Yes, MSIE has a really bad support for PNG, but if you don't need animation, MSIE can do everything with PNG it can do with GIF. That is, PNG must be 8-bit version with only one color totally transparent. Yes, you loose transludency and true colors and if you use some b0rken software like Photoshop to produce those PNGs the resulting filesize will be larger than with GIF. I repeat: if you don't need animations, PNG can do everything the GIF can do even with MSIE. Other browsers can do 24bit colors with 8 bit transludency with gamma correction, though.
In addition, you you can hack some support to MSIE: just use some javascript combined with "behavior" CSS attribute. Can you see the irony of using non-standard feature to fix non-standard behaviour? I have yet to have any luck with this hack combined with absolute positioning, so that isn't perfect. And as far as I know, one cannot use transcludent PNG as a background with MSIE, with hacks or not.
I'm wonderfully happy to live in a world where the only large-scale communication network is prone to mass disruption and/or destruction at the drop of a pin.
I think that should have been moderated as "Sad" instead of "Funny"... However, the issue here isn't that one can easily cause mass distruption to the world wide web, but one can easily ask "other people" to automatically execute a piece of code that causes some distruption. The "other people" refers to users who use b0rken software.
It's like your car had a "feature" which caused it to go on full throttle and flash lights continuously in case somebody flashed lights at them. You'd see nationwide transportation systems go haywire very fast after someone flashing lights once. In this case the problem wouldn't be that the highways were baddly designed but majority of the users of those had malfunctioning devices (cars). It's the same thing here - except that software is much harder to do right than cars and practically any piece of software can send information to other computers at will. Firewalls cannot do a thing when the traffic is routed through HTTP and personal firewalls cannot do a thing if the worm asks the browser to send the information for it (browsers usually have access to other computers through the firewall).
For example, why do Ethernet cards allow promiscuous mode? It makes diagnosing certain problems easier, but it also represents a very big opportunity for all sorts of security vulnerabilities. Or why can MAC addresses be changed so easily?
I think it's a good thing that even so many NICs allow promiscuous mode and changing MAC address. If they didn't some lame-ass designer would think those are security features and rely on them while designing a new security model. Have you noticed how some very old protocols rely on IP numbers for "authentication" or "security". Perhaps when those protocols were designed the IP numbers were fixed and couldn't be changed easily? Or perhaps the security wasn't an issue back then when everybody had to trust everybody in the net to make it work, or something. Compare fixed MAC or IP to public key security and you'll immediately notice that those two are from totally different ballparks when it comes to the security. For example, the RSA is even in theory hard to crack to the best of our knowledge -- fixed MAC or IP is "hard" to crack because the "hardware doesn't allow you to change it". In cases like these, it's always the blackhats who have have the hardware that allows those "impossible changes" and whitehats just think they're safe.
You cannot trust on obscure hardware if you want security - it just quarantees that every issue is much harder to notice in advance, not that those issues didn't exist. (See also: DeCCS)
I have an admitted tendency towards speeding, ~10 miles an hour over the posted limit in town and cruising at 80 on the interstate are the norm for me.
Regardless of how good your vehicle handling is, speeding in town is never safe. You know the little thing called reaction time? No matter how much you drive, the experience never helps to get the reaction time much lower and it's the reaction time that kills the pedestrians in town. Real world reaction time including the time you notice the reason for breaking, moving your feet to start breaking until the wheels actually start the breaking is roughly one second [1]. If the limit is 45 km/h (~30 mph) and you run 60 km/h (~40 mph) the reaction time gets your car forward 16.7 m instead of 12.5 m. So you have 4.2 meters (~13.7 feet) less space to break on. Breaking to stop takes about 15 meters from the speed of 60 km/h and considering that the movement energy is squared when the speed is doubled, breaking from the allowed speed of 45 km/h takes about 7.5 meters - increase both numbers for a SUV. So taking the car from 60 km/h to zero takes 31.7 m versus 20 m from 45 km/h to zero. If a child runs to the lane 25 m before you the difference between hitting him or stopping safely many meters before him depends on whether you were speeding or not [2]. Saying it another way, staying under the legal limit allows you to completely stop in the same position where has breaking has continued for only 3.3 meters (~10 feet) in case you were speeding - and I'm pretty sure your car doesn't go from 60 km/h to zero in 3.3 meters. Considering that the remaining few meters would be part of effective breaking, the speed of the vehicle during the hit is still somewhere near 30 km/h (~20 mph). Think about bicycling quite fast and then hitting a concrete wall...
I agree that taking the raw reading from the box doesn't say if you're a safe driver or not but it does say if you're guilty after not being able to stop the vehicle before the crash.
