They had NT 4.0 out not very long after Win95 -- and NT4 was an upgrade (some like to call it a downgrade because of some of Microsoft's design decisions when rewriting the code, like allowing third party device drivers to run in ring 0) of NT3.51, not a rewrite of Win95. Then of course there was Windows 98, ME and then Windows 2000 (by many thought to be an upgrade to the Win9x family), followed by Windows XP.
What Microsoft did with Windows 95, was create first the OEM Service Releases (OSR2 actually worked pretty well, at least for me), then Win98 and WinME. Windows 2000 was developed out of NT 4.0, which in turn came from Windows NT 3.51, the origins of which can be traced to the old days of OS/2.
And I actually liked OS/2. PM has (had) its own ways of doing things (like the Shift, Alt and Ctrl mouse-drag combinations to copy, move or create a shadow of a file, which were dependent on the source and destination media), but the system was actually pretty usable. It really is too bad that it failed. I still have an OS/2 Warp 3 CD plus the Bonus Pak around here somewhere along with OS/2 2.11 floppies... might install it some day just for the heck of it...
Don't you think that is something that could be called a Win-Win situation?
I can't help but think back on the days when one could create their own WIN.COM file...
"The activation process authenticates licensed users of "shrinkwrapped" products (for retail purchase) without hindering their ability to use the software the way they always have."
I personally believe that software should written be for the user, not someone else. But the above quote after all doesn't deal with use of the software. It only deals with activation, a one-time-per-installation process.
Now, what if I want to reinstall my copy of Photoshop in ten years? Can Adobe guarantee that I will be able to do that? In 20 years? If they cannot guarantee that, it is indeed hindering my ability to use the software, regardless of marketing hype.
No, thanks. I will stick with software that doesn't require activation, and doesn't limit by design what I can do with it.
And unlike Linux, it's not just the kernel... it's got its own shell, programming language and God knows what else. It even includes a built-in text editor, lo and behold!
The difference is that fax machines, cars and many other "closed" appliances in the everyday world work according to very well defined principles. The protocols used for transfering a fax over the phone lines is documented and no one will prosecute you if you make an alternative implementation of it. Cars just get you from one place to another, sometimes with bells and whistles, but that's what they do. So do bicycles, buses, trains, and a number of other means of transportation.
In contrast, closed source software is very rarely documented enough so that complete interoperability is easily achievable. Imagine if only vehicles with a particular wheel spacing were usable on public roads, but the information on the necessary spacing was secret and deducing it from available, usable cars was punishable with severe fines or even jail time? That, in my opinion, is a better analogy than just saying that governments should not mandate use of one particular piece of technology simply because another exists.
The government has several functions, two of which are: (a) serve the public, and (b) preserve documents for the future. Microsoft, which must be said to be the most common provider of proprietary desktop software, has a tendency to change their file formats every few major revisions, sometimes more often than that. This wouldn't be so much of a problem if there was a reliable means of reading the old file formats in a new version, but unfortunately that is far from always the case. Coupled with a potential inability to use old software in a year, a few years or even a few decades (mandated software activation, on-line forced upgrades and the "Trusted" Computing Platform), this may grow into a very serious problem. When converting the documents, can you guarantee that no information or formatting is lost or altered? I wouldn't want to bet on it.
Open source software will not automatically mean that old documents can be read. But if the lowest level specification for a file format, the source code for the application used to create it, is freely available, that means that one is always guaranteed the possibility of creating tools to read and/or convert the data to another format in the future. (At least as long as the source code, or at the very least extensive documentation of the file format, is preserved.)
So, I think that saying that mandating government use of open source software is wrong because mandating government use of cars is wrong, is a rather flawed logic. As I understand it most governments are already required to use the best, most cost effective tool for the job. "Best" in terms of handling documents subject to long term preservation requirements would seem to include the possibility to read those documents in both five, ten and fifty or even a hundred years, with a minimum of hassle.
Just like at times, a first class jet airline ticket might be more cost effective than a second class train ticket for getting from one place to another.
Sure! "You can trust us, because we don't trust you." (Isn't that how it is in the NGSCB (Palladium) world, though?)
Re:Moore's Law vs. Evolution
on
Real Security?
·
· Score: 1
The problem is that it is always easier to be insecure than it is to be secure. If you want to be secure, you need to consider every possible attack vector. An attacker, or anyone else wanting to prove your systems insecure, only need to find one attack vector that you didn't think of.
Myself, I keep my passwords on my Palm Vx, encrypted (using software specifically designed for the purpose and released under the GPL) and protected using a single "master password". I remember my passwords most of the time, but this gives me a chance to refresh my memory if I can't, without using the (often horribly insecure) "password recovery" mechanisms provided by some systems. Since the Palm Vx is almost always right with me - in my pocket or next to me on the desk - and the data is encrypted while in storage and only decrypted upon accessing it, I believe this is a fair compromise between security and convinience.
And no, I don't use completely random passwords, either. They might work if you have just one that you need to remember. Anything beyond that, it's a lost cause. Better to come up with a good password that you can remember, than an excellent that you have to write down and refer to the note every time you need it.
