They were exactly what MS needed - some humanization. They were sly. Some people here didn't seem to get or like them (well, they hated them for the most part) - but they made sense. They were the kind of ads you put out when you're winning, which, make no mistake, MS is.
This new ad is just sad, they seem like a desperate response. The kind of commercial you do when you're losing. That's not the vibe they need.
Hate to just pile on - but completely agree. For me, Pandora died a while ago (since I'm in Canada).
I haven't bought a ton of music lately, but what I have has been:
1. Stuff I found on Pandora. Their model was such that you were constantly bumping into new things that I actually liked (because of their excellent related music search). I've found 4 or 5 artists I really like, and have bought at least one album from each. 2. Stuff I ran into on YouTube - most recently, episodes of Mitchell & Webb. After seeing a few bits on Youtube, I just had to get all their stuff (and if there was more to buy on iTunes, I would - but of course there's only certain things they'll sell in Canada). 3. Stuff I found on eMusic, or on the (very) old MP3.com.
What's keeping me from buying more media?
1. It's hard to be find other stuff I like. The radio is useless, I don't have much time to watch TV, and the bulk of online content doesn't have samples, doesn't play in Canada, etc. I'm sure there's more stuff I want out there, but I'll have to be very lucky to run into it and I don't have time to search through garbage. 2. It's hard to buy what I want. Good luck, for example, getting new episodes of Dexter here (or Mitchell & Webb) online. I promise I would pay. There's all sorts of stuff I do pay for. Just take my money!
Completely agree. I've installed Starcraft probably 20 to 30 times. More if you count multiplayer "spawn" installs and what not.
(And I've bought it 3 times now. Not because it quit working, but because I lost the discs once, and once gave my copy away.)
Spore is intended as a huge, open-ended game with user-created content and lots of gameplay "meat": the kind of game that should be lasting 5 or 10 years. 3 installs isn't going to work for that for a lot of people. Not just people who upgrade a lot or move around or crash a lot - just regular people who like the game for more than the first year or two - are going to get bitten by this.
Sure there's differences in how much code it will take to implement an idea, and there's differences in runtimes, available libraries and what not.
I guess my point would be that for an interesting algorithm contest, none of those things are going to be much more than tie breakers. Now obviously if the problem is, "multiply these 100 digit numbers" it's going to be easier in a language that an arbitrary precision math class - but I wouldn't count that as much of an interesting contest. Similarly there might be problems that are nigh unsolvable in a certain language due to speed, but again that's not the case for a well designed contest. Things like garbage collection and what not are only going to enter in to a contest if it's very poorly designed (too tight of time bounds or a bad judge, etc..).
I guess my other point is that if "TopCoder only provides the option of using some of the worst tools for any job", and "good programmer picks the best tool for the job", then it's strange that the same set of guys that win at TopCoder tend to win at more open programming competitions (GCJ, the last ICFP, ICPC), and they do it - in large part - with those same horrible languages.
To the extent that language is important, and that the best programmers would tend to choose the best language I think empirically we'd have to conclude that some C variant is the best language for algorithm competitions.
My own view is that that's not really the case, and that its dominance is a combination of unrelated circumstance about it's availability and popularity combined with the fact that language choice just isn't that important.
Meh. The currently running Google Code Jam let people use whatever they wanted. In the last round (round 3), there was 1000 people. 3 used Haskell, 3 used Lisp, 1 used OCaml. Of those, 1 Haskell and 2 Lisp users got through to the round of 500 (and one of those was reid, who could have advanced using baling wire and twine). The remainder of advancers used C++, Java, Pascal, Python, and a couple other boring procedural languages. One poor fool used VB (me).
I think a good programmer can solve these kind of problems in any language. Sure some competitors might be more comfortable in one language or another, but in the end the meat of the solution is going to be the same anyways.
