While I agree that I think WoW will start seeing active accounts in the US decline in the next year or two. I disagree with the reason. The questing in WoW is what they do great. While I have tons of complaints about the game, I'm sure I'll renew for the WotLK to play those 10 levels. The questing, exploration, and general leveling process is a lot of fun in WoW. My biggest complaint is that at max level the game goes from being a fun game of frequent character improvements to soul crushing long grind for even the smallest improvement. My 2nd complaint is that support characters (talented for teamplay) do not have a way to support themselves or significantly lag behind dps characters. (Some daily quests help, but they don't really balance the game, but I won't go into that long diatribe that I have.) My 3rd complaint is about grouping in general, it takes too long to get people together for group activities, and group activities are not dynamic for variable size groups or variable skill levels, the game is too ridged when it comes to grouping. The game creates too sharp a divide between people of different time commitments and skill levels.
Remote Control: Check
AutoPlay DVD/CD: Check
Large Screen: 19" monitor or S-Video to 36" TV (via Home Theater Receiver). Check. Check.
Couch: One in Living Room. One in Computer Room. Check. Check.
Ease of Use: While it takes more work to get all this setup and working it wasn't much extra effort than figuring all the wiring for a Home Theater System.
Cost: about the same. (Say three grand for a good gaming PC with monitor, software, joystick, and gamepad. $500 big screen TV, $300-$1000 for a good Receiver, $300-$1000 good speaker system, $50-200 DVD player, $50-100 CD player, $25-200 universal remote, $100-$300 for console system. Either system can be done cheaper depending on where you want to cut corners.)
I can do the same things at LAN events. Or on Voice Chat (Teamspeak2, Ventrilio, etc).
I have to agree with the parent poster. My PC is my multimedia center. I watch more DVDs on my computer and laptop than I do on my TV. I listen to more MP3/OGGs on them than other audio devices. I game on them. The only thing my TV has over my computers is that I don't have a capture card and PVR software on one of them (yet) and the laptop is more portable.
My only complaint about gaming on the PC is I haven't found a controller I like as much as the Nintendo or Sega controllers. (all the PC control pads I can find are modeled after the blasted PS2 controller.)
Watch for hacks. There are plenty of servers that allow cheats. Read the MOTD. Often servers will identify their exploit policies and installed mods. It won't take you long to realize if you've fallen victim to wall hacking or other exploits. Observe the other players during your dirt-nap to see if they're cheating.
Unless a person is blantently cheating a 5-30 second spectating of the person you aren't really going to be able to determine if they are cheating. With Punkbuster I've only seen a few I could say with certainty were cheating. (I've watched people for 30+ minutes sometimes)
Some tips for spectating players:
Good Sound System. Headphones or surround, with a decently loud volume. With the advent of directional sound what may seem like a wallhack to one person is often good ears. Maybe they heard you reload, maybe they heard you running, maybe they heard that gun fight you just won, etc.
Play with video settings. PB allows servers to limit certain settings to reasonable values, but this doesn't mean that people won't be using them within those limits to see things better. For example, turning up your brightness even a smidge may help you spot other players against difficult textures better. Also playing with these can help you achieve things like jumping further/higher/faster (quake), or spread reduction (HalfLife?)
Make a demo (RTFM). Ask someone else with more knowledge to watch and learn a little from them.
If nothing else you will learn a ton spectating people. Map secrets, tactics, etc.
I'm sure there are lots of more details on what to look for out there on the web, but I find these things will help in most cases.
Correct me if I'm wrong, but most of Netscape's initial employees were the students who worked on Mosaic. Not so much killing a free competitor, but one upping their previous effort. I think you read to much into "killer". This history is also why they continued to offer a free version.
The Practice of Programming by Brian W. Kernighan and Rob Pike. ISBN: 020161586X The Practice of Programming (reviewed sometime in the past by slashdot) is an excellent read, I hand it out to people the right crappy code.