Slashdot Mirror
They had NT 4.0 out not very long after Win95 -- and NT4 was an upgrade (some like to call it a downgrade because of some of Microsoft's design decisions when rewriting the code, like allowing third party device drivers to run in ring 0) of NT3.51, not a rewrite of Win95. Then of course there was Windows 98, ME and then Windows 2000 (by many thought to be an upgrade to the Win9x family), followed by Windows XP.
What Microsoft did with Windows 95, was create first the OEM Service Releases (OSR2 actually worked pretty well, at least for me), then Win98 and WinME. Windows 2000 was developed out of NT 4.0, which in turn came from Windows NT 3.51, the origins of which can be traced to the old days of OS/2.
And I actually liked OS/2. PM has (had) its own ways of doing things (like the Shift, Alt and Ctrl mouse-drag combinations to copy, move or create a shadow of a file, which were dependent on the source and destination media), but the system was actually pretty usable. It really is too bad that it failed. I still have an OS/2 Warp 3 CD plus the Bonus Pak around here somewhere along with OS/2 2.11 floppies... might install it some day just for the heck of it...
Don't you think that is something that could be called a Win-Win situation? I can't help but think back on the days when one could create their own WIN.COM file...
Well...
I personally believe that software should written be for the user, not someone else. But the above quote after all doesn't deal with use of the software. It only deals with activation, a one-time-per-installation process.
Now, what if I want to reinstall my copy of Photoshop in ten years? Can Adobe guarantee that I will be able to do that? In 20 years? If they cannot guarantee that, it is indeed hindering my ability to use the software, regardless of marketing hype.
No, thanks. I will stick with software that doesn't require activation, and doesn't limit by design what I can do with it.
And unlike Linux, it's not just the kernel... it's got its own shell, programming language and God knows what else. It even includes a built-in text editor, lo and behold!
The difference is that fax machines, cars and many other "closed" appliances in the everyday world work according to very well defined principles. The protocols used for transfering a fax over the phone lines is documented and no one will prosecute you if you make an alternative implementation of it. Cars just get you from one place to another, sometimes with bells and whistles, but that's what they do. So do bicycles, buses, trains, and a number of other means of transportation.
In contrast, closed source software is very rarely documented enough so that complete interoperability is easily achievable. Imagine if only vehicles with a particular wheel spacing were usable on public roads, but the information on the necessary spacing was secret and deducing it from available, usable cars was punishable with severe fines or even jail time? That, in my opinion, is a better analogy than just saying that governments should not mandate use of one particular piece of technology simply because another exists.
The government has several functions, two of which are: (a) serve the public, and (b) preserve documents for the future. Microsoft, which must be said to be the most common provider of proprietary desktop software, has a tendency to change their file formats every few major revisions, sometimes more often than that. This wouldn't be so much of a problem if there was a reliable means of reading the old file formats in a new version, but unfortunately that is far from always the case. Coupled with a potential inability to use old software in a year, a few years or even a few decades (mandated software activation, on-line forced upgrades and the "Trusted" Computing Platform), this may grow into a very serious problem. When converting the documents, can you guarantee that no information or formatting is lost or altered? I wouldn't want to bet on it.
Open source software will not automatically mean that old documents can be read. But if the lowest level specification for a file format, the source code for the application used to create it, is freely available, that means that one is always guaranteed the possibility of creating tools to read and/or convert the data to another format in the future. (At least as long as the source code, or at the very least extensive documentation of the file format, is preserved.)
So, I think that saying that mandating government use of open source software is wrong because mandating government use of cars is wrong, is a rather flawed logic. As I understand it most governments are already required to use the best, most cost effective tool for the job. "Best" in terms of handling documents subject to long term preservation requirements would seem to include the possibility to read those documents in both five, ten and fifty or even a hundred years, with a minimum of hassle.
Just like at times, a first class jet airline ticket might be more cost effective than a second class train ticket for getting from one place to another.
How many will remember who SCO really was, in ten years, and what they did?
Unlike Sweden, which has a right-leg left-wing government.
Sure! "You can trust us, because we don't trust you." (Isn't that how it is in the NGSCB (Palladium) world, though?)
The problem is that it is always easier to be insecure than it is to be secure. If you want to be secure, you need to consider every possible attack vector. An attacker, or anyone else wanting to prove your systems insecure, only need to find one attack vector that you didn't think of. Myself, I keep my passwords on my Palm Vx, encrypted (using software specifically designed for the purpose and released under the GPL) and protected using a single "master password". I remember my passwords most of the time, but this gives me a chance to refresh my memory if I can't, without using the (often horribly insecure) "password recovery" mechanisms provided by some systems. Since the Palm Vx is almost always right with me - in my pocket or next to me on the desk - and the data is encrypted while in storage and only decrypted upon accessing it, I believe this is a fair compromise between security and convinience. And no, I don't use completely random passwords, either. They might work if you have just one that you need to remember. Anything beyond that, it's a lost cause. Better to come up with a good password that you can remember, than an excellent that you have to write down and refer to the note every time you need it.
I believe that 127.59.16.233 will work... :)