For TopCoder, their framework and style makes it hard to support a lot of different languages. And in terms of whether they picked the right ones, the statistics of what people pick when they have the choice suggests that they chose well.
I hated the old RPGs with their mindless teleporter mazes and talk to everyone on the continent riddles. I hated the adventure games where you click every item on your inventory on every object on the screen, and hope that your pie will take out the yeti (or whatever). I also hate the loads of action or adventure games that have a random "sliding brick" , "connect the pipes" or "towers of hanoi" puzzle. I never need to see any of those again.
Perhaps puzzles in games aren't as popular now not because the new generation isn't used to them, but because the old generation (the one that's writing games) learned to hate them.
(Note - if you actually like puzzles in a game, I recommend Space Rangers 2).
Actually, the computer WOULD be predictable, exactly because it will always choose some "optimal" move.
No, it wouldn't. Read the article I linked to - it's a good introduction to the relevant game theory.
But, to simplify: an "unbeatable" strategy doesn't imply one move for each game state. Take rock/paper/scissors. The computer doesn't have to analyze your moves and try to guess what you're going to do. All it has to do is randomly pick a move and, over time, it will come out even. It's an unbeatable strategy.
Poker is more complex than RPS obviously, but the principle is the same. When looking at poker, instead of just one move, we have to consider an entire strategy as a move (ie. a complete list of reachable game states and an associated play for each of those states).
Given strategies for each player, there's an expected value of a hand over all possible cards. For each hand, we choose a strategy probabilistically such that our expected value across possible opposing strategies is maximized. Note that a given hand may have a negative expected value since one player goes first, etc - but overall the expected value over many games will be zero.
If you want to play around with this a bit, try the TopCoder Marathon Match problem "TwoCardDraw". It's a simplified version of poker, but it's nonetheless a challenging problem and an interesting coding strategy. It involves resolving the expected value of strategies, coming up with "perfect counter" strategies, and lots of other fun, relevant stuff. (I did OK - my submitted solution was a static strategy, but I experimented quite a bit before that).
And if you leak your source code, anyone who knows they're playing against your bot will have a huge advantage.
No. Knowing the source code of an ideal poker program would do you no good. You can't win against an agent following the Nash equilibrium for a game. He's going to be choosing from a library of strategies such that for every complete game, whatever strategy you choose, the expected value of the game is zero.
Naturally such an ideal program is hard to write, but conceptually it's very simple - poker is equivalent to any other zero sum game.
Pros may have an advantage against other humans (that they could read) but they wouldn't have any information advantage against an ideal computer player.
You have to understand: the ideal computer would not be predictable - it wouldn't always be choosing some "optimal" move in each situation. Rather, it would choose strategies probabilistically such that there is no strategy that would win against it over many hands.
Poker is complicated, but it can still be solved as a zero sum game treating every possible strategy as a single move.
Well hidden is a lot different to a human than an automated filter. Most people are going to have keyword searches that span newsgroups, so making your material findable to someone will still be as easy as putting a description in the subject line.
In terms of "they're watching", I'm sure these people are plenty concerned about anonymity anyways - and apparently to this point have found ways to post that hide their identity. What they're doing is already against the law, and there are already people trying to find them.
I'll grant perhaps removing some newsgroups might make the material more difficult to find for some users - but I also think it will put the material in front of more people who weren't looking for it.
No, there won't be more bandwidth. They'll still post it - it's just now instead of posting to alt.underage.porn (which won't exist) they'll put it in alt.porn (where people will be exposed to it that otherwise wouldn't have been).
And if they shut down alt.porn (because it's full of child porn now) they'll start putting it in the rec.ham.radio.
If you have a smelly toilet, you can't just say: "oh, well, we'll just board up that door so nobody poops there anymore". People are going to poop, and if there's no bathroom they're going to poop in the kitchen.
And the firewall analogy fails. A firewall makes it harder to attack my computer. Having no alt.underage.porn doesn't make it harder to post kiddy porn - it just makes it harder to avoid.
