Now, armed with all those definitions of all the unfamiliar terms in that paragraph, complete with links to the terms used in the definitions (which are themselves complete with links to all the terms used in the definitions of the definitions, ad nauseum), you've got all you need to understand those two paragraphs! Isn't the Internet great?
The problem is that user interface design is not something that can be done by a geek sitting alone in his/her room coding. Even so-called user-interface experts can't fix a user interface by themselves. The one and only key to designing a good user interface is USER TESTING. This means finding other people who have never used your software and observing them as they learn how to use it. It really is crucial to get actual people to use your software while you watch. Without user testing, your user interface will be crap no matter how many self-proclaimed experts pontificate on the merits of your various design choices. 10 minutes of user testing is worth days of speculation about how to make your interface better. However, nobody writing open-source software does usability testing; they make their GUIs by themselves based soley on their preconcieved notions of what a GUI should be like. No matter how well-intentioned the developer is, this process won't produce easy-to-use software, and it won't produce new innovations in user interface design. I am convinced that this is the reason open-source software interfaces suck and are mostly copycats of other software.
Another big problem with UI design in general is that when things go wrong, there is a tendancy to blame the user instead of the software. "You should have clicked this other button" or "You should have seen this option" or the ever-popular "You should have read the manual". This attitude is not restricted to open-source software developers; you see it everywhere. The fact of the matter is, if one person makes a stupid mistake, it's quite likely that other people will too, so you should account for it in your design no matter how stupid the mistake seems. If you want a good user interface, you have to make sure that even the stupidest mistakes people make are accounted for in your design. The attitude you need to have to design a good UI is: _every_ mistake a user makes is entirely the fault of the interface, because a truly good interface would either eliminate the possibility of making a mistake or at least be smart enough to indicate that you're making a mistake. Obviously it's not possible to meet this ideal, but a lot can be done to eliminate most mistakes users make. On-the-fly spell/grammar checking is a good example of this philosophy.
That's exactly what it is. It's "we're going to keep doing all the things we've already been doing, but we're going to call them XNA now." XNA is not a product, an API, a hardware specification, or anything tangible at all. It's just a brand name for MS marketing to slap on all their gaming-related stuff to make it sound cooler. The only real news items in these two articles were the plans to bring XBox live to PC games and the introduction of a single controller for XBoxes and PCs.
I doubt a card that was too long would get past the hardware, and if it did, any extra data would probably be thrown away by the firmware before it got to the software. Of course there is a possibility that this kind of attack would work, it's just not very probable.
That idea is rediculous. Blogs and personal homepages are the richest source of the quality links that make PageRank work. Links on commercial sites are the worst kind becuase the linkers have a vested interest in affecting PageRank. Commercial sites can and do exploit the heck out of PageRank. OTOH, most of the time blogs link exclusively to stuff that's useful, cool or interesting to their owners (and therefore in aggregate all web surfers). The only two search terms I know that have been deliberately affected by bloggers are "miserable failure" and "litigious bastards", and only a tiny percentage of bloggers have participated in those efforts. I could easily point to a million searches unfarily influenced by commercial sites through exploitation of PageRank.
Says who? Did you read the contract? I thought not. It would be perfectly legal for the contract to require them to use the product, and that appears to be what happened here. The story says the lawsuit revolves not around whether or not MLB is obligated to use Real, but whether spring training is a part of the "baseball season" as defined in the contract. There is apparently no question that the contract does obligate them to use Real to broadcast the season.
His site explains that the games don't, in fact, run faster. Most Genesis games must actually be based on a clock instead of the processor speed. The only effect of the overclocking is that slowdown is eliminated. Don't you remember in Sonic games how if you had more than 20 or so rings and you got hit, the Genesis would slow to a crawl as it drew all the rings bouncing around on the screen? In two-player mode slowdown was even more common. Well if you overclock your Genesis, that can apparently be fixed.