When I said Because they knew this repetition existed they were able to create machines that could "Brute Force" those 6 letters. But they still had to look to see that they created a readable message. I really should have said that those machines they created tried every initial wheel setting and combinations of wheels till they found ones that matched the ciphertext. (I'm leaving off the plug-board settings because I don't remember off hand how they worked those out, but I think they found a way to negate their effect) Since their was a possiblity for multiple keys and multiple decryptions of the 3 letters, they took each guess at the 3 letters and decrypt the rest of the message using the 3 letters as the key. If the message isn't giberish they take the original key and try to decrypt other messages with it to get their 3 letter keys followed by an attempt to decrypt the rest of those message with their own 3 letter key.
As you can see this is a lengthy process and can understand why it took so many people at Bletchley park to do it. Even with the help of machines it was difficult. And as somebody else pointed out the keyspace for Enigma is extremely small compared to today's standards.
Most recent advances in cryptography have come from Diffie-Hellman and RSA type advances (called public key encryption or asymmetric ciphers). Which deal more with key exchange then the mangling of the message itself. Even DES and other newer symmetric ciphers are still basically substitution ciphers. Where one thing is substituted for another. The algorithms to do the substitutions have gotten much more complex and the keyspaces have gotten larger, but the basic principle has stayed the same. The fact that the mangle(encryption) and de-mangle(decryption) could be different algorithms was a huge advancement, but unfortunately as of yet these are very slow algorithms to implement and so usually RSA is used to exchange a DES (or some other symmetric ciphers key) to do the encryption of the actual message with.
Using the "Brute Force" method is an effective method of cryptanalysis. It just isn't a very good one. You will most likely get more than one possible solution and without some sort of crib you may not know which one is correct. This is what makes the One-Time Pad effective.
These Distributed.net contests aren't that stupid. They do show what is possible if you do have a crib. Initialy Enigma was broken by the Polish because they got a hold of the schematics and some german codebooks. They were able to find out that the first three letters were the same as the next three lettters and those were the rotator settings for the rest of the message. (the first enigma machines only had 3 wheels in use at one time) Because they knew this repetition existed they were able to create machines that could "Brute Force" those 6 letters. But they still had to look to see that they created a readable message. Turning took this information and many decrypted messages and figured out a way to decipher engima messages without this repetition because he knew the germans would eventually realize that it is a weekness. And they did! To crack these new messages Turning used cribs, guesses at what words would be where in the message. See Turing realized many messages the germans sent adhered to a strict standard format. IIRC especially some weather reports that were sent out at the same time everyday. Knowing the format of these messages and which messages (of the many messages that are sent during a war) turing was able to create a machine that could use these cribs to find the key to the message.
I haven't read Kahn's book yet, but I want to. I have read The Code Book by S. Singh.
From lurking on the tomcat-dev mailing list it sounds like this was an test pilot program. To see if OSS would work for them. It is good that they released NetBeans too. I guess they think the Jakarta project has done well. I think Sun is probably still evaluating this whole Open Source idea. So if you really want them to embrace it help with the Jakarta project or Netbeans and show them it is worth their time and a very good idea. (Hopefully this will eventually lead them to opening java up.)
Here is a more information about debugging java programs. It also has a list of applications that use the JDPA. Some of them being stand alone debuggers.
I tried one out last friday, JSwat. I thought it was ok, but still needs quite a bit of work and is more geared toward loading the class itself and debugging locally then attaching to a running VM. But it can do that.
I was just meaning to look up how to do this. I found this.
To summerize, if you can start the servlet engine's VM in debug mode you can then connect to it with a remote debugger like jdb. You have to get the password from wherever the VM's standard out goes. The page I listed up above gives directions on how to do this with the JSDK's Java Web Server.
As far as other developement. I've been using iPlanet's iWS 4.1 beta. It works pretty good. I haven't tried setting Tomcat up yet, but on the Tomcat mailing list they are getting together a beta for version 3.1 which is VERY close to meeting the Servlet 2.2 specs and will be the reference implementation once it is done. The beta release canidate 1 should be up today.
As somebody else mentioned standard out and the servlet response are also good places to send information especially if you don't want to run your VM in debug mode or can't. One thing about iWS 4.1, I don't think the standard out and standard error are directed to a file by default. I had to open my own servlet logging files and redirect System.out and System.err to them myself from within a servlet.
