Q: Why the jump from 4 to 7? . . . I think it's clear that some other distributions inflated their version numbers for marketing purposes, and I've had to field (way too many times) the question "why isn't yours 6.x" or worse "when will you upgrade to Linux 6.0" which really drives home the effectiveness of this simple trick. . . Sorry if I haven't been enough of a purist about this. I promise I won't inflate the version number again (unless everyone else does again;)
> I'm surprised by the majority here that is against this. What kind of nerds exactly are you?
We're the kind of nerds that understand that almost all computers are based on bits - binary digits - capable of holding exactly 2 values.
We're the kind of nerds that understand that a fixed number of bits - as found in the registers and address buses of bit-based-computers, and in the data items used in uncountable storage media and file formats - can store, or address, exactly some power-of-2 number of values.
We're the kind of nerds that understand *why* binary quantities are the fundamental units of computing, and why the binary system is the natural and proper way to work with them.
People developing analog, quantum, or decimal-based systems should certainly do whatever's right for them, but for the vast majority of systems, binary is what's right.
Ok. So when the next gruesome, mutilated corpse turns out to be your own child... you won't go and bitch about it, right? After all, this is life and it sucks.
> Moreover, they would take ownership of not just what they paid for, but also my changes leading up to this moment,...
And they're prepared to compensate you for that, right?
Calculate how many hours you've put into the work you've done so far. Multiply that by minimum wage (assuming you're in the U.S.; currently $6.55 per hour). Are they planning on paying you anywhere near that amount of money for the already completed work, above and beyond your wages covering future work?
The Shuttle flight software, and the fire control system and missile firmware you worked on, may very well be the best engineered, highest quality software on the planet. (I hope so). It may be as close to perfection as is humanly possible. But if you think there are zero bugs in it, I guarantee that you are mistaken.
No bugs at all in the requirement specs, the design, code, development tools, software interfaces, hardware interfaces? No bugs at all in the firmware on all the controller boards and interface cards? No floating-point loss of precision issues in any of the math code? No latent issues with bus traffic timing? No potential for unsafe data usage across rate groups in the face of task overruns?
No unusual operational scenarios not properly accounted for in the failure recovery algorithms? No gaps in the test program that verifies all these things?
There is no nontrivial software with zero bugs. Humans are not capable of creating it. The systems are too large and complex, and there are far too many kinds of potential errors, for even the most talented and disciplined team, following the most rigorous engineering process, to produce zero error systems.
I bought CF lamps when they started becoming available around here. Imagine my surprise when I found that a "100 watt equivalent" CF lamp didn't produce enough light to read by. Literally, a magazine held 2 feet from my face could not be read. The CF produced maybe the same illumination as a 40 watt incandescent light bulb. I'll give it another try sometime, but for now, my house has incandescent bulbs, and I can see fine.
This is kind of an inverse GPL for patents. I would say that IBM has understood the strategy of the GPL at a deep level and applied it in a completely novel (some might say, perverted) way. Frankly, it's brilliant.
There are (at least) two ways IBM can apply such a meta-patent. (1) They can use it to become the only legally permitted patent troll in the United States. (2) They can use it to sue existing patent trolls out of business on the grounds that those trolls are violating IBM's patent on patent trolling.
I'm willing to give them the benefit of the doubt on this one. Time will tell.
http://en.wikipedia.org/wiki/De_facto Agreed, but... It doesn't matter whether the eventual final standard differs from the currently implemented specification. My current 802.11n-like wireless networking gear works fine *now* and will continue to do so until it breaks or gets thrown away. At some point the standard may be ratified, and chances are that new gear produced using the final standard will be 99% backward compatible with my current gear. To sum up: if it walks like 802.11n, and quacks like 802.11n, then it's 802.11n to me.
> Personally, I would like my desktop and my window manager to get out of my way as soon as I log in. Absolutely. That's the bottom line.
> A system with sensible defaults should Just Work(tm). Unfortunately the developers never consult me when deciding on the defaults. That's why decent software has options/preferences/settings/whatever - the developers realize that the user has the best understanding of their own needs.