I use newsgroups quite a bit. Once alt.underage.porn (or whatever) is shut down, that material is just going to be posted somewhere else - and probably end up being seen by more people. If they ban keywords, they'll move onto new euphemisms. No automatic filter will do this job - and the results of the attempt will be worse in every way than if no filter was used.
All it is is scoring political points, and providing the illusion of action while really making the situation worse.
When you're identifying a bird, you might say that it matches the "hue and cry" of a certain species. Hue means color, and naturally cry refers to the noise the bird makes. That's where this term came from - and usually it's used in the same situations where you might say "Well, if it walks like a duck..."
It doesn't make sense for there to be a "hue and cry" as a reaction to something. Unless, perhaps, you imagine people are going to change color in anger.
But I'm guessing your interpretation is as though both words are verbs, like people are "hew"ing and crying in anger or something.
I posted one example, and it would easy for you to find others. For myself, I don't feel any particular obligation to further establish this.
As to the guy's benchmarks, they seem fairly reasonable to me. I don't think it's unreasonable to include things like vectors and hashmap implementations when talking about the speed of a language - and the benchmarks this guy was countering did the same thing.
You're also confused on my motivation, I think.
I don't hate Java. I was arguing against the idea that it's so much faster than C++ that nobody measures it anymore. Even if Java would outpace C++ if it was let to run for longer or something, we'd still be left with the conclusion that there are some tasks C++ is faster for (which is all I was trying to establish).
You're correct in that people don't generally choose C++ based on performance. But it does happen that non-C++ coders will switch for certain problems.
As an example, As here is Kawigi (a fairly high level competitor who usually uses Java), switching to C++ so he can brute-force a problem instead of using DP.
Notwithstanding that, all TopCoder problems (well, all correct ones) have a Java solution that runs in under 1 second - as that's the requirement on the reference solution. Where you'll see switches to C++ is to make other approaches feasible.
This may be a significant factor for some of his smaller tests, but I don't think it jeopardizes the overall conclusion. For example, the Ackerman test was 15s vs. 60s. There isn't going to be more than a second of startup, and most of the tests are showing differences on the order of many seconds.
In terms of "is it harder to write fast code in C++", I guess it depends on what you're good at. Looking at the changes he did to the C++ code, none of it is esoteric - but I'm also more interested in this kind of thing than most of the programmers I know. I think it's fair to say many coders would produce faster code for these functions in Java than C++.
All in all, it's a subject that probably gets more attention than it deserves. Like you say, Java is plenty fast. Whenever I see performance problems at work, I never think "Ooh, let's rewrite this in a lower-level language" (and very seldom would that help). I think, let's look at the SQL, look at the bandwidth, look for other resource conflicts, or change overall approach.
Java is perfectly fast for real world applications, and I'd agree that the "Java is Slow" idea is outdated.
But it's not conclusively faster than C++, at least not in a general sense. In terms of a small task involving lots of simple operations, you'll still often see a significant speed increase using C++. This is a good example. Now I'm sure there's more optimizations available on both sides - and plenty of stuff to tweak - but C++ is going to come out ahead by a significant margin on a lot of these tasks.
A good example where the participants on both sides have some motivation is on TopCoder (where I spend a fair bit of time). Performance isn't usually the driving factor in language choice there - but sometimes it is, and when it is the answer is pretty much always C++ (unless it's a comparison between Java BigInteger and a naive implementation of the same in C++).
Reasonably often you'll see people write an initial solution in Java, find it runs a bit too slow, and quickly port it to C++ (or pre-emptively switch to C++ if they think they'll be near the time limit). It's not uncommon to see a factor of two difference in performance.
To be clear - these are not usually "real world" tasks. As more memory and objects come into play, Java is going to do better and better. But these kinds of tasks still exist - there's still plenty of places where C++ is going to be the choice because of performance.