For developement I use JBuilder on NT, and vim occasionally under linux. I have had no platform problems with moving my Java class files from NT to linux. On linux I am using the latest blackdown release of the JDK. In fact for one application I have the web server (iWS 4.1 beta) running under linux and it loads pre-compiled servlets that exist on an SMB mounted directory that exists on an NT machine. It hasn't acted any different from the web server running on NT with JRun as the servlet engine. (well I had some JSP problems, but that was a difference in implementation of the JSP specs between iWS and JRun)
It isn't the importing of the PS2 that is illegal, it is the exporting it from Japan that is illegal. Sheesh. I'm starting to wonder if the key to getting a story posted on slashdot is to sensationallize the story and blow it out of proportion...
Just incase here it is again: Fortran in some instances can be the better language for your project depending on your requirements. Let me rephrase that: I don't believe in the "universal hammer".
You said: All the best languages are niche languages.
What like COBOL? Seriously, there are way to many variables that go into making a good language, deciding what the best tool for the job is, and what a person's favorite language is to argue about those. But I will argue that for most jobs the non-niche languages are the best fit. That is why they are not niche languages, they solve a wide variety of problems and do a good job at them.
Your idea of what "all the best languages" are may be completely different from somebody else's idea of "all the best languages". "Best" is largely subjective.
Don't worry about the dinosaur remarks. I developed on VMS systems (not the alpha version, but the old DEC mainframes) for a while and still did part-time until November. I get pointed out by other programmers saying "that guy used to program in Fortran on VMS". Being 25, that isn't very cool.
Yes there are products that are still being developed, but see my point in the last post, what is coming out that is new? I'm sure thare are a few things, but how much? Give it up they have been replaced as the choice for new development by workstations. So with your experience if you were given the task of developing a new system with no hardware requirements (the bank/client/whatever doesn't care what it runs on and they don't have any current hardware) are you honestly going to develop on a mainframe? The mainframe developer pool is shrinking, the hardware is hard to find, the clients won't get those pretty pictures they love so much, and you're going to have to support the hardware too because they aren't going to hire a competent mainframe sysadmin if they can even find one.
Sure my bank account probably is on an IBM mainframe, but that machine nor its programs are probably very new. Maybe the software had a new version come out, but that isn't a new product.
I will admit I could be wrong, maybe there is a big underground mainframe movement, so prove me wrong then and tell me about all these newly developed comercial products for mainframes. I only ever hear about people trying to replace them.
Maybe fortran wasn't the best example, but it pretty much is a niche language now. Fortran in some instances can be the better language for your project depending on your requirements. But what does Motif and CDE have for them besides being the current standards? Do they do anything better? Are they available on more platforms? Do they have any niche at all that will keep them alive as good choices for any type of new development?
I honestly don't know the answers to those questions, but the article implies that they don't have much going for them and they are not being used for any new development. At least none that the author could find.
One of the things that I've noticed about linux and GNU software seemingly "pushing things out of the way" is that generally, it happens in a spot where a company isn't *too* afraid of giving ground.
The article doesn't say it is replacing exisiting Motif applications at the moment, but that there is no NEW applications being written. This isn't to say if a company is already entrenched in Motif or CDE they won't add another program that fits into their world. But probably that program either isn't a commercial product or is only a piece of a larger system that already exists.
Just like FORTRAN programs and mainframes still exist so will Motif and CDE for quite some time. But are any new commercial products being developed using FORTRAN or on a mainframe? I can't think of any. C and workstations have replaced them respectively.
There are lots of idiots out there who want to sell you substandard components and switch your ordered part for something 'just as good'. Always send it back. I almost exclusively order parts from websites (the only exception is when the price listed at pricewatch.com doesn't match the website price). I usually order something simple first, just to test the site out, then order components. Sure, it takes longer to do things this way, but you save hundreds of dollars.
From (cr|h)ackers getting your servlets to do bad things: Java can't execute system commands from my knowledge unless you do some serious work to call native methods. Perl you have to check your eval() and other such calls, but I believe this has been automated for you.
From users putting in naughty servlets: From my understanding servlets run in the servlet sandbox(the servlet engine). Much like applets have a sandbox(your browser) that lets you can set what they are allow to do, you are supposed to be able to do this with servlets also. But I haven't actually seen this put into action. While glossing through a Java Security book I noticed this, but didn't read into depth about it. Last I checked there isn't a way to do something similar with perl.