[1] Don't believe me? Do an experiment: ask your friend to shout "break!" some time during the driving (of course, first making sure that no-one is immediatly behind you) and starting the clock. Ask him to stop the clock once he notices you've started the breaking. Sure, if you know the test is taking the place you can get great reaction times (much less than one second) but ask your friend to delay the test for a couple of hours while you're adjusting the car stereo or something and see how good your reaction time is then.
[2] In addition, if you weren't speeding, the kid could have a change to run over the whole lane even without you even breaking. If you're speeding, the kid might not made it.
PS. Speed limit in most Finnish towns and cities is 40 km/h, changed from the previous 50 km/h for exactly the reason that this way the pedestrians have much better survival rate.
Mensa membership apparently doesn't require you to be smart enough to know that you shouldn't brag about your IQ in a forum with ~ half a million readers.
I would not brag about anything in which you can gain membership by taking an extended matchbook quiz.
I'm pretty sure that he was suggesting that if a forum has half a million readers, the changes are pretty high that you're not the brightest person in the whole forum. Especially when the forum is something like slashdot where the average member probably has IQ much higher than 100.
Or perhaps your post was a simple flamebait or you tried to be funny. Posting AC didn't help even a little bit.
If IBM's lawyers have any balls whatsoever, they will countersue for anticompetitive practices.
So the big picture is: SCO sues IBM without any real evidence -> SCO looses -> IBM countersues SCO for anticompetitive practices and lost revenue -> IBM wins -> SCO cannot pay -> IBM owns SCO -> the problem is gone.
In my car, the audio system sits inside an elliptic box, thus making the resale value of it minimal.
You've misunderstood. Ford using non-standard audio source has nothing to do with preventing theifs but with maximizing their profit margins. Do you think they could ask such prices for "their" players if those were standard and the car buyer could purchase other one from another manufacturer? You're right about the resale value being minimal but I wouldn't count that as a plus.
Users always blindly open zipfiles and double-click (cringe) the files found in side to open them,
despite for years Microsoft telling you that double-clicking on executable files is not the way to install programs.
Hmm... where Microsoft or any other party has claimed that you should install Windows software some other way but running executables? Are you trying to hint that the Windows autorun mechanism is somehow more secure? Yes, some MS extensions use.msi packages but just check how one installs, for example, Microsoft Internet Explorer -- not only is the user required to execute setup binary but the user is expected to feel comfortable while to install binary downloads and runs binaries from the internet.
Most Linux users think twice before running some packages disguised as shell scripts that only decompress the package. Most Windows users just double click everything and see what happens.
When I want to test-drive a new car, do I build a driving simulator based on the car and load in my route to work so I can see how it would handle during a typical commute? Of course not, I just take the car for a test drive, because sometimes the best way to test something is just to use it.
Sure, and that's how hard drive benchmarks are done. With newest graphics cards it's the other way around: you have the car but you have no road to run on with it.
How do you test a train that requires new kind of track? You build a test track and try to guess how the train performs in real use after testing it on that test track. It's the same thing here.
The DualShock2 is the best controller out there for just about everything else. It's comfortable, it has intelligently placed buttons, it has enough buttons, but not *too* many
It might be the best, but it's far from good. I personally prefer DualShock2 analog sticks over other products but DualShock2 has really bad buttons when compared to competitors. Have you ever tried triggers under the XBox controller? You just hate to use DualShock2 controller ever again.
One thing I especially hate with game console controllers is that those're designed to be used with thumbs only. Hello? See, I've four other fingers in both hands, why not use those? (Yep, some games support triggers but still using all 10 buttons in the DualShock controller is real pain in the ass.)
DualShock2 (and any controller that comes with any game console) is something that allows you to play all games but it's not perferct for anything. Take driving games for example: the only driving wheel that works even moderately well with PS2 is Logitech's newer model. AFAIK, there's not a single one usable force feedback wheel controller for XBox. Nevermind the fact that the MS FF Wheel for PC is about the best wheel there is. And this is the situation even though driving games is one of the biggest console gaming genres.
And all the first person shooters... anybody that claims that console controller is more suitable for that genre of games than keyboard and mouse should go online with some game that allows plaing against PC players and get their feet wet.
The problem with all the consoles is lowest common dominator: if something doesn't come with every console it's supported by practically no game. How many games support, for example, more than 2 players on PS2 even though the console can support 8?
I did have a laptop in college. You'll never use it. Really. Professors tend to talk in a highly non-linear manner, go back and correct themselves, make heavy use of diagrams, generally lecture in a manner not friendly to taking notes on a laptop.
I agree. If you're trying to make notes, laptops get you nowhere. I'd say, get a digital camera and a small tripod (one that can sit on a table) instead. It'll be money much better spent. Remember to get a couple of extra battery packs and lots of memory cards.
As for myself, I make notes on paper but I never read those notes. I've noticed that if I write things down while I'm listening the lecture I can remember things much better - in fact, so well that I don't need to re-read those notes. I usually read some related papers instead.
