It depends, because failures can lead to mis-attribution of the problem. Say any of Internet Explorer, Modem, ISP, DNS, etc. fail while a customer is trying to access Amazon.com but, by chance, the failure is over when they access Barnesandnoble.com. In this case, the person will probably think something is wrong with Amazon.com, when that isn't true (computers really are black boxes to nearly everyone).
Conversely, POV-Ray is more mature and more portable. For example, Yafray required the very most recent point release of GCC when 0.0.6 came out...sigh). Also, POV-Ray comes with documentation (fancy that!).
Yet it only became popular in the last decade. Remember, Sun was a *workstation* company before the mid-1990s. The big-ass UNIX servers we see today really are a fairly recent development--before them were the mainframes. I think it is fair to say that SMP really is only recently proven, regardless of how long various implementations existed.
Whatever you do, don't ever take on more unsecured educational debt (e.g., loans for tuition) than you can pay off within three or four years after graduating. Too many people go to an expensive private school blinded by idealism to only find themselves imprisoned to their debt for ten or more years. This is serious stuff, because school debt is a very huge distraction from much more important post-school things like buying a car or home or starting a family. Don't fall into the loan trap!!!
I'm not made of money, and sometimes buying a used game, even one without a manual, for really cheap is exactly what I need for my gaming habit. At least I'm not "stealing" the game by copying the disk and paying almost nothing at all. Just be happy that I'm helping the game store stay open in these days of Amazon.com et. al.
The game developers really need to understand that the revenue generated from used game sales is not revenue that would otherwise go to them. Well, it might if they sell their new games for $15, but they don't. IMO, they are just a bunch of whiners who sound like the losers from the RIAA: "But but were losing 15 quadrillion dollars in sales to these people too cheap to buy our new expensive games!" Give it up, guys.
Oh, so now I can purchase MYOB? Adobe software? Macromedia Software? Peach Tree Accounting?
For this stuff, go buy a Mac and get over it.
As far as corporations whose main applications are written in-house, JDS could be a perfect deployment platform for client-side operations. If you hired a secretary to keep the office running smoothly, does he/she really need more than JDS already has (StarOffice, Mozilla, etc.)? How about a tech-support person, who probably uses a web-based interface anyway? How about an engineer whose tools work under UNIX/Linux anyway? How about a system administration console for a server room?
For well under a thousand dollars per seat (including adequate PC hardware plus years of Sun support), you get a job done. It also allows saving the time wasted by chasing/testing/deploying the ceaseless patches from Microsoft that are always a case of russian roulette (round and round she goes, will she boot, nobody knows).
SunCom is mediocre even beyond your itemized billing struggle. I hope your girlfriend can actually cancel her plan when that time comes without getting bills for the next several months. Also, their phone displays were very vague about subtle variants of "in network"...oh boy, did I appreciate that $160 bill! Do they still do TDMA only? I wonder if being part of AT&T is another liability for them. I couldn't find a decent TDMA/Palm combo phone to save my life; everyone else is CDMA/GSM.
Google Zeitgeist still reports Windows 98 at a solid 21% and Windows 95 still gets 1%. This isn't a perfect sampling, but it is telling that there a lot of Win 98 holdouts, probably due to the boom of the late 1990s. With Win 98 still so popular, software vendors would have to think long and hard whether to support only XP and/or 2000.
While Win 98 does feel rather primitive compared to more modern stuff, it and Windows 2000 are the last big releases that dont have all the activation nonsense.
Idea: how about carrying your wallet in the front pocket and putting a nice big wad of Brie in the back pocket. The expression on the pickpocket's face: priceless.
There's nothing wrong in looking a little ratty while in a city. Wear old jeans, no watch or jewelry, a t-shirt, etc. Put the laptop in an old backpack, etc. No one need know you actually have money or possessions. I've heard too many stories about people getting killed for their car/rolex/diamonds and it just seems that looking pretty isn't worth one's life.
Yes, we can't neglect a dog for fear of imprisonment, yet nothing stops us from creating unwanted human beings that suffer under a heartless social services system and arbitrary welfare. It seems politics just stops cold when it comes to regulating families. This isn't really a bad thing (slippery slopes, eugenics, etc.), but it really creates situations where dogs really are treated better than our own children. Kinda sad.
