What damage? Please give me an example of 80 lines of code that could cause the damage SCO claims they did. 80 lines changed Linux into a "luxury car" from a "bycicle"? Nonsense.
You're counting wrong. What that gives you is the number of *characters* in all the filenames that find finds. To count lines the argument to wc should be -l, and the input to wc should be the actual source code, not file names.
Even if those 80 lines were really copied from Linux, it's not a big deal. Any decent programmer can write 80 lines in less than an hour. Even if those 80 lines do something unusually cool and hard to come up with, I'm still sure that there's no way that just 80 lines of code can give Linux any significant advantage. Fix a bug, or get a driver to work? Maybe. Give Linux the quality SCO claims it wouldn't have if IBM hadn't copied the code from them? No way.
XWindows is not painful at any resolution by itself. The problem are the window manager, widgets, etc. It's perfectly possible to write a program that looks well when run on X at 320x200.
That's stupid. You're not a spider. Chemicals have different effects on different animals. For example aspirin is very dangerous to cats, IIRC. But it's just fine for humans, even in quite large amounts.
There's code reuse of the small stuff. Say, if you're coding VB you'll soon have a big library of nice stuff that should have been there in the first place. I have functions like Limit, FirstWord, Capitalize, an object that can do logging to a file, database or anything else, etc. All that is small stuff on its own, but adds up after a while. Using another developer's stuff is also problematic. When you write a function you have some idea of what might go wrong in it. If using somebody else's code you'll often blame yourself first, and only after that look at the library. And there's integration. My library uses this logger object, so that I get messages about possible bugs in the program's log.
Then there's the big stuff you've *got* to reuse, or you'd die of boredom or never finish it. Things like big Windows controls, spreadsheet control, fancy toolbar/statusbar control, installer, etc. In these cases reuse is a very good idea, especially if you're using a stable product that's been well tested. Then of course there's always the evil component that worked for 99% of what you needed, is used in a lot of places in your program and absolutely refuses to do the remaining 1% right. With enough time it can be much more satisfying to roll your own.
Hmm, I suppose I have the same problem (I take no medication), I keep switching between different thoughts randomly pretty much all the time. People find it a bit odd to see me sitting somewhere and changing between looking annoyed and happy every few minutes. I have a pretty good imagination and sometimes plan whole conversations in advance.
On the other hand, I can concentrate very well on the things I like. Reading a book in a day is not a problem. Same as spending hours writing code.
The thing here is what you interpret as "normal". If you define "normal" as "what most of the population is like" then no, we're not normal. But then geeks, homosexuals, and every other "deviation" from the norm is not. Rather than consider it like some kind of illness I just consider it to be a part of my normal behavior. It also depends on how hyperactive you are. I don't consider myself to be a very bad case. Now if I couldn't sit quiet for a minute then I'd be worried. The problem is where to place the line between what's okay and what's excessive.
OCR. Eventually I suppose somebody will just put a copy of the source online if typing it is so much effort. Then, why type it? If you want to study how it works then paper should be good enough, if you're evil and want to infect somebody then I imagine that finding the binary version isn't too hard.
Linux boot time would be easier to measure. You can measure it more accurately. How do you exactly know when Windows finished loading? With what drivers/software?
With Linux you can just measure the time spent on booting say, the minimum Debian install. It'd be easy to make a CD for benchmarking purposes that'd quicky make that miminim install on the hard disk.
Hmm, if that was the same show I saw a few years ago, then I don't see what's wrong with it, it was pretty cool. Although maybe it's a different one, I can't remember if it's the same or not. It had a small girl robot, at least.
I remember that my tecnology teacher (kind of a very basic introduction to electronics and general tech stuff) was telling us that they had a lot of fun at school. Turns out one of their favourite jokes was finding a capacitor that could stand 200 volts or so, charging it, and leaving it on the table. Evenutally somebody always picked it up and got a good shock.
That probably wouldn't kill him, IIRC there's some protection that turns the power off if you make a contact between the live wire and a ground that's not connected to the neutral wire. Sticking a knife in one hole should cause that. Or something of that kind, it was a long time ago and I forgot.