Some time ago I would have agreed with you, but not anymore, If media player crashes playing some video then the whole system becomes unstable and then even doing something like sending a file to the recyclebin freezes the UI
As I wrote in another thread, there's a difference between the whole system and single pieces of software. That is, it doesn't matter if the computer doesn't crash if the piece of software that the user is using crashes. If you're running your desktop on X and X crashes it doesn't make you happy that the system is running as you just lost all the apps you had open. If Windows Explorer is practically a single threaded process that blocks until a file is fully moved to recycle bin or a cd-rom drive has successfully read the TOC after inserting the disk then the whole UI seems halted or crashed to the user[1]. Crashing some game running on X so that keyboard doesn't work is exact equivalent to system crash to casual user -- not all of us know, or have enabled, Alt+SysRq+R hack.
My kernel (be it linux or w2k) hasn't crashed for a long time but my apps are crashing every now and then. Crashing a single app isn't that bad as crashing the whole system but it's still far from perfect. The problem is, it's quite expensive to make all software as reliable as our current OS kernels are.
[1] At least, with windows 2000, this keeps happening all the time.
Debian has 8,710 packages available to do anything a comercial comercial software does, mostly better. Not just one or two pieces of it, every piece. My systems never crash under their stable release and I run all sorts of services.
Did you mean that the system didn't crash, or that not a single process crashed? I guess it's the former. Unfortunately, it isn't usually enough. If any of my processes has a bug that causes that process to crash then part of my work is lost -- in best case it might be just some lost time. If you're running your desktop on X and X crashes then it's pretty much equivalent to system crash from your point of perspective even though the system is still running. The same thing if your server process needs database access and your database process goes to deadlock.
Writing software is complex and if everything must work then everything must be perfect[1]. Think about writing a whole series of detective stories with references back and forth where every little detail mathes perfectly with all the other details. Writing perfect software isn't that far from proving complex math formulas other than that software is usually much more complex as a whole.
[1] Sometimes it's possible to write redundant code where first run fails and the process automatically tries something else. In that case the code isn't perfect but the software works perfectly. One might say that the system is perfect but it has non-optimal code because it does some unneeded work (it should have automatically executed the failover code immediately instead of trying the default one first).
C++ [...] In that code were a couple of news, and I couldn't find the deletes which matched them. So I asked the original developer. Turns out the deletes were in a base class of the class that I was moving the code to.
Let's get this straight... you have a base class and a class inherited from it and the inherited class acquires (with new) some memory and trusts that the base class implementation frees (delete) that memory? It sounds that the code is seriously broken -- if the base class implementation is changed then the memory isn't deleted? IMO, if the inherited class acquires the memory with new, it must take care to delete that memory. If, on the other hand, it acquires the memory by calling some specific method then it should call matching method unless the aforementioned method guarantees to take care of freeing the memory. If you've broken code it really doesn't matter which programming language was used to write that code. Or if all the memory was designed to be freed by the base class then there's a bug in the documentation.
In C++ you have to take care of freeing memory. That is by design and if you can't handle it, you should use some other language. It's not like you don't have null pointer exceptions in Java code either. Those languages are a little different but in both cases you have to know what you're doing.
But more to the point, if one company changes things and names it correctly (so the game that comes out in 2k3 is called 2k3, not 2k4) then they are at a LOSS compared to the competition.
There's some truth in there. However, I think that if I released Extreme Football 2005 not that many people tough it were better than the 2004 offerings by other companies. Using next year's title makes your product look "from the future" instead of "current". Lying it by two years makes your product to look fake. I've no idea why it's so, but I think it has something to do with car and electronics advertisement where they always advertise next year's offerings (and the most important thing is the looks, not the functionality). Software companies don't advertise next year's offerings, instead they sell those (though, with current year's functionality and looks).
Most have seek bars that move smoothly under the mouse but aren't connected to the video while moving, or only seek to discrete points that are quite far apart. [...] VCRs have an excuse for being unresponsive, they are based on physical tape that must move.
You know, most video formats have so called key frames and it's only those key frames from which the player can really start the playing. If you select a position between keyframes the decoder must compute all frames from previous keyframe to the point you selected. And some formats require the decoder to compute all frames after the selected frame and next keyframe, too! When some people encode movies with 10 minute keyframe interval [1] you can be sure that no player can quickly skip to any given location.
There's your physical excuse for the lack of displaying video live while you move that track bar -- there's no way to compute the results fast enough. Yep, most players could do better than they do but if the underlaying decoder engine is designed with the playback in mind, it might be that it doesn't do backwards seek that fast. Once you have backend engine that can do smooth seeking , coming up with a nice frontend is a non-issue. Or, it might be hard to decide if those buttons should have "silver" or "grey" finish...
