You could get a much better CRT monitor than LCD for not much bucks. I have an iiyama 19" which does 85hz at 1600x1280 and is as sharp as a tack. The brightness levels are brilliant and there is no ghosting at all. All for about 450 of your American dollars - not a bad deal. The *only* thing which is not so great is the sheer size of the beast. It looms on my desk and since my desk is a crappy Ikea job (just a plain desk, not a workstation or anything that awful) I am constantly worried the weight will break the desk. NOTE: I also have bookshelf speakers and 6 units of rackmount on said desk, so it's getting pretty loaded down now.
I know most guys out there would just love to put together a three LCD screen like the traders have, but I've worked on multimonitor screens before and they are more distracting than useful. You are constantly craning one way or the other (neck strain) to see a window that is on some other screen. Let's not forget the way pop-ups appear in the gap between the two screens if you have a two monitor setup. I'd say there's plenty of life in the CRT yet.
You would develop some pretty nasty RSI issues if you used this a lot...but who's going to do that. I think the purpose of the technology is to allow you to bang out a quick (and irrelevent) SlashDot comment while on the move. This would be great on the train home from work for example. You could reply to all your email of the day in otherwise unused time - then spend the 30 minutes you normally take to email people with your family instead.
If a single article in Times is enough to equip sniper rifle carrying serial killers in your country then you're in a bad way. Seriously, do you think that just because people now know of one model of sniper rifle they are ready to become expert killers.
They would get much better info from a copy of Guns and Ammo magazine, and an address where they could also purchase said weapons. This magazine even carries stories in it from rabib readers on "How guns save lives". I'm not kidding!
Moore's law is meant to bunk out around 2010 (or so) so we can expect hardware to improve about 20 fold before then. Add to that the ability to all run quad processor and quad vid processor rigs and I think we could be making out way towards final fantasy in real time. We certainly won't be too far off the mark.
And before you all scoff, remember that DSP and custom chip logic for specific jobs (e.g. 3d chips and 3d audio chips) can get a lot more bang for buck from the hardware.
There is so much potential computing power that hasn't been harnessed. What about video cards that don't concentrate on triangle rendering speed, but instead work on rendering fractals at high speed. If you ever got to see fractal image compression in action then you will be aware of how much more powerful the fractal is than block oriented compression systems.
The idea of an mp3 hacking the computer through the player is only slightly more credible than that of a txt hacking the computer through the text editor.
Note the following bugtraq announcement.
- Sandblad advisory #5 -
Title: Mp3 file can execute code in Winamp.
Date: [2002-04-26]
Software: Nullsoft Winamp 2.79
Rating: High because mp3 files are widely trusted as safe.
Impact: Specially crafted mp3 file can execute
arbitrary code when played in Winamp due to a
buffer overflow condition.
Vendor: Nullsoft has confirmed the vulnerability.
Patch: Winamp 2.80 released 02-04-25 will fix the issue.
Download at: http://www.winamp.com/
Workaround: Disable the minibrowser
(enabled by default)
Author: Andreas Sandblad, sandblad@acc.umu.se (o o)
NON TECHNICAL DESCRIPTION:
It is possible to modify an existing mp3 file in such a way that it can
carries a virus. The virus is activated when the mp3 file is played in
Winamp and can then infect other mp3 files found on harddrives or network
shares. In order to protect yourself you need to upgrade to Winamp 2.80 or
disable the minibrowser.
The code he posted is used to infect the MP3 files, rather than XMMS. It is simply using a buffer overflow to execute some arbitrary code. You should get an MD5 hash of your MP3 files instead. Tripwire would detect this of course, if you are paranoid enough to run it on your workstation as well as your servers.
You wouldn't need to infect each MP3 the same. You could use a rota system to infect each file with one exploit from a list. e.g. user has 2000 MP3's on machine, 500 get WinAmp exploit. 500 get MPG123 exploit, 500 get Xine exploit. Infect some with Windows code, and others with Linux code as desired.
Just because you're poisoning the well, doesn't mean you have to use the same poison in each well.
All quite true. The problem lies in the way patents are handed out with little regard for prior art or sense. For examples see the BT hyperlink patent, OneClick, and the laughable set of patents on "e-commerce" functionality currently doing the rounds. These can all be contested, but only by an entity with sufficient cash to battle the courts and laywers. Not all small companies can obtain the VC that you talk about to defend these patents.
e.g. If you were sued for ecommerce functionality what VC would bother to invest to defend you? There would be no bottom line return on the investment if you are only a small company.
