...I've seen people with MBAs struggle with how to turn on a computer, let alone use it. Anyone who has worked at a larger corporation has probably seen this level of ignorance. This just makes me kick myself for not thinking of it first...
Ok, I'll admit up front that I don't own a portable lossy audio device. I do, however, use Vorbis on my PC and think it is great. But, does anyone else think that these things are insanly expensive? I mean, $399 for a portable audio player?! I'd be happy with an Ogg Vorbis player that's about the size of a pager that I can take to the gym or wear while cutting the lawn. I'm thinking like 1-4 hours of music, maybe use a CF card or just sync via USB into on-board flash or RAM. I'm also of the belief that as much as vorbis rocks the world, I won't be getting rid of my CDs or non-lossy versions of the recordings.
Anybody else think this way, or am I in the minority?
- my comments apply to both Linux and WinXP
- my conecpt of painfully slow is likely different than most of the folks with the reply "I've got xxx machine which is slower than yours and I don't notice any issues." You want an example of speed? Load Opera sometime. Reliability isn't any better, but blows Moz out of the water in terms of speed
- I never compared anything to IE. I would rather use Lynx or have snails render my pages than touch IE.
My point was (like many of the other posts) that the "browser" would be better served to not have every feature under the sun and concentrate on reliability and speed. The rest will follow.
Mozilla.org released Mozilla v1.5 alpha today, with flavors available for Linux, Mac OS X, and Windows. Some of the new features include Composer enhancements, Chatzilla logging, multiple tab window closing confirmation, and quicksearch support in about:config. A more detailed rough changelog is also available.
...usability without having to have a monster machine. I use Mozilla, but damn is it slow for even the most mundane of pages. If you want to load a plugin, forget it. I'm running a 1.3GHz Athelon with 512MB of RAM and it gets bad. Personally, I couldn't care less about a good deal of the wiz-bang bloat features...make it fast and reliable.
Everyone else seems to have covered the concept of allegations vs. facts and date of IBM's entry into Linux, so I won't bother...
"SCO maintains that the code, if checked line by line, matches their original design and sometimes syntax."
Uh huh. So their argument is that given 2 developers trying to accomplish the same thing there is no way they can come up with the same or similar code? Please. While there may be more than one way to skin a cat, nothing says that 2 people won't skin them the same way. Especially when the goal is to provide an implementation of an existing technology (i.e. SysV type calls.)
On the other hand, if they are correct and the code is theirs and was illegally placed in the kernel tree, the ones approving what code goes into the kernel (still just Linus?) will really need to sit and think about how to prevent this in the future.
And the problem with that is? Don't get me wrong, I hate pop-ups, etc. more than anyone; but, I also have no problem with the credit of who provided you with a piece of software you value being displayed to the user. It can be done in a tasteful way. For example, for years I've seen the name of the author of my 3Com card's driver module in my system log when the driver loads. Do I care that it's there? Nope. Do I think that Donald Becker is a swell guy because now I can use my 3Com card in the OS of my choice? Damn skippy, I do. Would I have known his name if it wasn't in my system log? Probably not since it works great and I have no reason to look at the code. On the other hand, if the driver was logging something once a day or every time it sent 1MB worth of data, I probably would not be a happy camper.
I say kudos to those who take time out of their lives to allow me to use something I value. I'm glad I get to know a few of their names even though I may not be developing code in their neck of the woods.
Free software is not about egos, it is about keeping software free. Forcing something like this through licensing makes the software non-free.
So, you get the software free (as in beer) and you believe that if someone displays their name telling you they wrote it, that makes it non-free (again, as in beer)? The notion of free software (pay attention, this is the "as in speech" part) is that the person who writes the code shares the code for all to use, not so you don't have to pay money for it. If released under the GPL, the code must maintain credit to the original author. The only new part here is that the credit has visibility beyond those who care to look at the code.
Ok, unless I'm missing something here (and I could be), how are they going to prove someone is using a VPN or firewall? The VPN should look like normal IP traffic between 2 machines. Same thing with the firewall. While you could interpret the proposed (or approved in MI's case) law as making these illegal, if they are setup correctly it should be undetectable. From the ISP's point of view, all traffic from that subscriber is coming from the firewall and they can't detect the protected machines.
It's funny...I went back and read the service agreement for my ISP and while it prohibits creating a LAN with "un-approved" equipment, it also states that it is the subscriber's responsiblity to secure the machines he/she places on the network. So, a firewall used to shield the subscriber from the ton of port scans received daily, but really shouldn't be there because the subscriber has 3 machines on the LAN? Seems like a paradox to me and pretty much impossible to prove. Not that it makes the law, written as it is, good or valid.
Agreed, though it may not even take weeks for months. One company I worked at used relational databases for subscriber records throughout the system. We were in the middle of development and the subscriber database was corrupted by a bug in the new site controller software. They first saw the bug after bringing the system online for a weekend to stress test. By Monday morning the whole thing was dead and they couldn't go to the backups because they were backing up once a day, rotating between 2 tapes.
You'd be surprised how many places either:
- Don't back up - Back up using the same media over and over
End result, as you pointed out, is that the model for back ups can be a real problem. Not to mention, a month later your backup isn't real accurate.
I hope that Intel's documentation is better than TI's. TI has a huge chance to further penetrate the cellular/handheld market but their documentation, tools, etc. are sparse, vague, and poorly written. You have to document something well beyond the marketing material to get people to use it. The pretty pictures and snappy acronyms will only get you so far.
Somebody at TI should crack open a Motorola, ARM, Phillips, or Atmel manual or datasheet and see how to document a part so someone can use it.