I also remember that once I stuck a piece of wire in both holes of the socket as a kid, and all I got was a scary shock that my parents didn't find out about. I suppose that since the wire conducts better than a human not much harm could have been done.
I suppose I could have died if I held a wire in each hand, and plugged that into the socket, but that's pretty hard to do.
This is probably very inexact, it's from memory and translated from spanish:
A man had to take a plane but he was very nervous thinking there might be somebody with a bomb on board. He went to the pilot and asked how likely it was that there was somebody with a bomb on board.
The pilot answered "Well, I wouldn't worry about it at all. It's very unlikely. Probably something like 1 chance in a million". The man feeling somewhat better then asked: "And what is the chance of that there are two bombs on board?". The pilot answered "Oh, now that is extremely unlikely. I'd say it's so unlikely that you could say it's practically impossible"
The man feeling relieved then said "That's what I thought as well, which is why I brought this bomb with me".
Many Unix config files are just shell scripts that get executed to set some variables. That's hardly harder than reading and writing a few keys.
The registry has its share of problems, too. It's slow, ugly, inefficent, space limited (see registry size limit in NT), some sections (DLLs especially) can't be longer than 64K, and for complex storage needs you'll need to write your own parser anyway, especially with VB, which has a really limited access to the registry.
Even without fancy transparency features PNG is still very useful.
First, unlike GIF it supports more than 256 colors Second, just like GIF it's lossless (It could be said that conversion to 256 colors is a loss though) Third, it's a bit smaller than GIF
JPG can't deal properly with thin lines. Which means that if you want a good encoding of a picture that contains lines and text, like color comics you have to use JPG or GIF. If the comic happens to have many colors then PNG is the only thing that will look good.
Let's do a calculation: 1GB transferred with 128 byte packets gives 8388608 packets. With 56 bytes of TCP/IP data per packet that makes 448MB of overhead. Yeah, the download will be going slower, but a lot of bandwidth will be lost on TCP/IP.
The whole idea is useless, anyway. Many tools like Snort can already reassemble fragments to avoid being foiled by tricks like this.
Oh, and you can tell the remote host to send smaller packets by changing the MTU.
Forgot to say that it's not that hard to work around it, anyway. I already use SSL everywhere I can. It'll be just a matter of adding SSL support to programs, or writing some kind of proxy, which would add some compression as a side effect.
BTW, your ISP would do well by blocking telnet. There's SSH, and it's not that hard to use.
As if they couldn't do this already. Some imaginative packet filtering can do wonders, and IIRC, Linux has some support in iptables for examining the content of the packets. I'm also sure that it's not too hard for an ISP to pay a programmer to create a Linux module like this. Cisco probably has that as a feature too in some expensive routers.
No, I have rearranged it for Dvorak, although I still type qwerty. Confuses the hell out of people. Even the guy who came to install the ADSL couldn't type on it.
I've been thinking about switching for a while, but hadn't had time. Maybe this summer.
Yeah, but I was talking about the keyboard as a device, not the layout. The russian layout for example has nothing to do with QWERTY or Dvorak, anyway, so the speed argument applies only to keyboards with a QWERTY or similar layout.
Well, I think that this means that the keyboard practically reached perfection.
CPUs can be improved a lot, but the keyboard has been doing its job just fine for a long time. If we're still using the same design as 10 years ago then it means that simply no improvement is needed.
Other fancy replacements didn't gain any noticeable market share. Voice recognition is still far from perfect and is noticeably slower. I suppose a direct interface to the computer might finally replace the keyboard, but I doubt that will become available any time soon.
While I agree that laws are getting insane lately, the way how things currently are it actually makes sense. Currently it seems that paying $800 for a license, which may not even need to be a physical object, so what's so strange about owning something in a virtual world?
Security can't be just added like that to a product. Security is not a state, it's a process, and has to be had in mind right from the beginning. Closing a few buffer overruns won't magically make Windows safe. There are other kinds of security problems, and I'm pretty sure there's a good number of them that exist due to the design of the API itself.