[1] Say you have 24fps movie with the resolution of 720x360 pixels. With 10 minute keyframe interval the decoding engine needs to compute 10*60*24/2 = 7200 frames per seek on average. Considering that uncompressed frame takes about 0.5 MB, the player needs to handle about 7200 * 0.5 MB = 3600MB of data per seek... Of course, the situation isn't that bad usually but there're still lots of little bits to compute.
IBM PC XT 4.7 Megahertz to Pentium 4 at 3 Gigahertz. (3,000 Megahertz) It seems a little shy of 10,000 times unless you factor going from an 8 bit processor to a 32 bit processer.
You don't need to go that far back to history to see a really big difference. Just compare the FPU speed of i287 and Athlon. i287 took minimum of 90 cycles for FMUL, minimum of 70 cycles for FADD and at least 30 cycles for a floating point load. Compare that to Athlon that can do two loads, FMUL and FADD every cycle. So, something that took i287 at least 90+70+2*30 = 220 cycles, Athlon can do every clock cycle. In addition to that, Athlon is running at 2GHz instead of 10MHz. So one could argue that current Athlon is 2000/10*220 = 44000 times faster than about a 20 year old FPU (when was 287 released anyway?). In addition to that, we have MMX, SSE and SSE2 that can further boost best case scenarios but I think it's safe to say that current x86 CPUs are at least 10000 times faster than 20 year old ones. Not to count more advanced caches -- not too many years ago L2 cache was external and optional. Of course, if you compare 20 year old Gray and a CPU inside modern portable device the difference is much smaller.
A 2.2 Ghz P4 (I assume) is not 88 times faster than a 25 MHz 386/486. You're buying into the megahertz myth; the modern processor has many more optimizations that make it more faste.
I wrote "we can modify SMTP servers to return '555 Advertisements not allowed'" and you wrote
How would it save bandwidth? The Subject: header is considered content, and follows the DATA SMTP command.
It doesn't save bandwidth between the spammer and initial access point but it would save bandwidth between that mail server and everything else (mail usually goes through more than one server, you know). In addition, spammers would quite quickly get a hint and they would need to rethink their strategy. Perhaps they could narrow their target audience so much they can take the hit from the bounty. As a result the number of spams would decrease heavily[1] even though there would be some spam still. Or perhaps they would simply change to some honest work.
[1] In this case they would probably use valid headers because otherwise the bounty would be too expensive. If not, we would need to readjust the forged headers punishment.
The link you provided contains firmware for
"DVD-116, DVD-106S and DVD-A06S drives." Unfortunately for me, I've DVD-115 which is locked to 16x. Perhaps I should go digging to web in case somebody has hacked firmware to work with 115 too...
Page even says: "Note. works with Pioneer 6 series models only." Why is that? Why isn't any other model supported? I wish manufacturers would release source code to firmware too so I could fix the issue by myself when the manufacturer is just too lazy.
Firmware upgrades can add support for it in the future.
Well, perhaps they can, but it's up to the manufacturer. For example, I have a Pioneer DVD drive that wants to read all DVDs with 16x speed. Needless to say, it's a loud drive to watch movies with. I complained to Pioneer and asked if they could release a firmware patch to slow down the drive. They replied something along the lines "it would be impossible because the hardware couldn't support it". I consider that reply as bullshit as the drive already slows down if there're scratches on the disk so they are just too lazy to support older drives. So, if there isn't firmware patch available already, I wouldn't count on it being available in the future either. YMMV.
I wrote "The problem, as I see it, isn't the protocol per se, but the fact that we trust in every step on the path from the sender to receiver." and you replied:
The Verisign way you describe sounds like what I'm talking about, although I don't know about the specific implementation details (could you provide me with a link or two where I could read up more?)
Sorry, but just come up with the idea as I was reading the discussion so I don't have any links. I don't think that the idea of requiring that all messages are signed with a key verified by a known CA is going to fly. We have already PGP (and GPG) so people can send digitally encrypted or signed messages without extra costs (not counting CPU time) and still not that many even know how to make it work. Considering all unsigned mail as a spam is much too coarse filter rule. Counting all signed mail as legimate is 100% sure rule so you can use that for filtering already (you don't even need any specific CA, any one will do just fine for now). Just tell on your web page, or wherever you distribute your email address, that the sender should digitally sign the message. If you require that all messages are signed, you also quarantee that you receive messages from other nerds. Tell the visitor that the mail is quaranteed to get to your inbox if it's signed.
IMHO, the problem isn't the filtering part but to make all users to sign their messages. As we have seen, people do not upgrade their mail software to block security rules that allow the attacker to take full control of the system -- how do you think they would upgrade to digitally sign messages? And to pay for identification by some well-known CA if this catches on?
As for current system wasting many man years with dealing with spam. Yes, that's unfortunate but I really believe that the problem with a new system isn't that designing and coding the software would be expensive but that getting the rest of the world to get it way too expensive.