...I know, the answer to the company directive that "security is the #1 priority": find someone else to blame. Classic. They can't really get their arms around security, so make someone else (the hardware mfrs.) take care of it. They must keep a whole farm of scapegoats somewhere. The sad part is, people will believe it. Nevermind that has security built int from the start.
Slashdot Mirror
...I've seen people with MBAs struggle with how to turn on a computer, let alone use it. Anyone who has worked at a larger corporation has probably seen this level of ignorance. This just makes me kick myself for not thinking of it first...
...SCO has hired Naji Sabri (of Iraqi foreign minister fame) as part of their defense team...
Ok, I'll admit up front that I don't own a portable lossy audio device. I do, however, use Vorbis on my PC and think it is great. But, does anyone else think that these things are insanly expensive? I mean, $399 for a portable audio player?! I'd be happy with an Ogg Vorbis player that's about the size of a pager that I can take to the gym or wear while cutting the lawn. I'm thinking like 1-4 hours of music, maybe use a CF card or just sync via USB into on-board flash or RAM. I'm also of the belief that as much as vorbis rocks the world, I won't be getting rid of my CDs or non-lossy versions of the recordings.
Anybody else think this way, or am I in the minority?
- my comments apply to both Linux and WinXP
- my conecpt of painfully slow is likely different than most of the folks with the reply "I've got xxx machine which is slower than yours and I don't notice any issues." You want an example of speed? Load Opera sometime. Reliability isn't any better, but blows Moz out of the water in terms of speed
- I never compared anything to IE. I would rather use Lynx or have snails render my pages than touch IE.
My point was (like many of the other posts) that the "browser" would be better served to not have every feature under the sun and concentrate on reliability and speed. The rest will follow.
Everyone else seems to have covered the concept of allegations vs. facts and date of IBM's entry into Linux, so I won't bother...
"SCO maintains that the code, if checked line by line, matches their original design and sometimes syntax."
Uh huh. So their argument is that given 2 developers trying to accomplish the same thing there is no way they can come up with the same or similar code? Please. While there may be more than one way to skin a cat, nothing says that 2 people won't skin them the same way. Especially when the goal is to provide an implementation of an existing technology (i.e. SysV type calls.)
On the other hand, if they are correct and the code is theirs and was illegally placed in the kernel tree, the ones approving what code goes into the kernel (still just Linus?) will really need to sit and think about how to prevent this in the future.
That's a HUGE leap since it becomes advertising.
And the problem with that is? Don't get me wrong, I hate pop-ups, etc. more than anyone; but, I also have no problem with the credit of who provided you with a piece of software you value being displayed to the user. It can be done in a tasteful way. For example, for years I've seen the name of the author of my 3Com card's driver module in my system log when the driver loads. Do I care that it's there? Nope. Do I think that Donald Becker is a swell guy because now I can use my 3Com card in the OS of my choice? Damn skippy, I do. Would I have known his name if it wasn't in my system log? Probably not since it works great and I have no reason to look at the code. On the other hand, if the driver was logging something once a day or every time it sent 1MB worth of data, I probably would not be a happy camper.
I say kudos to those who take time out of their lives to allow me to use something I value. I'm glad I get to know a few of their names even though I may not be developing code in their neck of the woods.
Free software is not about egos, it is about keeping software free. Forcing something like this through licensing makes the software non-free.
So, you get the software free (as in beer) and you believe that if someone displays their name telling you they wrote it, that makes it non-free (again, as in beer)? The notion of free software (pay attention, this is the "as in speech" part) is that the person who writes the code shares the code for all to use, not so you don't have to pay money for it. If released under the GPL, the code must maintain credit to the original author. The only new part here is that the credit has visibility beyond those who care to look at the code.
Ok, unless I'm missing something here (and I could be), how are they going to prove someone is using a VPN or firewall? The VPN should look like normal IP traffic between 2 machines. Same thing with the firewall. While you could interpret the proposed (or approved in MI's case) law as making these illegal, if they are setup correctly it should be undetectable. From the ISP's point of view, all traffic from that subscriber is coming from the firewall and they can't detect the protected machines.
It's funny...I went back and read the service agreement for my ISP and while it prohibits creating a LAN with "un-approved" equipment, it also states that it is the subscriber's responsiblity to secure the machines he/she places on the network. So, a firewall used to shield the subscriber from the ton of port scans received daily, but really shouldn't be there because the subscriber has 3 machines on the LAN? Seems like a paradox to me and pretty much impossible to prove. Not that it makes the law, written as it is, good or valid.
Agreed, though it may not even take weeks for months. One company I worked at used relational databases for subscriber records throughout the system. We were in the middle of development and the subscriber database was corrupted by a bug in the new site controller software. They first saw the bug after bringing the system online for a weekend to stress test. By Monday morning the whole thing was dead and they couldn't go to the backups because they were backing up once a day, rotating between 2 tapes.
You'd be surprised how many places either:
- Don't back up
- Back up using the same media over and over
End result, as you pointed out, is that the model for back ups can be a real problem. Not to mention, a month later your backup isn't real accurate.
I hope that Intel's documentation is better than TI's. TI has a huge chance to further penetrate the cellular/handheld market but their documentation, tools, etc. are sparse, vague, and poorly written. You have to document something well beyond the marketing material to get people to use it. The pretty pictures and snappy acronyms will only get you so far.
Somebody at TI should crack open a Motorola, ARM, Phillips, or Atmel manual or datasheet and see how to document a part so someone can use it.
...I know, the answer to the company directive that "security is the #1 priority": find someone else to blame. Classic. They can't really get their arms around security, so make someone else (the hardware mfrs.) take care of it. They must keep a whole farm of scapegoats somewhere. The sad part is, people will believe it. Nevermind that has security built int from the start.