The precision of a float or double isn't defined anywhere in the ANSI spec.
I assume this is specified in Java though.
Java is very thorough in specifying the precision for its types. C is not, but the question above was regarding the "epsilon" that Perl uses for fudging equality. In C, it is still bad practice to use "==" for floating point numbers.
(3.0 can be represented exactly in IEEE-like formats)
This isn't the point, rather the problem is using the equality operator for comparing floating point numbers. The odds that a floating point variable would actually be exactly equal to an integer are pretty low, in general.
Going with the Social Services costs argument, then, yes. There are way too many unwanted children around. All recreational sex should require a condom or other effective birth control. Only when parents decide they want the responsibility of a child do they go without birth control.
on my computer 16 9s will suffice whereas in reality you would need an infinite number of 9s
Yes, but is that number of nines variable from architecture to architecture or is the epsilon "fudge factor" built into Perl defined in the language's specification? I know this is nit-picking, but these are questions that wouldn't have to be asked in C or Java.
Rejecting convenience because someday you may have to work without said convenience does not seem to me to be a logical behavior.
It is logical. How much trouble does Microsoft's "convenience" need to cause before people learn to not execute their e-mail attachments? Bash's "export VAR = value" syntax is convenient, but it is broken on all genuine Bourne shells (e.g., most UNIX). Configure scripts are convenient, but they break very badly when they break making debugging near impossible. Auto-completion for text is convenient for some people who can't type so they never need to learn how. It is just a matter of debate where convenience ends and luxury begins, and my argument is that mixing floating point and integer types in equality expressions is just a luxury that leads to a compacency that can haunt future debugging efforts. Quite honestly, I think programming languages stopped evolving in usefulness and productivity once garbage collection became popular (memory management is by far the worst problem in C). Beyond getting rid of assembly coding and the advent of garbage collection, most everything else has been incremental improvements rather than true leaps forward.
You obviously never studied mathematics, 3 and 2.999... are exactly the same.
Programming != what is theoretically true in math. All I got is these 32 or 64 or 80 bits and there isn't any mathemetician of any IQ that can make 2.9999... in IEEE floating point equal 3.0 exactly.
That's unfortunate. Hopefully, Perl programmers can unlearn this habit when they work with just about every other programming language ever. You just can't get around the fact that floating point numbers are constrained to a certain number of bits in RAM, unless special math libraries are used (and those libraries are not used often).
This is my opinion, of course, but the fact their installer, for example, has been so inflexible for so long and always second-guessing the user that who outside of corporate red tape would use Red Hat for anything? I've worked with computers for over a decade and still have to use fdisk to figure out a dual-boot setup under Red Hat, and sometimes their installer still thinks it knows better (oops, there goes Windows...again). When they say to back up before the install, they mean it.
Give me Slackware, Debian, etc. any day before Red Hat/Fedora. Red Hat really was a big deal in the 1990's, but, now, I've moved on.
Even numerically, good programmers assume this to be always false. Why? Because it isn't deterministic whether "3.0" is actually 2.999999... or 3.0000...0001. Also, "==" does not imply "almost"; it is exact.
This might be a way for people near slow rivers to use the river water since they cannot use it for hydroelectric.
There is a tree-hugger for every type of pollution...even thermal pollution in rivers. Sigh. Still, using solar heat to drive a turbine is a good idea, and I think there are steam-driven ones somewhere (I recall picture of a desert with many many mirrors).
Thus in the long run a logarithmic decrease cannot overcome an exponential increase.
This is true, which means the only long-term solutions are either solar or fusion. Any fossil fuel-based plan is doomed from the start. Solar can be anything from using ocean thermals to tapping the Sun's core itself, but it has to use the Sun's output in real-time.
Slashdot Mirror
Some people will just come back and order later.
It depends, because failures can lead to mis-attribution of the problem. Say any of Internet Explorer, Modem, ISP, DNS, etc. fail while a customer is trying to access Amazon.com but, by chance, the failure is over when they access Barnesandnoble.com. In this case, the person will probably think something is wrong with Amazon.com, when that isn't true (computers really are black boxes to nearly everyone).
