is easier than programming. In math, you only express a relationship between things in numerical terms. In the more robust logic systems of a computer, you also have to take into account many more variables (unless, of course, you're doing programming purely on paper).
Math is more formal than computer science thanks to the push towards better proofs in the 19th and 20th centuries; Computer Science is currently around where math was in the 15th and 16th century -- lots of hacks working at it, proofs for sale, "instinctive" and geometrical proofs unbacked up by any kind of algebra. Just because that makes it seem easier doesn't mean it is. I'm sure Dijkstra, an accomplished mathematician, would have harsh words for those that say math is harder.
Math is a subset of programming. People who can't see that the logical expression in terms up pure numerical values is a subset of higher-order logic are like Leopold Kronecker, who opposed set theory as having any relationship to mathematics in the 19th century. Nowadays, most mathematicians agree that set theory and mathematics are correct.
There are proofs that are very hard to do with traditional mathematical techniques that end up being only a few lines when you apply Turing's approaches.
Apparently they're charging roughly 35 to 40$ USD a month, and 40$ CDN a month for their "North America Unlimited" plans.
Currently, VOIP outgoing providers sell time for about a cent per minute in North America, or as high as 1.5-2.0 cents for an incoming number. My roomate and I have a 1-888# which comes in to the house for a few cents a minute, with outgoing numbers going through 3 different VOIP providers of varying cost (cheapest first, of course). I think it costs us a grand total of 8$/month-2 months depending on usage.
Basically, Vonage is making so much money because traditional POTS is a cash cow, and VOIP makes it ridiculously cheap since you don't even need to run fibre -- BYOC!
"You can do simple things easily - and in particular, you can do things where the information only passes in one direction quite easily, but anythign else is much much harder, because there is no "shared state" (by design). And in the absense of shared state, you have a hell of a lot of problems trying to make any decision that spans more than one entity in the system."
I think you're looking at this the wrong way around.
There has been a lot of research into this over the past 40 years, ever since Dijkstra first talked about coordination on a really big scale in the THE operating system. Any decent CS program has a class on distributed programming. Any decent SW architect can break down these different parts of the OS into weakly-connected pieces that communicate via a message passing interface (check out this comment by a guy talking about how Dragonfly BSD does this).
It's obvious that breaking something like your process dispatcher into a set of processes or threads is silly, but that can be easily separated from the core context switcher. Most device driver bottom halves live fine as a userland process (each with a message-passing interface to their top-halves).
If you're compiling for an embedded system, I'm sure you could even entirely remove the interface via some #define magic; only debug designs could actually have things in separate address spaces.
The point I'm trying to make is: yes, you can access these fancy data structures inside the same address space, but you still have to serialize the access, otherwise your kernel could get into a strange state. If you mapped out the state diagram of your kernel, you'd want the transistions to be explicit and synchronized.
Once you introduce the abstraction that does this, how much harder is it to make that work between processes as well as between threads in the kernel? How much of a benefit do you gain by not having random poorly-written chunks pissing over memory?
How about security benefits from state-machine breakdowns being controlled and sectioned off from the rest of the machine? A buffer overflow is just a clever way of breaking a state diagram and adding your own state where you have control over the IP; by being in a separate address space, that poorly written module can't interact with the rest of the system to give elevated privileges for the attacker (unless, of course, they find flaws in more of the state machines and can chain them all together, which is highly unlikely!).
Clearly there is a security benefit as much as there is a consistency benefit. Provably correct systems will always be better.
"All popular OS kernels are monolithic."
I must've skipped into a parallel world where Windows XP, Windows 2000, and Windows 2003 were not somehow based on the microkernel NT Exective.
"Fact: OSX is sooooo slow that the only thing it is faster than is OpenBSD. And you cant even blame its slowness on it being a microkernel. How pathetic... Wow, that says it all in my book :)"
Actually, OS X was within a few percentage points of Linux on all hardware tested; actually outperforming it on memory throughput on PowerPC and some other tests. It's also faster than NT.
"But there are ALWAYS situations where it is going to be desirable for seperate parts of an OS to directly touch the same memory in a cooperative manner, and when this is the case a microkernel just gets in your damn way..."