In Debian, the Bash package comes with a totally awesome collection of customized tab completions. For some reason, they are not turned on by default. To turn them on in a single account, you can put the line "source/etc/bash_completion" in your ~/.bashrc file, or you can turn them on globally by editing the/etc/bash.bashrc file and uncommenting the relevant lines. You'll get magic smart tab completion for cd, apt-get, ssh, mplayer, and bajillions of other programs, and you'll wonder how you ever did without it. apt-get tab completion in particular rocks like nothing else. For example, if you type "apt-get remove x[TAB]" you'll get a complete list of installed packages starting with x. When installing, you'll get a list of available but not yet installed packages. I can't stand using apt-get without tab completion anymore.
This is a good thing. If you encrypt without authentication, the encryption is questionably useful at best because the connection is always vulnerable to a man-in-the-middle attack. At worst, providing this kind of mode that is known to be attackable could cause big problems if people assume it's secure (it's encrypted, of course it's secure, right? Wrong).
This is why SSH always asks you to verify the host computer's public key the first time you login and stores it for later connections. If it didn't, an attacker could come in at any other time you connect, impersonate the server and get your password. With authentication, the attacker can't impersonate the server and man-in-the-middle fails.
Maybe I'm just not a big enough open-source fanboy, but IMHO a local root exploit in Solaris is a much bigger problem than a denial-of-service attack on FreeBSD. One problem can cause security breaches in important systems that are assumed to be secure, leaking sensitive information, passwords, etc, and the only way to be sure your multi-user system is still secure now is to wipe it and reinstall. The other problem can temporarily disable your webserver.
If we can have 1Mbps up/down to the Internet anywhere there's a power jack in the entire US, I personally think that's worth sacrificing a part of the wireless spectrum. If it stomps all over amateur radio operators, that's too bad, but universal broadband Internet access is much higher on my list of things that matter. If this technology can bring broadband to the places that aren't served by cable or DSL, then it's a great thing. If this alleged interference caused problems with anything that really mattered, the FCC would be all over it. You guys are whining over nothing.
Better yet, for about the price of a Roomba, you can get an ER1 with a camera and robotics-oriented software. An ER1 would make a much better platform for a tablet-PC based robot, since that's basically what it and its software are designed for.
There are already several of those that work exactly as you describe. The reality is that they just don't work well for games, because games always have some special requirements for features or optimizations that those kinds of "environments" don't provide. For example, you couldn't make Prince of Persia with one of those, because of the "rewind time" ability (unless that was already built-in, but I doubt it would be...). I doubt it would be possible to make a game with networked multiplayer with one of those, because decent networked multiplayer requires gazillions of special-purpose optimizations that depend on the requirements of the game itself. It would be extremely hard to design a plugin API powerful enough to allow you to add these kinds of things after-the-fact. Plus, the amount of effort involved in even writing the plugins for these kinds of things would probably put one-man operations out of the picture. Yes, one-man shops making games could still use these environments, but their games still wouldn't compare to big game houses.
Actually, newbies are the ones who need documentation the least. As has been said before, if your user interface needs documentation it is too confusing. On the other hand, experts who are doing something unusual or complex need documentation on how to do those things. Documentation written for newbies is useless because newbies can't find it and don't want to spend time reading it, while experts won't get any benefit from it.
You don't have to right-click on the document to get the standard utility menus any more. Gimp 1.3 has the menus at the top of the document window by default.
Ofoto isn't exactly what I mean by a web hosting service. Ofoto doesn't let you view the original images you uploaded, only much lower-resolution previews. It doesn't let you link your album from other sites, and people have to give their email address to Ofoto before they can even view your pictures (in low-resolution preview versions). You certainly can't use it as a web hosting service to post pictures on your blog or whatever. Basically, it's a lame attempt to make people buy prints from them at $.29 each.
Kodak should offer a real image hosting service, where you can actually link to your full-resolution pictures or embed them in other sites, with a monthly fee. They could differentiate themselves by being extremely easy to use and integrated with their camera software. When you buy a Kodak digital camera you should get a year of this hosting service free. If they did that, I might actually recommend it to my parents.