In any case, your contention that Java is so much faster that nobody does benchmarks anymore is unsupported and wrong.
The control is a Forms style control. Obviously I can set properties on it at the server side. And I was using components on the control to communicate with the server - the problem was getting information from the control to the script (without an egregious amount of polling or ridiculousness).
And I'm not "automatically assuming" there is security - this control runs only on one Intranet page where we have complete control of security and full trust.
Anyways - as I replied to the other guy, it does appear there was a somewhat more reasonable solution I missed. You can manually write COM (ActiveX) style events for your.NET control, at which point it can signal to the remainder of the page successfully.
It's funny, Googling for this now I am getting some better results (than when I implemented it a while back). It appears you can get events to work, but you have to write the ActiveX event code yourself (rather than relying on.NET events, which apparently IE can't translate).
I'm sure that's what you meant.
In any case, the current solution works fine (and was probably actually easier to implement) - but still it seems a bit silly they didn't make this work reasonably, or at least provide some hints on how to get it working.
One of our recent projects involved hosting a.NET control in Internet Explorer as part of an Intranet page for editing and uploading photos. The control worked fine, and we were able to manage security permissions such that it could access the files it needed to (and even delete them after as desired - it has full trust).
The problem was that we wanted the page to refresh after the upload was complete. This seems like it should be fairly simple, but with how the security works, there's not a simple way to communicate between the.NET control and the surrounding page (you can't, for example, just call a function on the page, despite the fact that you have full trust).
With that being the case, I tried editing a property on the control once it was finished uploading, but that too didn't work as the page couldn't read them (even properties like "height" that it could infer couldn't be read directly). I thought about some fairly complicated solutions involving polling using XMLHTTP, partial refreshes or scripting or something - but what I ended up doing was this:
1. When the page loads, take note of the position of two elements on the page. 2. Poll the position of these elements a few times each second. 3. When the upload control finishes, it increases its height by 4 or 5 pixels. 4. This displaces the other elements on the page, and this can be used as a signal that an upload is complete and a refresh should be done.
This ended up working fine, but I felt ridiculous for having done it this way. There's no security being added here - the control and the page can communicate all they want via the server or scripting (or 100 other ways), it just means that using the controls is much more difficult and obscure than it needs to be.
I see what you're saying - but in practice I don't find it's much of a problem (the setup I mention is pretty much the same as I have set up now, which is why my I mention old ML models rather than whatever they have going now.. I'm also using an LCD until I finish the dedicated room in my new house).
I guess I'm not really much of an audiophile when it comes to movies. As long as you have sufficient power, a subwoofer, and surround, I don't notice too much else.
Where I do notice quality is in music - and having 2 good speakers alone has, to this point, been enough for that to sound great. I guess in the future we'll probably see more quality music with more channels, and I guess then I might upgrade the other speakers.
I wouldn't call this a home theater - and I don't think it's intended to be one. It's a technology showcase / bunch of useless crap. Besides that, how could you call it a "theater" with such a tiny screen?
Here's how I'd spend $18000:
1. Epson Powerlite 1080UB (projector) = $3000 2. Pair of Martin Logan Quest front speakers = $10000 3. Decent amp = $2000 4. Random center/rear channel speakers = $800 5. PS3 = $400 6. Decent 100" 16:9 screen = $500 7. Random subwoofer = $400
Now you're set up to watch movies, play games, listen to music, whatever - and your friends won't laugh at your pitiful 50" plasma.
And if you don't have $18000, substitute in a few cheaper alternatives and you can do a very decent theater for $3000 and still have a setup people will like more than the one in this article.
Slashdot Mirror
They were exactly what MS needed - some humanization. They were sly. Some people here didn't seem to get or like them (well, they hated them for the most part) - but they made sense. They were the kind of ads you put out when you're winning, which, make no mistake, MS is.
This new ad is just sad, they seem like a desperate response. The kind of commercial you do when you're losing. That's not the vibe they need.
