According to this article HTML is the best UI environment yet except for content creation.
Damn, and all this time I could've been playing Quake3 in HTML. Prolly would've had a much better UI. After all it's just a database interaction proggy.
And I didn't even know that HTML was great at controlling hardware. Dammit, all this time I could've been running my serial port controlled radio station in HTML. After all it's just a database interaction proggy.
And CRAP! My GPS navigation system in my car could be HTML pages!! I can't believe what I'm missing out on. Clicking pages, scrolling, and driving would so safe and fun. After all it's just a database interaction proggy.
etc...
And no, embeding some non-standard control in a HTML page does _NOT_ count. At that point you're opening a whole new can of worms, including the ability to use any of the crappy (not) controls the author mentions. It's arguable whether Javascript and friends count (definitely not Java applets).
It's too bad someone has not made the effort (AFAIK) to create the same type of integrated, stable, tight desktop environment on *BSD. On the server BSD is pretty good(*), but on the desktop it's a complete wash. It's worse than Linux because it's basically the same thing Linux has (GNOME, KDE, or whatever) except missing some really great software (nVidia lags behind with their BSD drivers and the only VMware you can run is 2.x).
(*) Only pretty good because I find BSD to be a big PITA when it comes time to upgrade the system. "Rebuild world" still requires a major effort to get the system in the correct state, not to mention the incredible time sink it is. Even just upgrading individual packages takes too much time and effort (same reason why I don't use Gentoo). A Debian apt-get upgrade or dist-upgrade is hella easier and fast.
I would really like to see a FreeBSD package system that uses dpkg to maintain everything (with the same or better quality put into it that FreeBSD normally sees) and a really good integrated desktop environment (At least as good as OS X or BeOS).
In any case, I'm a bit skeptical of the data. They seem to be monitoring the providers' own websites, not their clients' machines or sites.
That is totally true, and I'm skeptical also. They have Rackshack listed as NT/98, but in fact they offer servers of all sorts. My site is hosted on a RH Linux machine not Windows. Just because Rackshack's main site goes out doesn't mean any clients did (which I know for a fact has happened because once I couldn't get to the main site but my site worked fine).
Do you really need the operator syntax in every case?
I could rewrite the entire Linux kernel in 100% assembly language too. Doesn't mean I'd want to.
Operator overloading can be abused but when used correctly it goes a long ways towards making the code more readable. It allows constructs that are easy to code up and visualize. I mean, making the programmer's job easier is a good thing, no?
Think about this, in Java we don't need function overloading either. So why use it?
Hell, we don't need object-oriented stuff at all. Just use some flat language that doesn't support encapsulation, inheritance, function overloading, or... You don't need OOP to write software that does what you want.
If you did do that, it would still still work if you forced it into a value type: matrix matrix3 = matrix1 + matrix2;
I should qualify that and say it might work. Technically the variable is out of scope and matrix3 contains an invalid value. I'm not sure what the standard says about that though.
Everyone who keeps specifying the operator as operator+(matrix1, matrix2) is wrong. That's not the way you overload the + operator. This seems to indicate that many people on this thread are blasting away and have no idea how to program in C++. Anyway...
But what are you returning a reference to? For example (simplified):
matrix& operator+(const matrix& m) {
matrix temp_matrix;
temp_matrix[0] = m[0] + (*this)[0];
temp_matrix[1] = m[1] + (*this)[1];
etc...or use a for-loop, whatever...
return...What??? temp_matrix is the result }
You can't return a reference temp_matrix because it's about to go out of scope (a good compiler will warn you about this).
If you did do that, it would still still work if you forced it into a value type: matrix matrix3 = matrix1 + matrix2;
But at that point you are not using the reference. The reference would be copied by value into matrix3. Of course compilers these days can sometimes optimize that away, but not always.
If you could access the left side of the equation then you wouldn't need temp_matrix at all.
Take a look at java.awt.Graphics2D. Sound? The javax.sound package. Networking? java.net has everything you need.
Hehe... Sure, but those are not implemented in Java. It's only an interface to a C or C++ library. Java lacks the power to do it.
Yes, I freely admit that you can't write device drivers in Java, since that's about the only point you could possibly claim and actually be correct. For the 99.9999% of us that aren't actually writing device drivers, it's completely and utterly irrelevant. Use JNI if you really need something like that, and it's still easier than trying to write the whole program in C++.