Slashdot Mirror
What damage? Please give me an example of 80 lines of code that could cause the damage SCO claims they did. 80 lines changed Linux into a "luxury car" from a "bycicle"? Nonsense.
You're counting wrong. What that gives you is the number of *characters* in all the filenames that find finds. To count lines the argument to wc should be -l, and the input to wc should be the actual source code, not file names.
Even if those 80 lines were really copied from Linux, it's not a big deal. Any decent programmer can write 80 lines in less than an hour. Even if those 80 lines do something unusually cool and hard to come up with, I'm still sure that there's no way that just 80 lines of code can give Linux any significant advantage. Fix a bug, or get a driver to work? Maybe. Give Linux the quality SCO claims it wouldn't have if IBM hadn't copied the code from them? No way.
Easy, make keys for your children, and encrypt to them.
XWindows is not painful at any resolution by itself. The problem are the window manager, widgets, etc. It's perfectly possible to write a program that looks well when run on X at 320x200.
That's stupid. You're not a spider. Chemicals have different effects on different animals. For example aspirin is very dangerous to cats, IIRC. But it's just fine for humans, even in quite large amounts.
There's code reuse of the small stuff. Say, if you're coding VB you'll soon have a big library of nice stuff that should have been there in the first place. I have functions like Limit, FirstWord, Capitalize, an object that can do logging to a file, database or anything else, etc. All that is small stuff on its own, but adds up after a while. Using another developer's stuff is also problematic. When you write a function you have some idea of what might go wrong in it. If using somebody else's code you'll often blame yourself first, and only after that look at the library. And there's integration. My library uses this logger object, so that I get messages about possible bugs in the program's log.
Then there's the big stuff you've *got* to reuse, or you'd die of boredom or never finish it. Things like big Windows controls, spreadsheet control, fancy toolbar/statusbar control, installer, etc. In these cases reuse is a very good idea, especially if you're using a stable product that's been well tested. Then of course there's always the evil component that worked for 99% of what you needed, is used in a lot of places in your program and absolutely refuses to do the remaining 1% right. With enough time it can be much more satisfying to roll your own.
Hmm, I suppose I have the same problem (I take no medication), I keep switching between different thoughts randomly pretty much all the time. People find it a bit odd to see me sitting somewhere and changing between looking annoyed and happy every few minutes. I have a pretty good imagination and sometimes plan whole conversations in advance.
On the other hand, I can concentrate very well on the things I like. Reading a book in a day is not a problem. Same as spending hours writing code.
The thing here is what you interpret as "normal". If you define "normal" as "what most of the population is like" then no, we're not normal. But then geeks, homosexuals, and every other "deviation" from the norm is not. Rather than consider it like some kind of illness I just consider it to be a part of my normal behavior. It also depends on how hyperactive you are. I don't consider myself to be a very bad case. Now if I couldn't sit quiet for a minute then I'd be worried. The problem is where to place the line between what's okay and what's excessive.
OCR. Eventually I suppose somebody will just put a copy of the source online if typing it is so much effort. Then, why type it? If you want to study how it works then paper should be good enough, if you're evil and want to infect somebody then I imagine that finding the binary version isn't too hard.
Linux boot time would be easier to measure. You can measure it more accurately. How do you exactly know when Windows finished loading? With what drivers/software?
With Linux you can just measure the time spent on booting say, the minimum Debian install. It'd be easy to make a CD for benchmarking purposes that'd quicky make that miminim install on the hard disk.
Hmm, if that was the same show I saw a few years ago, then I don't see what's wrong with it, it was pretty cool. Although maybe it's a different one, I can't remember if it's the same or not. It had a small girl robot, at least.
I remember that my tecnology teacher (kind of a very basic introduction to electronics and general tech stuff) was telling us that they had a lot of fun at school. Turns out one of their favourite jokes was finding a capacitor that could stand 200 volts or so, charging it, and leaving it on the table. Evenutally somebody always picked it up and got a good shock.
That probably wouldn't kill him, IIRC there's some protection that turns the power off if you make a contact between the live wire and a ground that's not connected to the neutral wire. Sticking a knife in one hole should cause that. Or something of that kind, it was a long time ago and I forgot.