Conversely, POV-Ray is more mature and more portable. For example, Yafray required the very most recent point release of GCC when 0.0.6 came out...sigh). Also, POV-Ray comes with documentation (fancy that!).
"SMP technology" proved its worth decades ago.
Yet it only became popular in the last decade. Remember, Sun was a *workstation* company before the mid-1990s. The big-ass UNIX servers we see today really are a fairly recent development--before them were the mainframes. I think it is fair to say that SMP really is only recently proven, regardless of how long various implementations existed.
and yes, I do work for a university.
So do I (slaps stamp on latest loan payment).
Whatever you do, don't ever take on more unsecured educational debt (e.g., loans for tuition) than you can pay off within three or four years after graduating. Too many people go to an expensive private school blinded by idealism to only find themselves imprisoned to their debt for ten or more years. This is serious stuff, because school debt is a very huge distraction from much more important post-school things like buying a car or home or starting a family. Don't fall into the loan trap!!!
I'm not made of money, and sometimes buying a used game, even one without a manual, for really cheap is exactly what I need for my gaming habit. At least I'm not "stealing" the game by copying the disk and paying almost nothing at all. Just be happy that I'm helping the game store stay open in these days of Amazon.com et. al.
The game developers really need to understand that the revenue generated from used game sales is not revenue that would otherwise go to them. Well, it might if they sell their new games for $15, but they don't. IMO, they are just a bunch of whiners who sound like the losers from the RIAA: "But but were losing 15 quadrillion dollars in sales to these people too cheap to buy our new expensive games!" Give it up, guys.
Really? Where in SPARC v9 + VIS is PicoJava?
Oh, so now I can purchase MYOB? Adobe software? Macromedia Software? Peach Tree Accounting?
For this stuff, go buy a Mac and get over it.
As far as corporations whose main applications are written in-house, JDS could be a perfect deployment platform for client-side operations. If you hired a secretary to keep the office running smoothly, does he/she really need more than JDS already has (StarOffice, Mozilla, etc.)? How about a tech-support person, who probably uses a web-based interface anyway? How about an engineer whose tools work under UNIX/Linux anyway? How about a system administration console for a server room?
For well under a thousand dollars per seat (including adequate PC hardware plus years of Sun support), you get a job done. It also allows saving the time wasted by chasing/testing/deploying the ceaseless patches from Microsoft that are always a case of russian roulette (round and round she goes, will she boot, nobody knows).
SunCom is mediocre even beyond your itemized billing struggle. I hope your girlfriend can actually cancel her plan when that time comes without getting bills for the next several months. Also, their phone displays were very vague about subtle variants of "in network"...oh boy, did I appreciate that $160 bill! Do they still do TDMA only? I wonder if being part of AT&T is another liability for them. I couldn't find a decent TDMA/Palm combo phone to save my life; everyone else is CDMA/GSM.
Google Zeitgeist still reports Windows 98 at a solid 21% and Windows 95 still gets 1%. This isn't a perfect sampling, but it is telling that there a lot of Win 98 holdouts, probably due to the boom of the late 1990s. With Win 98 still so popular, software vendors would have to think long and hard whether to support only XP and/or 2000.
While Win 98 does feel rather primitive compared to more modern stuff, it and Windows 2000 are the last big releases that dont have all the activation nonsense.
France
Idea: how about carrying your wallet in the front pocket and putting a nice big wad of Brie in the back pocket. The expression on the pickpocket's face: priceless.
"I've switched my iPod headphones...."
There's nothing wrong in looking a little ratty while in a city. Wear old jeans, no watch or jewelry, a t-shirt, etc. Put the laptop in an old backpack, etc. No one need know you actually have money or possessions. I've heard too many stories about people getting killed for their car/rolex/diamonds and it just seems that looking pretty isn't worth one's life.
The consistency of our laws.