My basic understanding was that the patent is valid for 7 years. Unix is over 30 years old. What sort of patents could still be available after for exploitation after all this time?
Little guys need protection, and the patent office gives them that.
And that's why all the big companies own the patents? I don't own any patents and I'm a little guy...how is my right to innovate being protected when I can't even create a simple compression scheme similar to LZW without being sued.
It costs no extra money to develop a web site that is standards compliant (retrofitting an already severly broken site is a different matter).
Are you insane or do you not actually develop commercial websites and thus have no understanding of how much more effort it is to try and meet standards and cross browser issues? Support for older browsers (why won't Netscape curl up and die!) typically adds another 30% development time for HTML around here because we need to code around their foibles. Each needs it's own CSS and require other code changes to ensure consistency across the browsers.
Coding to a selected "standard" might get you basic support on IE5.5 and Netscape 6.0 and above, but it won't help if you need to support IE4 and Netscape 4.
Sadly, real role-playing is exactly what these games don't seem to engender. Mostly it is exactly about what those slacker friends of yours were about...the grabbing of big stacks of treasure and the slaying of high level monsters. Having recently worked my way through NWN I am hoping for much better from an online community, but are doubtful it exists. What's fun about role-playing is not the number or orcs you have slain, but the interaction and humour you share with a group of friends. When the online games become more condusive to player interaction then I'll be hooked hopelessly.
When you're writing server software then logging is essential and it stays in, even in release code. I write systems that need to function 24/7 and are deployed on servers where having a debugger is strictly forbidden (no compilers or other uneccessary tools). I log all important actions to plain text files, and all exceptions to the event log (and the plain text files) with the class and line of code that raised the exception. This allows me to determine what when wrong in a production system, even days after the fact. It's absolutely vital to the long term health of those systems.
Conversely, debugging is brilliant at the write and unit test stages. Stepping into the code and ensuring it does what you thought it should can be illuminating. However, I still see these debuggers as a crutch and prefer to read and model the code in my head prior to firing up a debugger.
Each technique has their place...horses for courses.
Couldn't you just draw the changed part of the window into the second buffer, then flip. If you then update the first buffer (in a background thread maybe) so it mirrors what's on screen you will be ready to flip again. This does mean drawing everything twice, but you are at least only drawing the dirty parts of the screen - and that will save CPU cycles.
I can't help but suspect it's the 1GB of RAM that really makes the difference. Event the jump from 256MB to 512MB can make a huge difference in apparent speed of machine.
Note: I am also a little weary of benchmarks, but Toms H. is usually pretty damn good at them.
Slashdot Mirror
I know most guys out there would just love to put together a three LCD screen like the traders have, but I've worked on multimonitor screens before and they are more distracting than useful. You are constantly craning one way or the other (neck strain) to see a window that is on some other screen. Let's not forget the way pop-ups appear in the gap between the two screens if you have a two monitor setup. I'd say there's plenty of life in the CRT yet.
Why would someone want a $500 leather seat on their motorbike?
From memory it was Butch Vig of Garbage fame that produced Nevermind. He's a kick as producer! Those drums and bass sounds rock out.
You would develop some pretty nasty RSI issues if you used this a lot...but who's going to do that. I think the purpose of the technology is to allow you to bang out a quick (and irrelevent) SlashDot comment while on the move. This would be great on the train home from work for example. You could reply to all your email of the day in otherwise unused time - then spend the 30 minutes you normally take to email people with your family instead.
They would get much better info from a copy of Guns and Ammo magazine, and an address where they could also purchase said weapons. This magazine even carries stories in it from rabib readers on "How guns save lives". I'm not kidding!
And before you all scoff, remember that DSP and custom chip logic for specific jobs (e.g. 3d chips and 3d audio chips) can get a lot more bang for buck from the hardware.
There is so much potential computing power that hasn't been harnessed. What about video cards that don't concentrate on triangle rendering speed, but instead work on rendering fractals at high speed. If you ever got to see fractal image compression in action then you will be aware of how much more powerful the fractal is than block oriented compression systems.
Note the following bugtraq announcement.