I fail to see how L4's shared-memory is somehow magically different from Linux's shared-memory. Once you let more than 1 process have access to a page via a TLB mapping, it's all the same.
Slashdot may be news for nerds, but it has a serious drawback when it comes to things such as this. The drawback is that what is accepted as "fact" by most people is never questioned.
"Fact": Micorkernel systems perform poorly due to message passing overhead.
Fact: Mach performs poorly due to message passing overhead. L3, L4, hybridized kernels (NT executive, XNU), K42, etc, do not.
"Fact": Micorkernel systems perform poorly in general.
Fact: OpenBSD (monolithic kernel) performs worse than MacOS X (microkernel) on comparable hardware! Go download lmbench and do some testing of the VFS layer.
Within the size of L1 cache, your speed is determined by how quickly your cache will fill. Within L2, it's how effecient your algorithm is (do you invalidate too many cache lines?) -- smaller sections of kernel code are a win here, as much as good algorithms are a win here. Outside of L2 (anything over 512k on my Athlon64), throughput of common operations is limited by how fast the RAM is -- not IPC throughput. Most microkernel overhead is a constant value -- if your Linux kernel us O(n) or O(1), then it's possible to tune the microkernel to be O(n+k) or O(1+k) for the equivalent operations. The faster your hardware, the smaller this value of k since it's a constant value. L4Linux was 4-5% slower than "pure" Linux in 1997 (See L4Linux site for the PDF of the paper).
But none of this is something the average slashdotter will do. No, I see lots of comments such as "micorkernels suck!" already at +4 and +5. Just because Mach set back microkernel research by about 20 years, doesn't mean that all micorkernels suck.
Graffiti v1 is very easy to learn, easy to read, and easy to decipher on a machine.
Check out this representation of the alphabet in Graffiti.
You can do X as a reverse of a K in that alphabet; U and V were a bit different (V is easier to do right-to-left for the machine to recognize the stroke, but you can make the shape the same as a "real" v). I actually did some of my paper notes in Graffiti (in University) since they tend to be mono-strokes (rather than the polystrokes to make up them more complicated letters when we write out traditional english), making it easier for shorthand note taking (which has a strict stroke form/order dictated by speed and readability!).
There you go: simple strokes, easy to read by eye, easy to read by my 68000-based Palm from 1998 as well as my current StrongARM Palm of 2004!
"Instead of cut and try, cut and try, cut and try, like the hot rod guys do, you have to do a whole bunch of computer analysis before you build it," he said. "We did (computerized) structural analysis and we did stability analysis. And by God, you know what happens? It works! Duh."
:)
I have to agree with him regarding hot-rodders. A lot of people seem to think the way to solve a problem is to frob at it until you get something that works. All the Motorola phone hacking kids, Xbox homebrewers, and PSP kiddies seem to think that the spackle approach (throw things at the wall until something sticks) is the best way to solve problems. You know, rather than solving them by understanding them
Straight from the innovations in software page, we have: "As patentability has increased, there's good evidence that the number of software innovations has decreased. Bessen and Maskin also demonstrated a statistical correlation between the spread of patentability in the United States and a decline in innovation in software. In particular, between 1987 and 1994 , software patents issuance rose 195%, yet real company funded R&D fell by 21% in these (software) industries while rising by 25% in industries in general. This paper gives additional evidence that software patents are inversely related to innovation; it's hard to not notice that as patenting become more common (e.g., 1987 and later) that the number of major innovations slowed down and are almost always not patented anyway."
The link supplied is to this PDF about patents. It's worth your time to read about this research.
But when you can make your whole house wired with 1Gbps ethernet in 2006 for half the cost what it cost your home to be wired with 100Mbps ethernet in 2000, isn't it reasonable to expect that ISPs aren't paying the same peering fees they used to?
Isn't it reasonable to expect that the interconnections are faster, and can handle more? Isn't there a near zero cost of adding more users, as long as the pipe isn't oversubscribed?
I think a lot of it is that business models for ISPs are based around a much longer cycle than technology will permit them to have. Anytime a disruptive technology comes along, must we have a debate about how the old, no longer valid assumptions still hold true?
Distributed content replication is one of the key features of the Internet, which is why the MPAA/RIAA hate it.