> The minute I must configure my computer, it is not doing what I want. Lack of decent configureability just means it's configured to someone else's preference and that person has no idea what I need. Just put the options in there - people who need them will be grateful, people who don't won't be bothered by them.
Incidentally, that's why I still use the SeaMonkey browser (i.e. Mozilla) instead of Firefox. Every time I try Firefix it's the same thing - looks great, but what happened to all the Preference options? Why can't I set it up to my exact needs the way I can with Seamonkey? It's even worse because the options were there originally, but were purposely removed - I can only assume, to dumb down the user experience.
I think you're missing the point here. The enforcement of freedom is the enforcement of the opportunity to act. Big difference. The person who has the freedom is the one that decides whether to take advantage of that opportunity.
"You know, we already tax most of the things we do for pleasure... drinking, smoking. What if we taxed... you know, thingy?" "Thingy?" "Yes, you know, thingy." "I'm afraid I don't. Could you please be more explicit?" " You know - thingy!"
Ultima II (on the Apple ][) had a somewhat similar bug. Can't remember the exact details but performing a certain action while on a ship would create a *duplicate* ship right beside the one you were on. It was dead simple to do. You could bridge continents together without too much work.
The Shuttle does indeed have two sets of flight software,
Primary Avionics Systems Software (PASS), and
Backup Flight System (BFS).
During critical phases of flight, PASS is loaded on four of the GPCs and BFS is loaded on the fifth. BFS doesn't have all the capabilities of PASS - it is intended to take over in case of an emergency.
Slashdot Mirror
I'm no audiophile, but I would have married my college roommate's NAD 2200 if the law allowed it. That thing was fine.
Preach brother, preach! +1.
Wizardry? Ultima II? Original Adventure? Hello???
Sure do.
http://www.slackware.com/faq/do_faq.php?faq=general#0
Q: Why the jump from 4 to 7? ;)
. . . I think it's clear that some other distributions inflated their version numbers for marketing purposes, and I've had to field (way too many times) the question "why isn't yours 6.x" or worse "when will you upgrade to Linux 6.0" which really drives home the effectiveness of this simple trick. . . Sorry if I haven't been enough of a purist about this. I promise I won't inflate the version number again (unless everyone else does again
At least Pat copped to it up front. IIRC "we promise not to it again unless everyone else does it first".
+1 Insightful
> I'm surprised by the majority here that is against this. What kind of nerds exactly are you?
We're the kind of nerds that understand that almost all computers are based on bits - binary digits - capable of holding exactly 2 values.
We're the kind of nerds that understand that a fixed number of bits - as found in the registers and address buses of bit-based-computers, and in the data items used in uncountable storage media and file formats - can store, or address, exactly some power-of-2 number of values.
We're the kind of nerds that understand *why* binary quantities are the fundamental units of computing, and why the binary system is the natural and proper way to work with them.
People developing analog, quantum, or decimal-based systems should certainly do whatever's right for them, but for the vast majority of systems, binary is what's right.
Some unknown Slashdotter said it best... when this issue comes up, it always gets resolved in the way that makes the numbers come out bigger.
When did Tim Berners-Lee go to the dark side?
People don't like assholes, film at 11...
Ok. So when the next gruesome, mutilated corpse turns out to be your own child... you won't go and bitch about it, right? After all, this is life and it sucks.
> Moreover, they would take ownership of not just what they paid for, but also my changes leading up to this moment, ...
And they're prepared to compensate you for that, right?
Calculate how many hours you've put into the work you've done so far. Multiply that by minimum wage (assuming you're in the U.S.; currently $6.55 per hour). Are they planning on paying you anywhere near that amount of money for the already completed work, above and beyond your wages covering future work?
I know which way I'm betting.
The Shuttle flight software, and the fire control system and missile firmware you worked on, may very well be the best engineered, highest quality software on the planet. (I hope so). It may be as close to perfection as is humanly possible. But if you think there are zero bugs in it, I guarantee that you are mistaken.