Hate to just pile on - but completely agree. For me, Pandora died a while ago (since I'm in Canada).
I haven't bought a ton of music lately, but what I have has been:
1. Stuff I found on Pandora. Their model was such that you were constantly bumping into new things that I actually liked (because of their excellent related music search). I've found 4 or 5 artists I really like, and have bought at least one album from each.
2. Stuff I ran into on YouTube - most recently, episodes of Mitchell & Webb. After seeing a few bits on Youtube, I just had to get all their stuff (and if there was more to buy on iTunes, I would - but of course there's only certain things they'll sell in Canada).
3. Stuff I found on eMusic, or on the (very) old MP3.com.
What's keeping me from buying more media?
1. It's hard to be find other stuff I like. The radio is useless, I don't have much time to watch TV, and the bulk of online content doesn't have samples, doesn't play in Canada, etc. I'm sure there's more stuff I want out there, but I'll have to be very lucky to run into it and I don't have time to search through garbage.
2. It's hard to buy what I want. Good luck, for example, getting new episodes of Dexter here (or Mitchell & Webb) online. I promise I would pay. There's all sorts of stuff I do pay for. Just take my money!
Completely agree. I've installed Starcraft probably 20 to 30 times. More if you count multiplayer "spawn" installs and what not.
(And I've bought it 3 times now. Not because it quit working, but because I lost the discs once, and once gave my copy away.)
Spore is intended as a huge, open-ended game with user-created content and lots of gameplay "meat": the kind of game that should be lasting 5 or 10 years. 3 installs isn't going to work for that for a lot of people. Not just people who upgrade a lot or move around or crash a lot - just regular people who like the game for more than the first year or two - are going to get bitten by this.
I assume what they mean is it's $84 each, when purchased in a lot of 1000.
Sure there's differences in how much code it will take to implement an idea, and there's differences in runtimes, available libraries and what not.
I guess my point would be that for an interesting algorithm contest, none of those things are going to be much more than tie breakers. Now obviously if the problem is, "multiply these 100 digit numbers" it's going to be easier in a language that an arbitrary precision math class - but I wouldn't count that as much of an interesting contest. Similarly there might be problems that are nigh unsolvable in a certain language due to speed, but again that's not the case for a well designed contest. Things like garbage collection and what not are only going to enter in to a contest if it's very poorly designed (too tight of time bounds or a bad judge, etc..).
I guess my other point is that if "TopCoder only provides the option of using some of the worst tools for any job", and "good programmer picks the best tool for the job", then it's strange that the same set of guys that win at TopCoder tend to win at more open programming competitions (GCJ, the last ICFP, ICPC), and they do it - in large part - with those same horrible languages.
To the extent that language is important, and that the best programmers would tend to choose the best language I think empirically we'd have to conclude that some C variant is the best language for algorithm competitions.
My own view is that that's not really the case, and that its dominance is a combination of unrelated circumstance about it's availability and popularity combined with the fact that language choice just isn't that important.
Meh. The currently running Google Code Jam let people use whatever they wanted. In the last round (round 3), there was 1000 people. 3 used Haskell, 3 used Lisp, 1 used OCaml. Of those, 1 Haskell and 2 Lisp users got through to the round of 500 (and one of those was reid, who could have advanced using baling wire and twine). The remainder of advancers used C++, Java, Pascal, Python, and a couple other boring procedural languages. One poor fool used VB (me).
I think a good programmer can solve these kind of problems in any language. Sure some competitors might be more comfortable in one language or another, but in the end the meat of the solution is going to be the same anyways.
For TopCoder, their framework and style makes it hard to support a lot of different languages. And in terms of whether they picked the right ones, the statistics of what people pick when they have the choice suggests that they chose well.
Most puzzles suck (now, and in the past).