Uh, but I'm not arguing how easy it is. I was just stating that Java is less powerful.
And it is most certainly not the only point where Java lacks power. Templates? (generics have less power) Multiple inheritance? What if I don't want garbage collection? The Java specification itself prevents Java programs from ever running as fast as the fastest C/C++ proggies. I could go on.
I write in Java too. I'm not arguing that Java isn't a good language or that it's not better than C++. I'm just saying it has less power.
I can understand where you're coming from. C++ is a complex beast. I've been using some form of C++ for over 10 years (well before it was standardised) and I still don't understand everything about it.
With that said, it's an extremely powerful and flexible language. Very much more powerful than Java or C#. The complexity is mostly due to its flexibility. You can do (almost) anything with it. Of course, we can argue whether that's good or bad.
I think C++ can learn from Java though. The default should be to pass all non-built-in-type function parameters by const reference and the programmer has to specify otherwise (basically opposite of the way it is now). That would clean up the code a whole lot since 99% of the time that's what you want anyway. And the standard C++ library should have some sort of garbage collector available.
Another problem I have with C++ is that even with all its power you have no way to get to the "left hand" variable of operations. For example, if you have a matrix class you can overload the "+" operator so that you can do things like "matrix3 = matrix1 + matrix2". However, that's not going to be very efficient (assuming that's why you're using C++ in the first place) because there is no way to get to the matrix3 variable from inside the + operator. That forces you to use a temporary variable to add the two matrices then copy by value the whole matrix after adding matrix1 and matrix2. There are tricks around this problem but none are clean.
Same here. However, while most of the mail I'm getting is directly sent and from DSL/cable accounts, none of the boxes have been Windows boxes.
I've examined some of the boxes (by either NMAP, SSH, or telneting into them) and there were a couple routers (Linksys or similar home routers) but many of the boxes are actually Linux.
This seems to suggest one of two things to me: Either Linux boxes are getting hacked, or the spammers are using (multiple?) DSL accounts and Linux to send out their spam (this seems more likely to me).
This is basically the philosophy of Extreme Programming.
And my answer to the "Quick 'n Dirty" or "Correct and Proper" qustion is to use some or all of the Extreme Programming practices. I had been using some of those XP techniques way before anyone decided to define what XP was (mostly from hands-on programming experience over the last 20 years) and have found it provides a great balance between a perfect design and something that "just works". Having something that works is way, way more important than having a perfect design.
For the most part, that's the software that runs everything right now. The software that works. It may not have a perfect design, but it works. XP helps mediate the "bad design" part of it.
Although I'd really like to see a driver that installs on something other than RedHat (how about a plain tar.gz maybe?). It shouldn't be that hard to repackage.
The driver works with kernel 2.4.18, so I don't know why so many people are complaining about only supporting old kernels. It should work with any system with that kernel (it might work with any 2.4 kernel), even newer and other distros.
I assume they recognized that there are tons of 3D movie, animation studios, and medical research labs that are all moving to Linux. 3DLabs provides some of the higher-end workstation-class graphics cards used for 3D work.
Which is kind of stupid really; I'd imagine computers with lower specs but increased stability, efficiency (wasting less power on warming the office) and lower costs would be popular in the corporate scene.
Except for the lower cost thing I'd swear you're talking about a Mac.
Personally, I want Doom3 to run at 1000 FPS just like I like my Quake3 (yes, I can tell the difference and it makes a difference in competition).
I want my 3D renderings to be completed instantly and I want my videos encoded in real time.
I don't want to ever give a damn about running out of disk space. I want everything stored in a never ending, version controled, file system (searchable and indexable of course, even with many items and a large version database).
Plus everything else I could do if I just had a computer that was faster and had more disk space.
The current hardware is too slow, the current drives are too small. We don't need slower stuff.
I remember poking around on PG not long ago but soon forgot about it.
If you're not looking for something specific then the site is kinda, meh. As you suggested, they need a news site, ratings, and other stats so you can see what's available.
And sections. "Technical", "Poetry", etc. Otherwise it's not very useful to the casual browser.
mmm, unless I am really missing something:) oopsie.
These are not LCD's. They are self illuminating and do not use backlighting. Also colorful. And from the look of the photo they do not appear to be translucent, but it's hard to tell.
If they could be made clear though... wow, it would look awesome. It would not be like a LCD because OLED's produce colorful light.
I wonder if they have a version that has or can make the pixels clear.