"It is not the same. The PSP is OFF while the DS remains on. Consuming less power to be sure but still draining the battery. The really big difference? If power fails because of an empty battery the PSP can resume where it left off as if nothing has happened. The DS loses its progress."
Not true. The PSP, if left in sleep mode, will die out. It still periodically refreshes its RAM, even if the cores are turned off entirely. I own(ed) a PSP and left it sleeping. When it was totally dead (would not resume from the button), it did the fully power on Sony logo the next time it was plugged in, and had lost all progress in the game. The PSP does not sleep to flash.
"As for speed of writing and reading. No competition. Nintendo has chosen not to have onboard save. That no doubt saves them money meaning that Nintendo makes a profit on a cheaper console but now the save file has to be in the game catridge and it is not going to be top of the line because of again the cost issue."
What!? I'm not sure what you're trying to say here. It's true the DS doesn't have a save slot in the device, because the carts can have a little flash spot in them.
"AC:WW takes about 30 secs to load and 30 secs to save. Since I love the DS lite for its ease of use in those lost 5 minutes that is a bit too long. The PSP is a bit to large to carry in my pocket but it allows me to play faster and quit faster."
I'm not sure you're quite aware of the time factor required to seek around on those PSP UMDs. They are no where near the speed of ROM or flash. I can fold my DS up instantly and have it sleep for about 4 days of standby time (equivalent to the PSP sleep mode), but startup is way faster. Getting into Brain Age is a matter of tapping the screen a few times. With Lumines, "Now Loading" pops up, and then a whole lot of seeking would happen. Some games have upwards of 45 second load times!
I load a lot more than I save, and it seems to me that the DS is better optimized for this usage.
claiming things like breakage sounds a lot more like the anus defence in terms of "getting off" with the artists' money.
Subverting legal agreements seems to be too acceptable to these types.
There's as fairly informative essay on work patterns I read recently.
"On the other hand, the "market" for free time hardly even exists in America. With few exceptions, employers (the sellers) don't offer the chance to trade off income gains for a shorter work day or the occasional sabbatical. They just pass on income, in the form of annual pay raises or bonuses, or, if granting increased vacation or personal days, usually do so unilaterally. Employees rarely have the chance to exercise an actual choice about how they will spend their productivity dividend. The closest substitute for a "market in leisure" is the travel and other leisure industries that advertise products to occupy, our free time. But this indirect effect has been weak, as consumers crowd increasingly expensive leisure spending into smaller periods of time."
It's taken from a book by a Harvard sociology professor, Juliet B. Schor.
I think it summarizes the weird dichotomy I see here. Some yanks just can't get enough of that 9 to 5, like it's their way of buying a stairway to heaven or something. I'd rather enjoy my time off, but if I were to live in the US, I'd have no real way of enforcing this since employers discourage less pay for more time off.
"We don't want to be like the Europeans. Generally speaking, the US produces far more and creates much more wealth than most European countries."
Since when was my main goal in life declared to be wealth creation, instead of something a bit more hedonistic and less puritan?
I mean, if you want no vacation time a year, go ahead. I'd much rather have a couple months of paid vacation to work on my hobbies and ideas. Google seems to understand, with their hobby Friday model.
DVD video requires CSS keys, licences, region management, DRM, etc.
Not including DRM in a device and not requiring to rebuy features I have is not a gaff. You are a troll.
Seems pretty redundant to me that I should have to pay the DVD playback royalty (we past costs on the the consumer, we're a business) to Nintendo regardless of whether or not I have 0 or 5 things that already play DVDs.
If the PS2 didn't play DVD out of the box, it'd be cheaper to make. The Xbox didn't play DVD out of the box because it was cheaper to do it that way.
Seriously, I can buy a progressive scan DVD player at Wal-mart better than the Nintendo Wii's will probably be for less than the price of a universal remote. Including it would be stupid.
My user CSS ensures that links are underlined. This is forced on. Otherwise, I get to play magic wand. Ditto for image links: if it's an image, it has a bright, blue border around it. Then I don't have to play magic cursor.
I have seen too many people, "is that a link?" (mouses over it, clicks at it), "I guess not."
I like taking the guessing out of the web. You can even set your user CSS so that you have to click on a flash before it can play! It's great.