Encryption/Decryption: Java's JCA and JCE is pretty cool. I've been learning this lately and it is fairly nice. There are many implementations of JCEs that provide a variety of algorithms. Also with the JCA and javax.crypto package you can switch out your Provider without changing your code, mearly a one line change to the file jre/lib/security/java.security. As long as the new Provider has an implementation of the algorithm you specified you're golden. I imagine perl has a wealth of crypto modules but I've never used any of them.
I like the idea that I built it myself. I know my machines inside and out. I know every component. I hand pick each component at my choosen price/quality ratio for my reasons. It is my hobby and I enjoy researching the parts, finding the bargins, and putting it all together. I'm very proud when I do a good job, and have noone to blaim but myself when I do a bad job. I like that.
I don't think it is any cheeper to build one yourself if you select all the same components. The big companies are buying at bulk prices. But you can do a pretty good job. And if you already have parts that you can reuse than you are saving money and getting the most out of previous purchases. You can also spread your purchases around to reduce the hit to your wallet without going into debt. Or you can say "I'll buy that part when it gets down to X dollars."
Plus I don't think the MBs in pre-builts are as overclocker friendly. If your into that sort of thing. It is also easier to make sure all your parts are Linux/BSD/(other alt OS)-friendly.
They have added a ton of new projects under them. From the Jakarta Project (Tomcat and Watchdog) to ECS. They are creating a lot of good tools if you want to build a web-based app with something more advanced than perl.
Slashdot Mirror
Hi there.
While I agree that I think WoW will start seeing active accounts in the US decline in the next year or two. I disagree with the reason. The questing in WoW is what they do great. While I have tons of complaints about the game, I'm sure I'll renew for the WotLK to play those 10 levels. The questing, exploration, and general leveling process is a lot of fun in WoW. My biggest complaint is that at max level the game goes from being a fun game of frequent character improvements to soul crushing long grind for even the smallest improvement. My 2nd complaint is that support characters (talented for teamplay) do not have a way to support themselves or significantly lag behind dps characters. (Some daily quests help, but they don't really balance the game, but I won't go into that long diatribe that I have.) My 3rd complaint is about grouping in general, it takes too long to get people together for group activities, and group activities are not dynamic for variable size groups or variable skill levels, the game is too ridged when it comes to grouping. The game creates too sharp a divide between people of different time commitments and skill levels.
I've had console games that crash, lockup, or have other bugs.
AutoPlay DVD/CD: Check
Large Screen: 19" monitor or S-Video to 36" TV (via Home Theater Receiver). Check. Check.
Couch: One in Living Room. One in Computer Room. Check. Check.
Ease of Use: While it takes more work to get all this setup and working it wasn't much extra effort than figuring all the wiring for a Home Theater System.
Cost: about the same. (Say three grand for a good gaming PC with monitor, software, joystick, and gamepad. $500 big screen TV, $300-$1000 for a good Receiver, $300-$1000 good speaker system, $50-200 DVD player, $50-100 CD player, $25-200 universal remote, $100-$300 for console system. Either system can be done cheaper depending on where you want to cut corners.)
I have to agree with the parent poster. My PC is my multimedia center. I watch more DVDs on my computer and laptop than I do on my TV. I listen to more MP3/OGGs on them than other audio devices. I game on them. The only thing my TV has over my computers is that I don't have a capture card and PVR software on one of them (yet) and the laptop is more portable.
My only complaint about gaming on the PC is I haven't found a controller I like as much as the Nintendo or Sega controllers. (all the PC control pads I can find are modeled after the blasted PS2 controller.)
Unless a person is blantently cheating a 5-30 second spectating of the person you aren't really going to be able to determine if they are cheating. With Punkbuster I've only seen a few I could say with certainty were cheating. (I've watched people for 30+ minutes sometimes)
Some tips for spectating players:
- Good Sound System. Headphones or surround, with a decently loud volume. With the advent of directional sound what may seem like a wallhack to one person is often good ears. Maybe they heard you reload, maybe they heard you running, maybe they heard that gun fight you just won, etc.