If we can get users to sign (or even encrypt) all their messages, the spam problem goes away. Perhaps we should campaign to get GPG integrated with Mozilla? Then we could just tell regular users to use Mozilla for their mail...
If all you want is to get rid of spam, simply auto-bounce all messages without approximate value of PI. Just ask the sender to include that in the message body to be able to send you mail. You can be sure not to receive any spam.
a better idea, compared to imposing taxes on email, would be to create a new infrastructure for exchanging of "email"
The problem, as I see it, isn't the protocol per se, but the fact that we trust in every step on the path from the sender to receiver. You can already enforce a rule to receive only mail signed with a key verified by VeriSign (or some other expensive CA). Because the key is expensive and contains some information about the sender one could easily block all the senders one doesn't want to hear about. And the spammer needs to be able to supply fake ID to a CA you trust to fake his sender information after you've blocked him once. In addition, digitally signing all messages requires much more computing power so sending enormous amounts of spam would be more expensive in that way too.
The "received" headers in mail would be enough to trace spammers even today if one could trust in those. And the reason we cannot trust in those is that some ISPs are run by spammers and they forge "received" headers in otherwise legimate mail servers (the worst case, usually spammers just find open relays). If you trust in ISPs in general, you're trusting spammers too.
The problem isn't how we can make the thing work but how we can make it work without speding more than we do today (that is, practically nothing).
Pretend to be the slowest talking person in the world...
Say "beep" each time the telemarketer says some often repeated word (like "the"). If the telemarketer asks you what's going on, say "nothing, why?" (Variation: say "me" every time the telemarketer says "you")
If they want to loan you money, tell them you just filed for bankruptcy and you could sure use some money.
Reply with a real husky voice: "What are you wearing?"
Say "Could you repeat that?" Repeat.
"I'm sorry, but I consider insurance a form of gambling, and my religion forbids it."
Slashdot Mirror
In addition, you you can hack some support to MSIE: just use some javascript combined with "behavior" CSS attribute. Can you see the irony of using non-standard feature to fix non-standard behaviour? I have yet to have any luck with this hack combined with absolute positioning, so that isn't perfect. And as far as I know, one cannot use transcludent PNG as a background with MSIE, with hacks or not.
I think that should have been moderated as "Sad" instead of "Funny"... However, the issue here isn't that one can easily cause mass distruption to the world wide web, but one can easily ask "other people" to automatically execute a piece of code that causes some distruption. The "other people" refers to users who use b0rken software.
It's like your car had a "feature" which caused it to go on full throttle and flash lights continuously in case somebody flashed lights at them. You'd see nationwide transportation systems go haywire very fast after someone flashing lights once. In this case the problem wouldn't be that the highways were baddly designed but majority of the users of those had malfunctioning devices (cars). It's the same thing here - except that software is much harder to do right than cars and practically any piece of software can send information to other computers at will. Firewalls cannot do a thing when the traffic is routed through HTTP and personal firewalls cannot do a thing if the worm asks the browser to send the information for it (browsers usually have access to other computers through the firewall).
I think it's a good thing that even so many NICs allow promiscuous mode and changing MAC address. If they didn't some lame-ass designer would think those are security features and rely on them while designing a new security model. Have you noticed how some very old protocols rely on IP numbers for "authentication" or "security". Perhaps when those protocols were designed the IP numbers were fixed and couldn't be changed easily? Or perhaps the security wasn't an issue back then when everybody had to trust everybody in the net to make it work, or something. Compare fixed MAC or IP to public key security and you'll immediately notice that those two are from totally different ballparks when it comes to the security. For example, the RSA is even in theory hard to crack to the best of our knowledge -- fixed MAC or IP is "hard" to crack because the "hardware doesn't allow you to change it". In cases like these, it's always the blackhats who have have the hardware that allows those "impossible changes" and whitehats just think they're safe.
You cannot trust on obscure hardware if you want security - it just quarantees that every issue is much harder to notice in advance, not that those issues didn't exist. (See also: DeCCS)
Regardless of how good your vehicle handling is, speeding in town is never safe. You know the little thing called reaction time? No matter how much you drive, the experience never helps to get the reaction time much lower and it's the reaction time that kills the pedestrians in town. Real world reaction time including the time you notice the reason for breaking, moving your feet to start breaking until the wheels actually start the breaking is roughly one second [1]. If the limit is 45 km/h (~30 mph) and you run 60 km/h (~40 mph) the reaction time gets your car forward 16.7 m instead of 12.5 m. So you have 4.2 meters (~13.7 feet) less space to break on. Breaking to stop takes about 15 meters from the speed of 60 km/h and considering that the movement energy is squared when the speed is doubled, breaking from the allowed speed of 45 km/h takes about 7.5 meters - increase both numbers for a SUV. So taking the car from 60 km/h to zero takes 31.7 m versus 20 m from 45 km/h to zero. If a child runs to the lane 25 m before you the difference between hitting him or stopping safely many meters before him depends on whether you were speeding or not [2]. Saying it another way, staying under the legal limit allows you to completely stop in the same position where has breaking has continued for only 3.3 meters (~10 feet) in case you were speeding - and I'm pretty sure your car doesn't go from 60 km/h to zero in 3.3 meters. Considering that the remaining few meters would be part of effective breaking, the speed of the vehicle during the hit is still somewhere near 30 km/h (~20 mph). Think about bicycling quite fast and then hitting a concrete wall...