That would be cool. Imagine a large wall that has graphics on it but is also translucent. You know, like those screens you always see in sci-fi movies. Add some touch sensors...
I think they're up to 177, but your point still stands... 17 more dB is a long way to go.
Uh, slightly. So you basically need ten times the power to increase 10 dB. If it takes 130,000 watts to do 177 dB then you'd need a nuclear reactor for 194 dB. What would it be? 8 million watts or so? (just a guess, it's too late for calculating)
Do you know of a laptop that will run more than 1GB?
Just curious. I've got 1GB in my Dell now and I run multiple VMware's all the time. If I'm not running anything memory intensive in Linux then it's not bad until I get past 3 or 4 Win2k sessions (which I don't need that often).
What sucks is when Linux swaps out the VMware process(es) for disk buffers. VMware tends to read a lot from the disk so all the VMware sessions end up fighting against each other. Not for memory or disk access, but against the damned disk buffering. I'd like to be able to tag the VMware process and give it 100-200MB of memory that can never, ever, be swapped out to disk.
For now I've found that giving the guest VMware sessions only around 64-128MB of memory (I usually do 96MB for Win2k) works best. The Linux disk buffering makes the guest session's swap access so fast that you can't tell it's swapping anyway. That frees up a lot of main memory and helps a little with the disk buffering problem.
They have traditionally been pretty good about letting people use the term Spam.
I have to agree with them on this. Anything like a company named "Spam Arrest" or "Fuck Spam, Inc." or something like that could be considered slander. I mean, if you ignore the e-mail side of things, it sounds like a company set up to make money by telling people how bad Spam ham is.
Slashdot Mirror
According to this article HTML is the best UI environment yet except for content creation.
Damn, and all this time I could've been playing Quake3 in HTML. Prolly would've had a much better UI. After all it's just a database interaction proggy.
And I didn't even know that HTML was great at controlling hardware. Dammit, all this time I could've been running my serial port controlled radio station in HTML. After all it's just a database interaction proggy.
And CRAP! My GPS navigation system in my car could be HTML pages!! I can't believe what I'm missing out on. Clicking pages, scrolling, and driving would so safe and fun. After all it's just a database interaction proggy.
etc...
And no, embeding some non-standard control in a HTML page does _NOT_ count. At that point you're opening a whole new can of worms, including the ability to use any of the crappy (not) controls the author mentions. It's arguable whether Javascript and friends count (definitely not Java applets).
What's the difference between this and SDL?
I've heard a lot about SDL but I've only ever heard anyone mention Allegro twice (including your post).
I have to agree.
It's too bad someone has not made the effort (AFAIK) to create the same type of integrated, stable, tight desktop environment on *BSD. On the server BSD is pretty good(*), but on the desktop it's a complete wash. It's worse than Linux because it's basically the same thing Linux has (GNOME, KDE, or whatever) except missing some really great software (nVidia lags behind with their BSD drivers and the only VMware you can run is 2.x).
(*) Only pretty good because I find BSD to be a big PITA when it comes time to upgrade the system. "Rebuild world" still requires a major effort to get the system in the correct state, not to mention the incredible time sink it is. Even just upgrading individual packages takes too much time and effort (same reason why I don't use Gentoo). A Debian apt-get upgrade or dist-upgrade is hella easier and fast.
I would really like to see a FreeBSD package system that uses dpkg to maintain everything (with the same or better quality put into it that FreeBSD normally sees) and a really good integrated desktop environment (At least as good as OS X or BeOS).
In any case, I'm a bit skeptical of the data. They seem to be monitoring the providers' own websites, not their clients' machines or sites.
That is totally true, and I'm skeptical also. They have Rackshack listed as NT/98, but in fact they offer servers of all sorts. My site is hosted on a RH Linux machine not Windows. Just because Rackshack's main site goes out doesn't mean any clients did (which I know for a fact has happened because once I couldn't get to the main site but my site worked fine).
Do you really need the operator syntax in every case?
... You don't need OOP to write software that does what you want.
I could rewrite the entire Linux kernel in 100% assembly language too. Doesn't mean I'd want to.
Operator overloading can be abused but when used correctly it goes a long ways towards making the code more readable. It allows constructs that are easy to code up and visualize. I mean, making the programmer's job easier is a good thing, no?
Think about this, in Java we don't need function overloading either. So why use it?