I hated the old RPGs with their mindless teleporter mazes and talk to everyone on the continent riddles. I hated the adventure games where you click every item on your inventory on every object on the screen, and hope that your pie will take out the yeti (or whatever). I also hate the loads of action or adventure games that have a random "sliding brick" , "connect the pipes" or "towers of hanoi" puzzle. I never need to see any of those again.
Perhaps puzzles in games aren't as popular now not because the new generation isn't used to them, but because the old generation (the one that's writing games) learned to hate them.
(Note - if you actually like puzzles in a game, I recommend Space Rangers 2).
Actually, the computer WOULD be predictable, exactly because it will always choose some "optimal" move.
No, it wouldn't. Read the article I linked to - it's a good introduction to the relevant game theory.
But, to simplify: an "unbeatable" strategy doesn't imply one move for each game state. Take rock/paper/scissors. The computer doesn't have to analyze your moves and try to guess what you're going to do. All it has to do is randomly pick a move and, over time, it will come out even. It's an unbeatable strategy.
Poker is more complex than RPS obviously, but the principle is the same. When looking at poker, instead of just one move, we have to consider an entire strategy as a move (ie. a complete list of reachable game states and an associated play for each of those states).
Given strategies for each player, there's an expected value of a hand over all possible cards. For each hand, we choose a strategy probabilistically such that our expected value across possible opposing strategies is maximized. Note that a given hand may have a negative expected value since one player goes first, etc - but overall the expected value over many games will be zero.
If you want to play around with this a bit, try the TopCoder Marathon Match problem "TwoCardDraw". It's a simplified version of poker, but it's nonetheless a challenging problem and an interesting coding strategy. It involves resolving the expected value of strategies, coming up with "perfect counter" strategies, and lots of other fun, relevant stuff. (I did OK - my submitted solution was a static strategy, but I experimented quite a bit before that).
And if you leak your source code, anyone who knows they're playing against your bot will have a huge advantage.
No. Knowing the source code of an ideal poker program would do you no good. You can't win against an agent following the Nash equilibrium for a game. He's going to be choosing from a library of strategies such that for every complete game, whatever strategy you choose, the expected value of the game is zero.
Naturally such an ideal program is hard to write, but conceptually it's very simple - poker is equivalent to any other zero sum game.
Pros may have an advantage against other humans (that they could read) but they wouldn't have any information advantage against an ideal computer player.
You have to understand: the ideal computer would not be predictable - it wouldn't always be choosing some "optimal" move in each situation. Rather, it would choose strategies probabilistically such that there is no strategy that would win against it over many hands.
Poker is complicated, but it can still be solved as a zero sum game treating every possible strategy as a single move.
Well hidden is a lot different to a human than an automated filter. Most people are going to have keyword searches that span newsgroups, so making your material findable to someone will still be as easy as putting a description in the subject line.
In terms of "they're watching", I'm sure these people are plenty concerned about anonymity anyways - and apparently to this point have found ways to post that hide their identity. What they're doing is already against the law, and there are already people trying to find them.
I'll grant perhaps removing some newsgroups might make the material more difficult to find for some users - but I also think it will put the material in front of more people who weren't looking for it.
No, there won't be more bandwidth. They'll still post it - it's just now instead of posting to alt.underage.porn (which won't exist) they'll put it in alt.porn (where people will be exposed to it that otherwise wouldn't have been).
And if they shut down alt.porn (because it's full of child porn now) they'll start putting it in the rec.ham.radio.
If you have a smelly toilet, you can't just say: "oh, well, we'll just board up that door so nobody poops there anymore". People are going to poop, and if there's no bathroom they're going to poop in the kitchen.
And the firewall analogy fails. A firewall makes it harder to attack my computer. Having no alt.underage.porn doesn't make it harder to post kiddy porn - it just makes it harder to avoid.