- Play with video settings. PB allows servers to limit certain settings to reasonable values, but this doesn't mean that people won't be using them within those limits to see things better. For example, turning up your brightness even a smidge may help you spot other players against difficult textures better. Also playing with these can help you achieve things like jumping further/higher/faster (quake), or spread reduction (HalfLife?)
- Make a demo (RTFM). Ask someone else with more knowledge to watch and learn a little from them.
- If nothing else you will learn a ton spectating people. Map secrets, tactics, etc.
I'm sure there are lots of more details on what to look for out there on the web, but I find these things will help in most cases.Correct me if I'm wrong, but most of Netscape's initial employees were the students who worked on Mosaic. Not so much killing a free competitor, but one upping their previous effort. I think you read to much into "killer". This history is also why they continued to offer a free version.
The Practice of Programming by Brian W. Kernighan and Rob Pike. ISBN: 020161586X The Practice of Programming (reviewed sometime in the past by slashdot) is an excellent read, I hand it out to people the right crappy code.
--
Here is a picture of the SNESp since you seem to have such a hard time finding it.
--
As you can see this is a lengthy process and can understand why it took so many people at Bletchley park to do it. Even with the help of machines it was difficult. And as somebody else pointed out the keyspace for Enigma is extremely small compared to today's standards.
Most recent advances in cryptography have come from Diffie-Hellman and RSA type advances (called public key encryption or asymmetric ciphers). Which deal more with key exchange then the mangling of the message itself. Even DES and other newer symmetric ciphers are still basically substitution ciphers. Where one thing is substituted for another. The algorithms to do the substitutions have gotten much more complex and the keyspaces have gotten larger, but the basic principle has stayed the same. The fact that the mangle(encryption) and de-mangle(decryption) could be different algorithms was a huge advancement, but unfortunately as of yet these are very slow algorithms to implement and so usually RSA is used to exchange a DES (or some other symmetric ciphers key) to do the encryption of the actual message with.
--
These Distributed.net contests aren't that stupid. They do show what is possible if you do have a crib. Initialy Enigma was broken by the Polish because they got a hold of the schematics and some german codebooks. They were able to find out that the first three letters were the same as the next three lettters and those were the rotator settings for the rest of the message. (the first enigma machines only had 3 wheels in use at one time) Because they knew this repetition existed they were able to create machines that could "Brute Force" those 6 letters. But they still had to look to see that they created a readable message. Turning took this information and many decrypted messages and figured out a way to decipher engima messages without this repetition because he knew the germans would eventually realize that it is a weekness. And they did! To crack these new messages Turning used cribs, guesses at what words would be where in the message. See Turing realized many messages the germans sent adhered to a strict standard format. IIRC especially some weather reports that were sent out at the same time everyday. Knowing the format of these messages and which messages (of the many messages that are sent during a war) turing was able to create a machine that could use these cribs to find the key to the message.
I haven't read Kahn's book yet, but I want to. I have read The Code Book by S. Singh.
--
--
I tried one out last friday, JSwat. I thought it was ok, but still needs quite a bit of work and is more geared toward loading the class itself and debugging locally then attaching to a running VM. But it can do that.
--
To summerize, if you can start the servlet engine's VM in debug mode you can then connect to it with a remote debugger like jdb. You have to get the password from wherever the VM's standard out goes. The page I listed up above gives directions on how to do this with the JSDK's Java Web Server.
As far as other developement. I've been using iPlanet's iWS 4.1 beta. It works pretty good. I haven't tried setting Tomcat up yet, but on the Tomcat mailing list they are getting together a beta for version 3.1 which is VERY close to meeting the Servlet 2.2 specs and will be the reference implementation once it is done. The beta release canidate 1 should be up today.
As somebody else mentioned standard out and the servlet response are also good places to send information especially if you don't want to run your VM in debug mode or can't. One thing about iWS 4.1, I don't think the standard out and standard error are directed to a file by default. I had to open my own servlet logging files and redirect System.out and System.err to them myself from within a servlet.