Yes, we can't neglect a dog for fear of imprisonment, yet nothing stops us from creating unwanted human beings that suffer under a heartless social services system and arbitrary welfare. It seems politics just stops cold when it comes to regulating families. This isn't really a bad thing (slippery slopes, eugenics, etc.), but it really creates situations where dogs really are treated better than our own children. Kinda sad.
The precision of a float or double isn't defined anywhere in the ANSI spec.
I assume this is specified in Java though.
Java is very thorough in specifying the precision for its types. C is not, but the question above was regarding the "epsilon" that Perl uses for fudging equality. In C, it is still bad practice to use "==" for floating point numbers.
(3.0 can be represented exactly in IEEE-like formats)
This isn't the point, rather the problem is using the equality operator for comparing floating point numbers. The odds that a floating point variable would actually be exactly equal to an integer are pretty low, in general.
Should we legislate condom use?
Going with the Social Services costs argument, then, yes. There are way too many unwanted children around. All recreational sex should require a condom or other effective birth control. Only when parents decide they want the responsibility of a child do they go without birth control.
...no-one has repeated that particular mistake since.
Then this thread of discussion must have never happened, either...
on my computer 16 9s will suffice whereas in reality you would need an infinite number of 9s
Yes, but is that number of nines variable from architecture to architecture or is the epsilon "fudge factor" built into Perl defined in the language's specification? I know this is nit-picking, but these are questions that wouldn't have to be asked in C or Java.
Rejecting convenience because someday you may have to work without said convenience does not seem to me to be a logical behavior.
It is logical. How much trouble does Microsoft's "convenience" need to cause before people learn to not execute their e-mail attachments? Bash's "export VAR = value" syntax is convenient, but it is broken on all genuine Bourne shells (e.g., most UNIX). Configure scripts are convenient, but they break very badly when they break making debugging near impossible. Auto-completion for text is convenient for some people who can't type so they never need to learn how. It is just a matter of debate where convenience ends and luxury begins, and my argument is that mixing floating point and integer types in equality expressions is just a luxury that leads to a compacency that can haunt future debugging efforts. Quite honestly, I think programming languages stopped evolving in usefulness and productivity once garbage collection became popular (memory management is by far the worst problem in C). Beyond getting rid of assembly coding and the advent of garbage collection, most everything else has been incremental improvements rather than true leaps forward.
You obviously never studied mathematics, 3 and 2.999... are exactly the same.
Programming != what is theoretically true in math. All I got is these 32 or 64 or 80 bits and there isn't any mathemetician of any IQ that can make 2.9999... in IEEE floating point equal 3.0 exactly.
3.0 is identical to 3 or you get your money back
That's unfortunate. Hopefully, Perl programmers can unlearn this habit when they work with just about every other programming language ever. You just can't get around the fact that floating point numbers are constrained to a certain number of bits in RAM, unless special math libraries are used (and those libraries are not used often).
This is my opinion, of course, but the fact their installer, for example, has been so inflexible for so long and always second-guessing the user that who outside of corporate red tape would use Red Hat for anything? I've worked with computers for over a decade and still have to use fdisk to figure out a dual-boot setup under Red Hat, and sometimes their installer still thinks it knows better (oops, there goes Windows...again). When they say to back up before the install, they mean it.
Give me Slackware, Debian, etc. any day before Red Hat/Fedora. Red Hat really was a big deal in the 1990's, but, now, I've moved on.
"3.0" == "3". Does this return true or not?
Even numerically, good programmers assume this to be always false. Why? Because it isn't deterministic whether "3.0" is actually 2.999999... or 3.0000...0001. Also, "==" does not imply "almost"; it is exact.
This might be a way for people near slow rivers to use the river water since they cannot use it for hydroelectric.
There is a tree-hugger for every type of pollution...even thermal pollution in rivers. Sigh. Still, using solar heat to drive a turbine is a good idea, and I think there are steam-driven ones somewhere (I recall picture of a desert with many many mirrors).
Thus in the long run a logarithmic decrease cannot overcome an exponential increase.
This is true, which means the only long-term solutions are either solar or fusion. Any fossil fuel-based plan is doomed from the start. Solar can be anything from using ocean thermals to tapping the Sun's core itself, but it has to use the Sun's output in real-time.