I think Kodak is doing amazingly well for a company whose main product is on the way to becoming obsolete. They saw the digital photography revolution before it actually happened, and they took preemtive steps to transition their brand name into the new market. They realize that technology is fundamentally changing their entire market, and they are attempting to adapt instead of being dragged kicking and screaming into bankruptcy by the inexorable forward march of technology.
They could have done it better, of course. Right now they are focusing on using digital cameras exactly like film cameras: making prints and organizing photos into albums for storage. Digital photography can be so much more. They should be focusing on the things that can be done better with digital photography: photo editing and distribution. They should offer a web hosting service for individual pictures or complete albums, and their camera software should come with extensive photo editing capabilities. (also it shouldn't suck quite so much). But there's a lot of inertia in a company like Kodak, and it's amazing that they've been able to adapt to changing technology as much as they have. Certainly better than some companies in other industries I could name...
The little-known DCOP can do the same for KDE. All KDE apps have a DCOP function interface which can be used for remote control from another C++ app, a shell script, a python/perl script, or any other kind of script. For example, I recently saw this perl hack that attatches a Konsole terminal emulator to the top of your screen. It does it all using DCOP.
Yeah. These icons are nothing compared to the work of Everaldo, KDE's chief icon guy. Plus, in the next KDE release KSVG may mature enough to render all the icons from the source SVG instead of using bitmap copies. This means fully resizable icons even better than OS X's.
82 characters of pure C goodness right there. You may wonder why I didn't have this in my sig all along. Well, last time I tried it, the lameness filter rejected it (yes, apparently the lameness filter rejects sigs too, or at least it used to). Now it seems to work; go figure. So now my sig will be the improved version (which is short enough to also allow <tt> tags and a <br> for nice formatting). I guess I should probably update my journal with this stuff too; I kinda forgot about it:-) After a while Slashdot disables posting in old journal topics, which kinda sucks. I'd rather post about this there than be offtopic here, but oh well.
I can't take full credit for this little gem. Another slashdotter was the first to suggest using ?: in this effective way. I can't find his post now though, so I can't credit him. I'll admit your trick of not giving the type of r or c is new to me. Also I didn't realize it would compile without including anything; that's pretty sweet. I wonder if there are any other shortcuts in C syntax that could be used to shorten this up even more? Under 80 chars would be pretty awesome.
Agreed. Even with Bittorrent this film isn't worth the time it takes to download. It has bad timing, bad acting, bad special effects, and bad jokes. Go watch Red vs. Blue instead.
Slashdot Mirror
Differential Operator
Vector Bundle
Fredholm Operator
Cokernel
Now, armed with all those definitions of all the unfamiliar terms in that paragraph, complete with links to the terms used in the definitions (which are themselves complete with links to all the terms used in the definitions of the definitions, ad nauseum), you've got all you need to understand those two paragraphs! Isn't the Internet great?
Another big problem with UI design in general is that when things go wrong, there is a tendancy to blame the user instead of the software. "You should have clicked this other button" or "You should have seen this option" or the ever-popular "You should have read the manual". This attitude is not restricted to open-source software developers; you see it everywhere. The fact of the matter is, if one person makes a stupid mistake, it's quite likely that other people will too, so you should account for it in your design no matter how stupid the mistake seems. If you want a good user interface, you have to make sure that even the stupidest mistakes people make are accounted for in your design. The attitude you need to have to design a good UI is: _every_ mistake a user makes is entirely the fault of the interface, because a truly good interface would either eliminate the possibility of making a mistake or at least be smart enough to indicate that you're making a mistake. Obviously it's not possible to meet this ideal, but a lot can be done to eliminate most mistakes users make. On-the-fly spell/grammar checking is a good example of this philosophy.