Hell, we don't need object-oriented stuff at all. Just use some flat language that doesn't support encapsulation, inheritance, function overloading, or
I won't reiterate what other's have posted in this thread but I'd like to add that TCC (Tiny CC) has an option for bound checking also.
TCC also lets you interpret C code on the fly as a scripting language (ie. without compiling it). Funky stuff.
If you did do that, it would still still work if you forced it into a value type:
matrix matrix3 = matrix1 + matrix2;
I should qualify that and say it might work. Technically the variable is out of scope and matrix3 contains an invalid value. I'm not sure what the standard says about that though.
I haven't read that book so I don't know specifically what you're talking about but... You can't return a reference to a temporary variable.
I just did another post on this here.
Everyone who keeps specifying the operator as operator+(matrix1, matrix2) is wrong. That's not the way you overload the + operator. This seems to indicate that many people on this thread are blasting away and have no idea how to program in C++. Anyway...
...What??? temp_matrix is the result
But what are you returning a reference to? For example (simplified):
matrix& operator+(const matrix& m)
{
matrix temp_matrix;
temp_matrix[0] = m[0] + (*this)[0];
temp_matrix[1] = m[1] + (*this)[1];
etc...or use a for-loop, whatever...
return
}
You can't return a reference temp_matrix because it's about to go out of scope (a good compiler will warn you about this).
If you did do that, it would still still work if you forced it into a value type:
matrix matrix3 = matrix1 + matrix2;
But at that point you are not using the reference. The reference would be copied by value into matrix3. Of course compilers these days can sometimes optimize that away, but not always.
If you could access the left side of the equation then you wouldn't need temp_matrix at all.
Take a look at java.awt.Graphics2D. Sound? The javax.sound package. Networking? java.net has everything you need.
Hehe... Sure, but those are not implemented in Java. It's only an interface to a C or C++ library. Java lacks the power to do it.
Yes, I freely admit that you can't write device drivers in Java, since that's about the only point you could possibly claim and actually be correct. For the 99.9999% of us that aren't actually writing device drivers, it's completely and utterly irrelevant. Use JNI if you really need something like that, and it's still easier than trying to write the whole program in C++.
Uh, but I'm not arguing how easy it is. I was just stating that Java is less powerful.
And it is most certainly not the only point where Java lacks power. Templates? (generics have less power) Multiple inheritance? What if I don't want garbage collection? The Java specification itself prevents Java programs from ever running as fast as the fastest C/C++ proggies. I could go on.
I write in Java too. I'm not arguing that Java isn't a good language or that it's not better than C++. I'm just saying it has less power.
How do you manipulate hardware devices (ie. pointers) in Java? Ain't got the "power", eh?
I can understand where you're coming from. C++ is a complex beast. I've been using some form of C++ for over 10 years (well before it was standardised) and I still don't understand everything about it.
With that said, it's an extremely powerful and flexible language. Very much more powerful than Java or C#. The complexity is mostly due to its flexibility. You can do (almost) anything with it. Of course, we can argue whether that's good or bad.
I think C++ can learn from Java though. The default should be to pass all non-built-in-type function parameters by const reference and the programmer has to specify otherwise (basically opposite of the way it is now). That would clean up the code a whole lot since 99% of the time that's what you want anyway. And the standard C++ library should have some sort of garbage collector available.
Another problem I have with C++ is that even with all its power you have no way to get to the "left hand" variable of operations. For example, if you have a matrix class you can overload the "+" operator so that you can do things like "matrix3 = matrix1 + matrix2". However, that's not going to be very efficient (assuming that's why you're using C++ in the first place) because there is no way to get to the matrix3 variable from inside the + operator. That forces you to use a temporary variable to add the two matrices then copy by value the whole matrix after adding matrix1 and matrix2. There are tricks around this problem but none are clean.
Same here. However, while most of the mail I'm getting is directly sent and from DSL/cable accounts, none of the boxes have been Windows boxes.
I've examined some of the boxes (by either NMAP, SSH, or telneting into them) and there were a couple routers (Linksys or similar home routers) but many of the boxes are actually Linux.
This seems to suggest one of two things to me: Either Linux boxes are getting hacked, or the spammers are using (multiple?) DSL accounts and Linux to send out their spam (this seems more likely to me).
This is basically the philosophy of Extreme Programming.
And my answer to the "Quick 'n Dirty" or "Correct and Proper" qustion is to use some or all of the Extreme Programming practices. I had been using some of those XP techniques way before anyone decided to define what XP was (mostly from hands-on programming experience over the last 20 years) and have found it provides a great balance between a perfect design and something that "just works". Having something that works is way, way more important than having a perfect design.