I agree that taking the raw reading from the box doesn't say if you're a safe driver or not but it does say if you're guilty after not being able to stop the vehicle before the crash.
[1] Don't believe me? Do an experiment: ask your friend to shout "break!" some time during the driving (of course, first making sure that no-one is immediatly behind you) and starting the clock. Ask him to stop the clock once he notices you've started the breaking. Sure, if you know the test is taking the place you can get great reaction times (much less than one second) but ask your friend to delay the test for a couple of hours while you're adjusting the car stereo or something and see how good your reaction time is then.
[2] In addition, if you weren't speeding, the kid could have a change to run over the whole lane even without you even breaking. If you're speeding, the kid might not made it.
PS. Speed limit in most Finnish towns and cities is 40 km/h, changed from the previous 50 km/h for exactly the reason that this way the pedestrians have much better survival rate.
I would not brag about anything in which you can gain membership by taking an extended matchbook quiz.
I'm pretty sure that he was suggesting that if a forum has half a million readers, the changes are pretty high that you're not the brightest person in the whole forum. Especially when the forum is something like slashdot where the average member probably has IQ much higher than 100.
Or perhaps your post was a simple flamebait or you tried to be funny. Posting AC didn't help even a little bit.
So the big picture is: SCO sues IBM without any real evidence -> SCO looses -> IBM countersues SCO for anticompetitive practices and lost revenue -> IBM wins -> SCO cannot pay -> IBM owns SCO -> the problem is gone.
You've misunderstood. Ford using non-standard audio source has nothing to do with preventing theifs but with maximizing their profit margins. Do you think they could ask such prices for "their" players if those were standard and the car buyer could purchase other one from another manufacturer? You're right about the resale value being minimal but I wouldn't count that as a plus.
Hmm... where Microsoft or any other party has claimed that you should install Windows software some other way but running executables? Are you trying to hint that the Windows autorun mechanism is somehow more secure? Yes, some MS extensions use .msi packages but just check how one installs, for example, Microsoft Internet Explorer -- not only is the user required to execute setup binary but the user is expected to feel comfortable while to install binary downloads and runs binaries from the internet.
Most Linux users think twice before running some packages disguised as shell scripts that only decompress the package. Most Windows users just double click everything and see what happens.
Sure, and that's how hard drive benchmarks are done. With newest graphics cards it's the other way around: you have the car but you have no road to run on with it.
How do you test a train that requires new kind of track? You build a test track and try to guess how the train performs in real use after testing it on that test track. It's the same thing here.
It might be the best, but it's far from good. I personally prefer DualShock2 analog sticks over other products but DualShock2 has really bad buttons when compared to competitors. Have you ever tried triggers under the XBox controller? You just hate to use DualShock2 controller ever again.
One thing I especially hate with game console controllers is that those're designed to be used with thumbs only. Hello? See, I've four other fingers in both hands, why not use those? (Yep, some games support triggers but still using all 10 buttons in the DualShock controller is real pain in the ass.)
DualShock2 (and any controller that comes with any game console) is something that allows you to play all games but it's not perferct for anything. Take driving games for example: the only driving wheel that works even moderately well with PS2 is Logitech's newer model. AFAIK, there's not a single one usable force feedback wheel controller for XBox. Nevermind the fact that the MS FF Wheel for PC is about the best wheel there is. And this is the situation even though driving games is one of the biggest console gaming genres.
And all the first person shooters... anybody that claims that console controller is more suitable for that genre of games than keyboard and mouse should go online with some game that allows plaing against PC players and get their feet wet.
The problem with all the consoles is lowest common dominator: if something doesn't come with every console it's supported by practically no game. How many games support, for example, more than 2 players on PS2 even though the console can support 8?
I agree. If you're trying to make notes, laptops get you nowhere. I'd say, get a digital camera and a small tripod (one that can sit on a table) instead. It'll be money much better spent. Remember to get a couple of extra battery packs and lots of memory cards.
As for myself, I make notes on paper but I never read those notes. I've noticed that if I write things down while I'm listening the lecture I can remember things much better - in fact, so well that I don't need to re-read those notes. I usually read some related papers instead.