For developement I use JBuilder on NT, and vim occasionally under linux. I have had no platform problems with moving my Java class files from NT to linux. On linux I am using the latest blackdown release of the JDK. In fact for one application I have the web server (iWS 4.1 beta) running under linux and it loads pre-compiled servlets that exist on an SMB mounted directory that exists on an NT machine. It hasn't acted any different from the web server running on NT with JRun as the servlet engine. (well I had some JSP problems, but that was a difference in implementation of the JSP specs between iWS and JRun)
--
--
You said: All the best languages are niche languages.
What like COBOL? Seriously, there are way to many variables that go into making a good language, deciding what the best tool for the job is, and what a person's favorite language is to argue about those. But I will argue that for most jobs the non-niche languages are the best fit. That is why they are not niche languages, they solve a wide variety of problems and do a good job at them.
Your idea of what "all the best languages" are may be completely different from somebody else's idea of "all the best languages". "Best" is largely subjective.
--
Yes there are products that are still being developed, but see my point in the last post, what is coming out that is new? I'm sure thare are a few things, but how much? Give it up they have been replaced as the choice for new development by workstations. So with your experience if you were given the task of developing a new system with no hardware requirements (the bank/client/whatever doesn't care what it runs on and they don't have any current hardware) are you honestly going to develop on a mainframe? The mainframe developer pool is shrinking, the hardware is hard to find, the clients won't get those pretty pictures they love so much, and you're going to have to support the hardware too because they aren't going to hire a competent mainframe sysadmin if they can even find one.
Sure my bank account probably is on an IBM mainframe, but that machine nor its programs are probably very new. Maybe the software had a new version come out, but that isn't a new product.
I will admit I could be wrong, maybe there is a big underground mainframe movement, so prove me wrong then and tell me about all these newly developed comercial products for mainframes. I only ever hear about people trying to replace them.
--
I honestly don't know the answers to those questions, but the article implies that they don't have much going for them and they are not being used for any new development. At least none that the author could find.
--
The article doesn't say it is replacing exisiting Motif applications at the moment, but that there is no NEW applications being written. This isn't to say if a company is already entrenched in Motif or CDE they won't add another program that fits into their world. But probably that program either isn't a commercial product or is only a piece of a larger system that already exists.
Just like FORTRAN programs and mainframes still exist so will Motif and CDE for quite some time. But are any new commercial products being developed using FORTRAN or on a mainframe? I can't think of any. C and workstations have replaced them respectively.
--
The print press won, but plastics (woohoo!) is mentioned.
--
- There are lots of idiots out there who want to sell you substandard components and switch your ordered part for something 'just as good'. Always send it back. I almost exclusively order parts from websites (the only exception is when the price listed at pricewatch.com doesn't match the website price). I usually order something simple first, just to test the site out, then order components. Sure, it takes longer to do things this way, but you save hundreds of dollars.
Also check out Reseller Rattings and read posted review.--
From (cr|h)ackers getting your servlets to do bad things: Java can't execute system commands from my knowledge unless you do some serious work to call native methods. Perl you have to check your eval() and other such calls, but I believe this has been automated for you.
From users putting in naughty servlets: From my understanding servlets run in the servlet sandbox(the servlet engine). Much like applets have a sandbox(your browser) that lets you can set what they are allow to do, you are supposed to be able to do this with servlets also. But I haven't actually seen this put into action. While glossing through a Java Security book I noticed this, but didn't read into depth about it. Last I checked there isn't a way to do something similar with perl.
Encryption/Decryption: Java's JCA and JCE is pretty cool. I've been learning this lately and it is fairly nice. There are many implementations of JCEs that provide a variety of algorithms. Also with the JCA and javax.crypto package you can switch out your Provider without changing your code, mearly a one line change to the file jre/lib/security/java.security. As long as the new Provider has an implementation of the algorithm you specified you're golden. I imagine perl has a wealth of crypto modules but I've never used any of them.
--
I don't think it is any cheeper to build one yourself if you select all the same components. The big companies are buying at bulk prices. But you can do a pretty good job. And if you already have parts that you can reuse than you are saving money and getting the most out of previous purchases. You can also spread your purchases around to reduce the hit to your wallet without going into debt. Or you can say "I'll buy that part when it gets down to X dollars."
Plus I don't think the MBs in pre-builts are as overclocker friendly. If your into that sort of thing. It is also easier to make sure all your parts are Linux/BSD/(other alt OS)-friendly.
--
--