For the most part, that's the software that runs everything right now. The software that works. It may not have a perfect design, but it works. XP helps mediate the "bad design" part of it.
LOL... it has nothing to do with the maximum limits what you can see or what the monitor will display. It's overall game performance that's important.
Obviously you've never played FPS games professionally.
Although I'd really like to see a driver that installs on something other than RedHat (how about a plain tar.gz maybe?). It shouldn't be that hard to repackage.
The driver works with kernel 2.4.18, so I don't know why so many people are complaining about only supporting old kernels. It should work with any system with that kernel (it might work with any 2.4 kernel), even newer and other distros.
I assume they recognized that there are tons of 3D movie, animation studios, and medical research labs that are all moving to Linux. 3DLabs provides some of the higher-end workstation-class graphics cards used for 3D work.
Which is kind of stupid really; I'd imagine computers with lower specs but increased stability, efficiency (wasting less power on warming the office) and lower costs would be popular in the corporate scene.
Except for the lower cost thing I'd swear you're talking about a Mac.
Personally, I want Doom3 to run at 1000 FPS just like I like my Quake3 (yes, I can tell the difference and it makes a difference in competition).
I want my 3D renderings to be completed instantly and I want my videos encoded in real time.
I don't want to ever give a damn about running out of disk space. I want everything stored in a never ending, version controled, file system (searchable and indexable of course, even with many items and a large version database).
Plus everything else I could do if I just had a computer that was faster and had more disk space.
The current hardware is too slow, the current drives are too small. We don't need slower stuff.
Yes, they need something like that badly.
I remember poking around on PG not long ago but soon forgot about it.
If you're not looking for something specific then the site is kinda, meh. As you suggested, they need a news site, ratings, and other stats so you can see what's available.
And sections. "Technical", "Poetry", etc. Otherwise it's not very useful to the casual browser.
mmm, unless I am really missing something :) oopsie.
These are not LCD's. They are self illuminating and do not use backlighting. Also colorful. And from the look of the photo they do not appear to be translucent, but it's hard to tell.
If they could be made clear though... wow, it would look awesome. It would not be like a LCD because OLED's produce colorful light.
I wonder if they have a version that has or can make the pixels clear.
That would be cool. Imagine a large wall that has graphics on it but is also translucent. You know, like those screens you always see in sci-fi movies. Add some touch sensors...
I think they're up to 177, but your point still stands... 17 more dB is a long way to go.
Uh, slightly. So you basically need ten times the power to increase 10 dB. If it takes 130,000 watts to do 177 dB then you'd need a nuclear reactor for 194 dB. What would it be? 8 million watts or so? (just a guess, it's too late for calculating)
I didn't say I wanted DPI to make things crisper.
All 1600x1200 on a 15" display does is make things smaller.
I agree, and that's why I want higher resolutions. I want to fit more stuff on the screen at the same time. More code, more windows.
Do you know of a laptop that will run more than 1GB?
Just curious. I've got 1GB in my Dell now and I run multiple VMware's all the time. If I'm not running anything memory intensive in Linux then it's not bad until I get past 3 or 4 Win2k sessions (which I don't need that often).
What sucks is when Linux swaps out the VMware process(es) for disk buffers. VMware tends to read a lot from the disk so all the VMware sessions end up fighting against each other. Not for memory or disk access, but against the damned disk buffering. I'd like to be able to tag the VMware process and give it 100-200MB of memory that can never, ever, be swapped out to disk.
For now I've found that giving the guest VMware sessions only around 64-128MB of memory (I usually do 96MB for Win2k) works best. The Linux disk buffering makes the guest session's swap access so fast that you can't tell it's swapping anyway. That frees up a lot of main memory and helps a little with the disk buffering problem.
They have traditionally been pretty good about letting people use the term Spam.
I have to agree with them on this. Anything like a company named "Spam Arrest" or "Fuck Spam, Inc." or something like that could be considered slander. I mean, if you ignore the e-mail side of things, it sounds like a company set up to make money by telling people how bad Spam ham is.
Because my 15" Dell looks and works great at 1600x1200. Hell, the 900 verticle resolution is smaller than a tiny 1280x1024 screen.
This thing should be pushing WUXGA (1920x1200) resolutions or better for that size screen.