Actually, when I was learning Pascal, I wrote a simple 3D modeling software with it. When I was learning C++ and STL, I wrote a simple SQL engine.
I'm still working on that text editor written in C...
As I wrote in another thread, there's a difference between the whole system and single pieces of software. That is, it doesn't matter if the computer doesn't crash if the piece of software that the user is using crashes. If you're running your desktop on X and X crashes it doesn't make you happy that the system is running as you just lost all the apps you had open. If Windows Explorer is practically a single threaded process that blocks until a file is fully moved to recycle bin or a cd-rom drive has successfully read the TOC after inserting the disk then the whole UI seems halted or crashed to the user[1]. Crashing some game running on X so that keyboard doesn't work is exact equivalent to system crash to casual user -- not all of us know, or have enabled, Alt+SysRq+R hack.
My kernel (be it linux or w2k) hasn't crashed for a long time but my apps are crashing every now and then. Crashing a single app isn't that bad as crashing the whole system but it's still far from perfect. The problem is, it's quite expensive to make all software as reliable as our current OS kernels are.
[1] At least, with windows 2000, this keeps happening all the time.
Did you mean that the system didn't crash, or that not a single process crashed? I guess it's the former. Unfortunately, it isn't usually enough. If any of my processes has a bug that causes that process to crash then part of my work is lost -- in best case it might be just some lost time. If you're running your desktop on X and X crashes then it's pretty much equivalent to system crash from your point of perspective even though the system is still running. The same thing if your server process needs database access and your database process goes to deadlock.
Writing software is complex and if everything must work then everything must be perfect[1]. Think about writing a whole series of detective stories with references back and forth where every little detail mathes perfectly with all the other details. Writing perfect software isn't that far from proving complex math formulas other than that software is usually much more complex as a whole.
[1] Sometimes it's possible to write redundant code where first run fails and the process automatically tries something else. In that case the code isn't perfect but the software works perfectly. One might say that the system is perfect but it has non-optimal code because it does some unneeded work (it should have automatically executed the failover code immediately instead of trying the default one first).
Let's get this straight... you have a base class and a class inherited from it and the inherited class acquires (with new) some memory and trusts that the base class implementation frees (delete) that memory? It sounds that the code is seriously broken -- if the base class implementation is changed then the memory isn't deleted? IMO, if the inherited class acquires the memory with new, it must take care to delete that memory. If, on the other hand, it acquires the memory by calling some specific method then it should call matching method unless the aforementioned method guarantees to take care of freeing the memory. If you've broken code it really doesn't matter which programming language was used to write that code. Or if all the memory was designed to be freed by the base class then there's a bug in the documentation.
In C++ you have to take care of freeing memory. That is by design and if you can't handle it, you should use some other language. It's not like you don't have null pointer exceptions in Java code either. Those languages are a little different but in both cases you have to know what you're doing.
There's some truth in there. However, I think that if I released Extreme Football 2005 not that many people tough it were better than the 2004 offerings by other companies. Using next year's title makes your product look "from the future" instead of "current". Lying it by two years makes your product to look fake. I've no idea why it's so, but I think it has something to do with car and electronics advertisement where they always advertise next year's offerings (and the most important thing is the looks, not the functionality). Software companies don't advertise next year's offerings, instead they sell those (though, with current year's functionality and looks).
You know, most video formats have so called key frames and it's only those key frames from which the player can really start the playing. If you select a position between keyframes the decoder must compute all frames from previous keyframe to the point you selected. And some formats require the decoder to compute all frames after the selected frame and next keyframe, too! When some people encode movies with 10 minute keyframe interval [1] you can be sure that no player can quickly skip to any given location.
There's your physical excuse for the lack of displaying video live while you move that track bar -- there's no way to compute the results fast enough. Yep, most players could do better than they do but if the underlaying decoder engine is designed with the playback in mind, it might be that it doesn't do backwards seek that fast. Once you have backend engine that can do smooth seeking , coming up with a nice frontend is a non-issue. Or, it might be hard to decide if those buttons should have "silver" or "grey" finish...
[1] Say you have 24fps movie with the resolution of 720x360 pixels. With 10 minute keyframe interval the decoding engine needs to compute 10*60*24/2 = 7200 frames per seek on average. Considering that uncompressed frame takes about 0.5 MB, the player needs to handle about 7200 * 0.5 MB = 3600MB of data per seek... Of course, the situation isn't that bad usually but there're still lots of little bits to compute.