I use newsgroups quite a bit. Once alt.underage.porn (or whatever) is shut down, that material is just going to be posted somewhere else - and probably end up being seen by more people. If they ban keywords, they'll move onto new euphemisms. No automatic filter will do this job - and the results of the attempt will be worse in every way than if no filter was used.
All it is is scoring political points, and providing the illusion of action while really making the situation worse.
This was a joke.
Please don't mod as informative.
When you're identifying a bird, you might say that it matches the "hue and cry" of a certain species. Hue means color, and naturally cry refers to the noise the bird makes. That's where this term came from - and usually it's used in the same situations where you might say "Well, if it walks like a duck..."
It doesn't make sense for there to be a "hue and cry" as a reaction to something. Unless, perhaps, you imagine people are going to change color in anger.
But I'm guessing your interpretation is as though both words are verbs, like people are "hew"ing and crying in anger or something.
Are you sure the idea wasn't that it's often fast enough where it's not worth the effort to port to C++ just to find out?
Here's the part of the original post I was replying to:
Java surpassed C++ performance many years ago, and by such a wide margin that no one even bothers running benchmarks anymore
I suppose one could read that in a number of lights, but it seemed a bit off to me.
Where do you see this?
I posted one example, and it would easy for you to find others. For myself, I don't feel any particular obligation to further establish this.
As to the guy's benchmarks, they seem fairly reasonable to me. I don't think it's unreasonable to include things like vectors and hashmap implementations when talking about the speed of a language - and the benchmarks this guy was countering did the same thing.
You're also confused on my motivation, I think.
I don't hate Java. I was arguing against the idea that it's so much faster than C++ that nobody measures it anymore. Even if Java would outpace C++ if it was let to run for longer or something, we'd still be left with the conclusion that there are some tasks C++ is faster for (which is all I was trying to establish).
Do you want to disagree with that?
You're correct in that people don't generally choose C++ based on performance. But it does happen that non-C++ coders will switch for certain problems.
As an example, As here is Kawigi (a fairly high level competitor who usually uses Java), switching to C++ so he can brute-force a problem instead of using DP.
Notwithstanding that, all TopCoder problems (well, all correct ones) have a Java solution that runs in under 1 second - as that's the requirement on the reference solution. Where you'll see switches to C++ is to make other approaches feasible.
The test includes start-up times in the results.
This may be a significant factor for some of his smaller tests, but I don't think it jeopardizes the overall conclusion. For example, the Ackerman test was 15s vs. 60s. There isn't going to be more than a second of startup, and most of the tests are showing differences on the order of many seconds.
In terms of "is it harder to write fast code in C++", I guess it depends on what you're good at. Looking at the changes he did to the C++ code, none of it is esoteric - but I'm also more interested in this kind of thing than most of the programmers I know. I think it's fair to say many coders would produce faster code for these functions in Java than C++.
All in all, it's a subject that probably gets more attention than it deserves. Like you say, Java is plenty fast. Whenever I see performance problems at work, I never think "Ooh, let's rewrite this in a lower-level language" (and very seldom would that help). I think, let's look at the SQL, look at the bandwidth, look for other resource conflicts, or change overall approach.
Java is perfectly fast for real world applications, and I'd agree that the "Java is Slow" idea is outdated.
But it's not conclusively faster than C++, at least not in a general sense. In terms of a small task involving lots of simple operations, you'll still often see a significant speed increase using C++. This is a good example. Now I'm sure there's more optimizations available on both sides - and plenty of stuff to tweak - but C++ is going to come out ahead by a significant margin on a lot of these tasks.
A good example where the participants on both sides have some motivation is on TopCoder (where I spend a fair bit of time). Performance isn't usually the driving factor in language choice there - but sometimes it is, and when it is the answer is pretty much always C++ (unless it's a comparison between Java BigInteger and a naive implementation of the same in C++).
Reasonably often you'll see people write an initial solution in Java, find it runs a bit too slow, and quickly port it to C++ (or pre-emptively switch to C++ if they think they'll be near the time limit). It's not uncommon to see a factor of two difference in performance.