I also remember that once I stuck a piece of wire in both holes of the socket as a kid, and all I got was a scary shock that my parents didn't find out about. I suppose that since the wire conducts better than a human not much harm could have been done.
I suppose I could have died if I held a wire in each hand, and plugged that into the socket, but that's pretty hard to do.
This is probably very inexact, it's from memory and translated from spanish:
A man had to take a plane but he was very nervous thinking there might be somebody with a bomb on board. He went to the pilot and asked how likely it was that there was somebody with a bomb on board.
The pilot answered "Well, I wouldn't worry about it at all. It's very unlikely. Probably something like 1 chance in a million". The man feeling somewhat better then asked: "And what is the chance of that there are two bombs on board?". The pilot answered "Oh, now that is extremely unlikely. I'd say it's so unlikely that you could say it's practically impossible"
The man feeling relieved then said "That's what I thought as well, which is why I brought this bomb with me".
Many Unix config files are just shell scripts that get executed to set some variables. That's hardly harder than reading and writing a few keys.
The registry has its share of problems, too. It's slow, ugly, inefficent, space limited (see registry size limit in NT), some sections (DLLs especially) can't be longer than 64K, and for complex storage needs you'll need to write your own parser anyway, especially with VB, which has a really limited access to the registry.
Even without fancy transparency features PNG is still very useful.
First, unlike GIF it supports more than 256 colors
Second, just like GIF it's lossless (It could be said that conversion to 256 colors is a loss though)
Third, it's a bit smaller than GIF
JPG can't deal properly with thin lines. Which means that if you want a good encoding of a picture that contains lines and text, like color comics you have to use JPG or GIF. If the comic happens to have many colors then PNG is the only thing that will look good.
Heh, wrong. The admin will hate you for that.
Let's do a calculation: 1GB transferred with 128 byte packets gives 8388608 packets. With 56 bytes of TCP/IP data per packet that makes 448MB of overhead. Yeah, the download will be going slower, but a lot of bandwidth will be lost on TCP/IP.
The whole idea is useless, anyway. Many tools like Snort can already reassemble fragments to avoid being foiled by tricks like this.
Oh, and you can tell the remote host to send smaller packets by changing the MTU.
Forgot to say that it's not that hard to work around it, anyway. I already use SSL everywhere I can. It'll be just a matter of adding SSL support to programs, or writing some kind of proxy, which would add some compression as a side effect.
BTW, your ISP would do well by blocking telnet. There's SSH, and it's not that hard to use.
As if they couldn't do this already. Some imaginative packet filtering can do wonders, and IIRC, Linux has some support in iptables for examining the content of the packets. I'm also sure that it's not too hard for an ISP to pay a programmer to create a Linux module like this. Cisco probably has that as a feature too in some expensive routers.
No, I have rearranged it for Dvorak, although I still type qwerty. Confuses the hell out of people. Even the guy who came to install the ADSL couldn't type on it.
I've been thinking about switching for a while, but hadn't had time. Maybe this summer.
So, what is the monetary value of an hour with your SO?
Yeah, but I was talking about the keyboard as a device, not the layout. The russian layout for example has nothing to do with QWERTY or Dvorak, anyway, so the speed argument applies only to keyboards with a QWERTY or similar layout.
Well, I think that this means that the keyboard practically reached perfection.
CPUs can be improved a lot, but the keyboard has been doing its job just fine for a long time. If we're still using the same design as 10 years ago then it means that simply no improvement is needed.
Other fancy replacements didn't gain any noticeable market share. Voice recognition is still far from perfect and is noticeably slower. I suppose a direct interface to the computer might finally replace the keyboard, but I doubt that will become available any time soon.
While I agree that laws are getting insane lately, the way how things currently are it actually makes sense. Currently it seems that paying $800 for a license, which may not even need to be a physical object, so what's so strange about owning something in a virtual world?
Security can't be just added like that to a product. Security is not a state, it's a process, and has to be had in mind right from the beginning. Closing a few buffer overruns won't magically make Windows safe. There are other kinds of security problems, and I'm pretty sure there's a good number of them that exist due to the design of the API itself.