You don't need to go that far back to history to see a really big difference. Just compare the FPU speed of i287 and Athlon. i287 took minimum of 90 cycles for FMUL, minimum of 70 cycles for FADD and at least 30 cycles for a floating point load. Compare that to Athlon that can do two loads, FMUL and FADD every cycle. So, something that took i287 at least 90+70+2*30 = 220 cycles, Athlon can do every clock cycle. In addition to that, Athlon is running at 2GHz instead of 10MHz. So one could argue that current Athlon is 2000/10*220 = 44000 times faster than about a 20 year old FPU (when was 287 released anyway?). In addition to that, we have MMX, SSE and SSE2 that can further boost best case scenarios but I think it's safe to say that current x86 CPUs are at least 10000 times faster than 20 year old ones. Not to count more advanced caches -- not too many years ago L2 cache was external and optional. Of course, if you compare 20 year old Gray and a CPU inside modern portable device the difference is much smaller.
Yep, current CPUs do much more in one clock than 386. For example, 386 took 9-38 clocks for a single 32b multiply. Even P3 can do effectively (pipeline) a single 32b multiply in 1 clock and floating point multiply in 2 clocks (or better if you use 'multimedia' instructions). Athlon can do both in 1 clock (in fact, it can do two loads, add and multiplication with floating point numbers in one clock). I couldn't find timings for P4, but I think it can do integer multiplication in one clock but requires quite more for floating point multiplication.
It doesn't save bandwidth between the spammer and initial access point but it would save bandwidth between that mail server and everything else (mail usually goes through more than one server, you know). In addition, spammers would quite quickly get a hint and they would need to rethink their strategy. Perhaps they could narrow their target audience so much they can take the hit from the bounty. As a result the number of spams would decrease heavily[1] even though there would be some spam still. Or perhaps they would simply change to some honest work.
[1] In this case they would probably use valid headers because otherwise the bounty would be too expensive. If not, we would need to readjust the forged headers punishment.
Page even says: "Note. works with Pioneer 6 series models only." Why is that? Why isn't any other model supported? I wish manufacturers would release source code to firmware too so I could fix the issue by myself when the manufacturer is just too lazy.
Well, perhaps they can, but it's up to the manufacturer. For example, I have a Pioneer DVD drive that wants to read all DVDs with 16x speed. Needless to say, it's a loud drive to watch movies with. I complained to Pioneer and asked if they could release a firmware patch to slow down the drive. They replied something along the lines "it would be impossible because the hardware couldn't support it". I consider that reply as bullshit as the drive already slows down if there're scratches on the disk so they are just too lazy to support older drives. So, if there isn't firmware patch available already, I wouldn't count on it being available in the future either. YMMV.
Sorry, but just come up with the idea as I was reading the discussion so I don't have any links. I don't think that the idea of requiring that all messages are signed with a key verified by a known CA is going to fly. We have already PGP (and GPG) so people can send digitally encrypted or signed messages without extra costs (not counting CPU time) and still not that many even know how to make it work. Considering all unsigned mail as a spam is much too coarse filter rule. Counting all signed mail as legimate is 100% sure rule so you can use that for filtering already (you don't even need any specific CA, any one will do just fine for now). Just tell on your web page, or wherever you distribute your email address, that the sender should digitally sign the message. If you require that all messages are signed, you also quarantee that you receive messages from other nerds. Tell the visitor that the mail is quaranteed to get to your inbox if it's signed.
IMHO, the problem isn't the filtering part but to make all users to sign their messages. As we have seen, people do not upgrade their mail software to block security rules that allow the attacker to take full control of the system -- how do you think they would upgrade to digitally sign messages? And to pay for identification by some well-known CA if this catches on?
As for current system wasting many man years with dealing with spam. Yes, that's unfortunate but I really believe that the problem with a new system isn't that designing and coding the software would be expensive but that getting the rest of the world to get it way too expensive.
If we can get users to sign (or even encrypt) all their messages, the spam problem goes away. Perhaps we should campaign to get GPG integrated with Mozilla? Then we could just tell regular users to use Mozilla for their mail...
If all you want is to get rid of spam, simply auto-bounce all messages without approximate value of PI. Just ask the sender to include that in the message body to be able to send you mail. You can be sure not to receive any spam.
The problem, as I see it, isn't the protocol per se, but the fact that we trust in every step on the path from the sender to receiver. You can already enforce a rule to receive only mail signed with a key verified by VeriSign (or some other expensive CA). Because the key is expensive and contains some information about the sender one could easily block all the senders one doesn't want to hear about. And the spammer needs to be able to supply fake ID to a CA you trust to fake his sender information after you've blocked him once. In addition, digitally signing all messages requires much more computing power so sending enormous amounts of spam would be more expensive in that way too.
The "received" headers in mail would be enough to trace spammers even today if one could trust in those. And the reason we cannot trust in those is that some ISPs are run by spammers and they forge "received" headers in otherwise legimate mail servers (the worst case, usually spammers just find open relays). If you trust in ISPs in general, you're trusting spammers too.
The problem isn't how we can make the thing work but how we can make it work without speding more than we do today (that is, practically nothing).
You'll be on their black list in no time.