To be clear - these are not usually "real world" tasks. As more memory and objects come into play, Java is going to do better and better. But these kinds of tasks still exist - there's still plenty of places where C++ is going to be the choice because of performance.
In any case, your contention that Java is so much faster that nobody does benchmarks anymore is unsupported and wrong.
The control is a Forms style control. Obviously I can set properties on it at the server side. And I was using components on the control to communicate with the server - the problem was getting information from the control to the script (without an egregious amount of polling or ridiculousness).
.NET control, at which point it can signal to the remainder of the page successfully.
And I'm not "automatically assuming" there is security - this control runs only on one Intranet page where we have complete control of security and full trust.
Anyways - as I replied to the other guy, it does appear there was a somewhat more reasonable solution I missed. You can manually write COM (ActiveX) style events for your
It's funny, Googling for this now I am getting some better results (than when I implemented it a while back). It appears you can get events to work, but you have to write the ActiveX event code yourself (rather than relying on .NET events, which apparently IE can't translate).
I'm sure that's what you meant.
In any case, the current solution works fine (and was probably actually easier to implement) - but still it seems a bit silly they didn't make this work reasonably, or at least provide some hints on how to get it working.
One of our recent projects involved hosting a .NET control in Internet Explorer as part of an Intranet page for editing and uploading photos. The control worked fine, and we were able to manage security permissions such that it could access the files it needed to (and even delete them after as desired - it has full trust).
.NET control and the surrounding page (you can't, for example, just call a function on the page, despite the fact that you have full trust).
The problem was that we wanted the page to refresh after the upload was complete. This seems like it should be fairly simple, but with how the security works, there's not a simple way to communicate between the
With that being the case, I tried editing a property on the control once it was finished uploading, but that too didn't work as the page couldn't read them (even properties like "height" that it could infer couldn't be read directly). I thought about some fairly complicated solutions involving polling using XMLHTTP, partial refreshes or scripting or something - but what I ended up doing was this:
1. When the page loads, take note of the position of two elements on the page.
2. Poll the position of these elements a few times each second.
3. When the upload control finishes, it increases its height by 4 or 5 pixels.
4. This displaces the other elements on the page, and this can be used as a signal that an upload is complete and a refresh should be done.
This ended up working fine, but I felt ridiculous for having done it this way. There's no security being added here - the control and the page can communicate all they want via the server or scripting (or 100 other ways), it just means that using the controls is much more difficult and obscure than it needs to be.
I see what you're saying - but in practice I don't find it's much of a problem (the setup I mention is pretty much the same as I have set up now, which is why my I mention old ML models rather than whatever they have going now.. I'm also using an LCD until I finish the dedicated room in my new house).
I guess I'm not really much of an audiophile when it comes to movies. As long as you have sufficient power, a subwoofer, and surround, I don't notice too much else.
Where I do notice quality is in music - and having 2 good speakers alone has, to this point, been enough for that to sound great. I guess in the future we'll probably see more quality music with more channels, and I guess then I might upgrade the other speakers.
I wouldn't call this a home theater - and I don't think it's intended to be one. It's a technology showcase / bunch of useless crap. Besides that, how could you call it a "theater" with such a tiny screen?
Here's how I'd spend $18000:
1. Epson Powerlite 1080UB (projector) = $3000
2. Pair of Martin Logan Quest front speakers = $10000
3. Decent amp = $2000
4. Random center/rear channel speakers = $800
5. PS3 = $400
6. Decent 100" 16:9 screen = $500
7. Random subwoofer = $400
Now you're set up to watch movies, play games, listen to music, whatever - and your friends won't laugh at your pitiful 50" plasma.
And if you don't have $18000, substitute in a few cheaper alternatives and you can do a very decent theater for $3000 and still have a setup people will like more than the one in this article.