That's exactly what it is. It's "we're going to keep doing all the things we've already been doing, but we're going to call them XNA now." XNA is not a product, an API, a hardware specification, or anything tangible at all. It's just a brand name for MS marketing to slap on all their gaming-related stuff to make it sound cooler. The only real news items in these two articles were the plans to bring XBox live to PC games and the introduction of a single controller for XBoxes and PCs.
To that I'd like to add that KStars is great planetarium program, and Celestia is just an awesome program in general.
I doubt a card that was too long would get past the hardware, and if it did, any extra data would probably be thrown away by the firmware before it got to the software. Of course there is a possibility that this kind of attack would work, it's just not very probable.
That idea is rediculous. Blogs and personal homepages are the richest source of the quality links that make PageRank work. Links on commercial sites are the worst kind becuase the linkers have a vested interest in affecting PageRank. Commercial sites can and do exploit the heck out of PageRank. OTOH, most of the time blogs link exclusively to stuff that's useful, cool or interesting to their owners (and therefore in aggregate all web surfers). The only two search terms I know that have been deliberately affected by bloggers are "miserable failure" and "litigious bastards", and only a tiny percentage of bloggers have participated in those efforts. I could easily point to a million searches unfarily influenced by commercial sites through exploitation of PageRank.
It would be hard; the amount of data that can be stored in a card's magnetic strip is very small. Format of magnetic strip data
Says who? Did you read the contract? I thought not. It would be perfectly legal for the contract to require them to use the product, and that appears to be what happened here. The story says the lawsuit revolves not around whether or not MLB is obligated to use Real, but whether spring training is a part of the "baseball season" as defined in the contract. There is apparently no question that the contract does obligate them to use Real to broadcast the season.
His site explains that the games don't, in fact, run faster. Most Genesis games must actually be based on a clock instead of the processor speed. The only effect of the overclocking is that slowdown is eliminated. Don't you remember in Sonic games how if you had more than 20 or so rings and you got hit, the Genesis would slow to a crawl as it drew all the rings bouncing around on the screen? In two-player mode slowdown was even more common. Well if you overclock your Genesis, that can apparently be fixed.
In Debian, the Bash package comes with a totally awesome collection of customized tab completions. For some reason, they are not turned on by default. To turn them on in a single account, you can put the line "source /etc/bash_completion" in your ~/.bashrc file, or you can turn them on globally by editing the /etc/bash.bashrc file and uncommenting the relevant lines. You'll get magic smart tab completion for cd, apt-get, ssh, mplayer, and bajillions of other programs, and you'll wonder how you ever did without it. apt-get tab completion in particular rocks like nothing else. For example, if you type "apt-get remove x[TAB]" you'll get a complete list of installed packages starting with x. When installing, you'll get a list of available but not yet installed packages. I can't stand using apt-get without tab completion anymore.
This is why SSH always asks you to verify the host computer's public key the first time you login and stores it for later connections. If it didn't, an attacker could come in at any other time you connect, impersonate the server and get your password. With authentication, the attacker can't impersonate the server and man-in-the-middle fails.
Which processors support NX? Does Linux support NX on these processors?
Maybe I'm just not a big enough open-source fanboy, but IMHO a local root exploit in Solaris is a much bigger problem than a denial-of-service attack on FreeBSD. One problem can cause security breaches in important systems that are assumed to be secure, leaking sensitive information, passwords, etc, and the only way to be sure your multi-user system is still secure now is to wipe it and reinstall. The other problem can temporarily disable your webserver.
If we can have 1Mbps up/down to the Internet anywhere there's a power jack in the entire US, I personally think that's worth sacrificing a part of the wireless spectrum. If it stomps all over amateur radio operators, that's too bad, but universal broadband Internet access is much higher on my list of things that matter. If this technology can bring broadband to the places that aren't served by cable or DSL, then it's a great thing. If this alleged interference caused problems with anything that really mattered, the FCC would be all over it. You guys are whining over nothing.
Well I hope they can get a tenth person to upgrade, I'll bet they need the money...