No bugs at all in the requirement specs, the design, code, development tools, software interfaces, hardware interfaces? No bugs at all in the firmware on all the controller boards and interface cards? No floating-point loss of precision issues in any of the math code? No latent issues with bus traffic timing? No potential for unsafe data usage across rate groups in the face of task overruns? No unusual operational scenarios not properly accounted for in the failure recovery algorithms? No gaps in the test program that verifies all these things?
There is no nontrivial software with zero bugs. Humans are not capable of creating it. The systems are too large and complex, and there are far too many kinds of potential errors, for even the most talented and disciplined team, following the most rigorous engineering process, to produce zero error systems.
I don't want MY health data "set free"...
I bought CF lamps when they started becoming available around here. Imagine my surprise when I found that a "100 watt equivalent" CF lamp didn't produce enough light to read by. Literally, a magazine held 2 feet from my face could not be read. The CF produced maybe the same illumination as a 40 watt incandescent light bulb. I'll give it another try sometime, but for now, my house has incandescent bulbs, and I can see fine.
This is kind of an inverse GPL for patents. I would say that IBM has understood the strategy of the GPL at a deep level and applied it in a completely novel (some might say, perverted) way. Frankly, it's brilliant.
There are (at least) two ways IBM can apply such a meta-patent. (1) They can use it to become the only legally permitted patent troll in the United States. (2) They can use it to sue existing patent trolls out of business on the grounds that those trolls are violating IBM's patent on patent trolling.
I'm willing to give them the benefit of the doubt on this one. Time will tell.
http://en.wikipedia.org/wiki/De_facto
Agreed, but... It doesn't matter whether the eventual final standard differs from the currently implemented specification. My current 802.11n-like wireless networking gear works fine *now* and will continue to do so until it breaks or gets thrown away. At some point the standard may be ratified, and chances are that new gear produced using the final standard will be 99% backward compatible with my current gear. To sum up: if it walks like 802.11n, and quacks like 802.11n, then it's 802.11n to me.
+1 Says it all
I agree with your basic premise, but...
> Personally, I would like my desktop and my window manager to get out of my way as soon as I log in.
Absolutely. That's the bottom line.
> A system with sensible defaults should Just Work(tm).
Unfortunately the developers never consult me when deciding on the defaults. That's why decent software has options/preferences/settings/whatever - the developers realize that the user has the best understanding of their own needs.
> The minute I must configure my computer, it is not doing what I want.
Lack of decent configureability just means it's configured to someone else's preference and that person has no idea what I need. Just put the options in there - people who need them will be grateful, people who don't won't be bothered by them.
Incidentally, that's why I still use the SeaMonkey browser (i.e. Mozilla) instead of Firefox. Every time I try Firefix it's the same thing - looks great, but what happened to all the Preference options? Why can't I set it up to my exact needs the way I can with Seamonkey? It's even worse because the options were there originally, but were purposely removed - I can only assume, to dumb down the user experience.
I think you're missing the point here. The enforcement of freedom is the enforcement of the opportunity to act. Big difference. The person who has the freedom is the one that decides whether to take advantage of that opportunity.
"You know, we already tax most of the things we do for pleasure... drinking, smoking. What if we taxed... you know, thingy?" "Thingy?" "Yes, you know, thingy." "I'm afraid I don't. Could you please be more explicit?" " You know - thingy!"
PJ won't run out of evildoers to investigate on Groklaw after SCO finally goes under.
We have a winner.
Ultima II (on the Apple ][) had a somewhat similar bug. Can't remember the exact details but performing a certain action while on a ship would create a *duplicate* ship right beside the one you were on. It was dead simple to do. You could bridge continents together without too much work.
The Shuttle does indeed have two sets of flight software, Primary Avionics Systems Software (PASS), and Backup Flight System (BFS). During critical phases of flight, PASS is loaded on four of the GPCs and BFS is loaded on the fifth. BFS doesn't have all the capabilities of PASS - it is intended to take over in case of an emergency.