- Sandblad advisory #5 - Title: Mp3 file can execute code in Winamp. Date: [2002-04-26] Software: Nullsoft Winamp 2.79 Rating: High because mp3 files are widely trusted as safe. Impact: Specially crafted mp3 file can execute arbitrary code when played in Winamp due to a buffer overflow condition. Vendor: Nullsoft has confirmed the vulnerability. Patch: Winamp 2.80 released 02-04-25 will fix the issue. Download at: http://www.winamp.com/ Workaround: Disable the minibrowser (enabled by default) Author: Andreas Sandblad, sandblad@acc.umu.se (o o) NON TECHNICAL DESCRIPTION: It is possible to modify an existing mp3 file in such a way that it can carries a virus. The virus is activated when the mp3 file is played in Winamp and can then infect other mp3 files found on harddrives or network shares. In order to protect yourself you need to upgrade to Winamp 2.80 or disable the minibrowser.
The code he posted is used to infect the MP3 files, rather than XMMS. It is simply using a buffer overflow to execute some arbitrary code. You should get an MD5 hash of your MP3 files instead. Tripwire would detect this of course, if you are paranoid enough to run it on your workstation as well as your servers.
Just because you're poisoning the well, doesn't mean you have to use the same poison in each well.
It appears to be suffering the latest 0-day release DDoS called "slashdotted".
That's 95% of hosts they are claiming...not networks.
All quite true. The problem lies in the way patents are handed out with little regard for prior art or sense. For examples see the BT hyperlink patent, OneClick, and the laughable set of patents on "e-commerce" functionality currently doing the rounds. These can all be contested, but only by an entity with sufficient cash to battle the courts and laywers. Not all small companies can obtain the VC that you talk about to defend these patents.
e.g. If you were sued for ecommerce functionality what VC would bother to invest to defend you? There would be no bottom line return on the investment if you are only a small company.
My basic understanding was that the patent is valid for 7 years. Unix is over 30 years old. What sort of patents could still be available after for exploitation after all this time?
Which patents...we need specific examples.
And that's why all the big companies own the patents? I don't own any patents and I'm a little guy...how is my right to innovate being protected when I can't even create a simple compression scheme similar to LZW without being sued.
NOTE: quoted with permission from some stupid money grubbing corporate entity
Are you insane or do you not actually develop commercial websites and thus have no understanding of how much more effort it is to try and meet standards and cross browser issues? Support for older browsers (why won't Netscape curl up and die!) typically adds another 30% development time for HTML around here because we need to code around their foibles. Each needs it's own CSS and require other code changes to ensure consistency across the browsers.
Coding to a selected "standard" might get you basic support on IE5.5 and Netscape 6.0 and above, but it won't help if you need to support IE4 and Netscape 4.
Sadly, real role-playing is exactly what these games don't seem to engender. Mostly it is exactly about what those slacker friends of yours were about...the grabbing of big stacks of treasure and the slaying of high level monsters. Having recently worked my way through NWN I am hoping for much better from an online community, but are doubtful it exists. What's fun about role-playing is not the number or orcs you have slain, but the interaction and humour you share with a group of friends. When the online games become more condusive to player interaction then I'll be hooked hopelessly.
When you're writing server software then logging is essential and it stays in, even in release code. I write systems that need to function 24/7 and are deployed on servers where having a debugger is strictly forbidden (no compilers or other uneccessary tools). I log all important actions to plain text files, and all exceptions to the event log (and the plain text files) with the class and line of code that raised the exception. This allows me to determine what when wrong in a production system, even days after the fact. It's absolutely vital to the long term health of those systems.
Conversely, debugging is brilliant at the write and unit test stages. Stepping into the code and ensuring it does what you thought it should can be illuminating. However, I still see these debuggers as a crutch and prefer to read and model the code in my head prior to firing up a debugger.
Each technique has their place...horses for courses.
Couldn't you just draw the changed part of the window into the second buffer, then flip. If you then update the first buffer (in a background thread maybe) so it mirrors what's on screen you will be ready to flip again. This does mean drawing everything twice, but you are at least only drawing the dirty parts of the screen - and that will save CPU cycles.
Come along doggie...
d @
I seem to recall someone saying that MS are actively blocking these sorts of sales on E-bay.
So you probably can't see your own comment then...since it's only got a 2.
What sort of slashdotter is capable of getting a date? Oh wait, are you a girl perhaps...?
Note: I am also a little weary of benchmarks, but Toms H. is usually pretty damn good at them.