Better yet, for about the price of a Roomba, you can get an ER1 with a camera and robotics-oriented software. An ER1 would make a much better platform for a tablet-PC based robot, since that's basically what it and its software are designed for.
There are already several of those that work exactly as you describe. The reality is that they just don't work well for games, because games always have some special requirements for features or optimizations that those kinds of "environments" don't provide. For example, you couldn't make Prince of Persia with one of those, because of the "rewind time" ability (unless that was already built-in, but I doubt it would be...). I doubt it would be possible to make a game with networked multiplayer with one of those, because decent networked multiplayer requires gazillions of special-purpose optimizations that depend on the requirements of the game itself. It would be extremely hard to design a plugin API powerful enough to allow you to add these kinds of things after-the-fact. Plus, the amount of effort involved in even writing the plugins for these kinds of things would probably put one-man operations out of the picture. Yes, one-man shops making games could still use these environments, but their games still wouldn't compare to big game houses.
Actually, newbies are the ones who need documentation the least. As has been said before, if your user interface needs documentation it is too confusing. On the other hand, experts who are doing something unusual or complex need documentation on how to do those things. Documentation written for newbies is useless because newbies can't find it and don't want to spend time reading it, while experts won't get any benefit from it.
You don't have to right-click on the document to get the standard utility menus any more. Gimp 1.3 has the menus at the top of the document window by default.
Kodak should offer a real image hosting service, where you can actually link to your full-resolution pictures or embed them in other sites, with a monthly fee. They could differentiate themselves by being extremely easy to use and integrated with their camera software. When you buy a Kodak digital camera you should get a year of this hosting service free. If they did that, I might actually recommend it to my parents.
They could have done it better, of course. Right now they are focusing on using digital cameras exactly like film cameras: making prints and organizing photos into albums for storage. Digital photography can be so much more. They should be focusing on the things that can be done better with digital photography: photo editing and distribution. They should offer a web hosting service for individual pictures or complete albums, and their camera software should come with extensive photo editing capabilities. (also it shouldn't suck quite so much). But there's a lot of inertia in a company like Kodak, and it's amazing that they've been able to adapt to changing technology as much as they have. Certainly better than some companies in other industries I could name...
The little-known DCOP can do the same for KDE. All KDE apps have a DCOP function interface which can be used for remote control from another C++ app, a shell script, a python/perl script, or any other kind of script. For example, I recently saw this perl hack that attatches a Konsole terminal emulator to the top of your screen. It does it all using DCOP.
Yeah. These icons are nothing compared to the work of Everaldo, KDE's chief icon guy. Plus, in the next KDE release KSVG may mature enough to render all the icons from the source SVG instead of using bitmap copies. This means fully resizable icons even better than OS X's.
main(r,c){for(r=-1,c=0;r<39;c++)printf(c<0?" ":c>r?c=r++-38,"\n":~r&c?" `":" #");}
82 characters of pure C goodness right there. You may wonder why I didn't have this in my sig all along. Well, last time I tried it, the lameness filter rejected it (yes, apparently the lameness filter rejects sigs too, or at least it used to). Now it seems to work; go figure. So now my sig will be the improved version (which is short enough to also allow <tt> tags and a <br> for nice formatting). I guess I should probably update my journal with this stuff too; I kinda forgot about it :-) After a while Slashdot disables posting in old journal topics, which kinda sucks. I'd rather post about this there than be offtopic here, but oh well.
I can't take full credit for this little gem. Another slashdotter was the first to suggest using ?: in this effective way. I can't find his post now though, so I can't credit him. I'll admit your trick of not giving the type of r or c is new to me. Also I didn't realize it would compile without including anything; that's pretty sweet. I wonder if there are any other shortcuts in C syntax that could be used to shorten this up even more? Under 80 chars would be pretty awesome.
Agreed. Even with Bittorrent this film isn't worth the time it takes to download. It has bad timing, bad acting, bad special effects, and bad jokes. Go watch Red vs. Blue instead.