We could compare Linux running on a WRT54G versus the cost of, say, a dual CPU P4 XEON system with 4 gbs RAM, SCSI array, redundant everything, and dual 19" LCD monitors.
...test this by putting twenty million random emails through the system, aimed at two million mailboxes plus 5% wrong addresses just in case. The WRT54G wouldn't even have room for the address mappings, the MS-Exchange box would be a puddle on the floor, and the IBM box would still be chugging along jess fahn.
If testing as a webserver, I think putting a pair of $500 whiteboxes up against the IIS monster server would make the point nicely, no need to go all out here.
If testing as a firewall, by all means use the WRT54G, it'll hammer the competition in bang for the buck.
Now how about the rest of Microsoft's false advertising?
8 minutes from/. posting to digital smoking crater. (-:
I remember putting an article up on a 64kb (yes, b not B) link some years back, then it got found and posted to./; within about 1/2 an hour the link was grinding (./ was much smaller in those days, this is not my first./ ID) and within two hours it died completely (pings went into hyperspace). For three days.
So heavy was the traffic that taking the webserver down didn't make a noticeable difference. Even if they'd been able to get through, the DNS queries alone would have been enough to smash the link flat. Think "trying to fill a thimble from a wide-open firehose".
I remember a story from mainframe days where the sysop had carefully explained how to log in on a decwriter, feed labels, and run the mailing-list-printing program to some blonde several times, then shortly afterwards he noticed that she'd logged on, but wasn't running the program. Remembering the mistake she always made, he found the terminal's location on a site map, found the nearest telephone extension, rang up, waited a looong time for her to answer, then said "Hi, you need to do such-and-such" and was instantly rewarded with a piercing scream 'coz Ms Blonde was totally freaked out that someone knew here every move (cue twilight zone music) without being there.
I got similar results with a girlfriend of old. I rang her prior to going out together for the evening, and she took a long time answering. Since she lived in a one-bedroom flat at the time, the likely reasons for the delay were either being on the 'loo or in the shower, so I took a punt and greeted her with "Hi Gail! Ooh! You're all wet!" Human nature being what it is, she didn't stop to think "how could he see me down the 'phone", just squealed and dropped the handset.
This was from the days when mobiles were expensive analogue briefcase-sized monsters, so I couldn't have been peeking through a window.
...dear Enter key: you will hereinafter transmit "Enter, 'echo', lock screen, lock keyboard, 'give my account J.ACT', sleep 1, reset Enter key to default, unlock screen, Enter, unlock keyboard". Sent to the console.
are Hot Keys for thinks like Copying and Pasting really over rated
Yes. I generally use the mouse for that (swipe, middle-click) and reverting to MS-Windows is always such an inordinate PITA for needing the extra two steps (swipe with mouse, copy with keyboard or right-click-and-menu, click with mouse, paste with keyboard or right-click-and-menu).
That and a few other trinkets to let ${FAVOURITEGAME} run anywhere without the WINE crew having to constantly play BlackBox to try and figure out what bizarre new tack MS've taken with their code this month.
Microsoft's problem has always been over control. Billy boy is buying up patents hand over fist for the public good. No? I didn't really think so, either. When I see MS stop doing that and start seriously opening up stuff like DirectX that allows others access to large markets they've fenced off, then I'll start trusting them.
Start trusting, mind you, not bare my soul. They've been screwing the market over for more than a decade, and they won't turn that much corporate inertia on a dime.
The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.
Preferred form, tick. All modules, debatable - are libraries for your gonzoscript included? Interface definitions, probably not. Scripts, neither here nor there. But your gonzoscript compiler/runtime or interpreter is not "normally distributed" with anyone's "major components", so technically you'd have to distribute that as well. To get a grey area, you'd have to have achieved a reasonable "market" penetration with gonzoscript - some hope of having it decreed "normally distributed... with the major components... of the operating system" by a judge.
Up to that point, you have an Open Source scripting example for gonzoscript - a trivial one, too - not a distinct Open Source program. Distributing examples with Visual BASIC didn't open source either the examples (go and read the copyrights on them) or VB.
In practical terms, an Open Source program written in VB is like the key to the toilet you get from some service stations, chained to a besser block. Yes, you have the key in your hand, but you're hardly able to pocket it or use it at will. Yes, the program can be argued to be fully Open Source because VB (or at least the runtime) can be reasonably argued to be "normally distributed" with the OS, and within the community of VB-capable platforms it functions as FOSS. But at some level it really is Clayton's FOSS and also at some level practically everybody realises this.
does OSS mean that it has to work wherever YOU want it to work?
If it's not portable, that limits its open-ness.
Of course, something like a dot-MSI maker ain't gonna be as widely useful running under Linux anyway, but even something as apparently silly as that means that someone compiling for MS-Windows using native-to-Linux tools can build an MSI using another native-to-Linux tool and push it across to his FTP server for his MS-Windows-using friends and testers to use.
It is easy to use COM to instantiate Word from your own code and manipulate documents throught the API, so ".doc format" is fully accessible and reusable from your own code, just as it would be if it was "open source".
No worries. Which APT or RPM repository should I look in for that? Or do I need to emerge it?
All that does if accepted as valid is effectively defer reasoning about the situation. Not very helpful.
There is indeed evidence of large-scale structure in the form of quantum redshifts for QSOs, but not of complexity. If there were high levels of complexity in the universe, we should be able to see precursors and aggregations of it at a scale perceptible to us. In short, if it's complex at one level, it is reasonable to expect complexity at all levels.
We have several theories about how self-replication could start. None of these are backed by experimental data. From there, the evidence for evolution is quite strong.
Er... theories aren't evidence?
It's kind of like saying, "If you'll cede me the existence of unicorns, the evidence for elves looks quite good." (-:
Short story is, multiply out the odds of forming any useful organics given all necessary amino acids in any required amounts versus all "natural" combinations (ie, only trialling arrangements known to be chemically possible in order to eliminate accusations of stretching the odds). Ignoring the constraints of distance (molecules getting to each other to interact) and gross structure, and contamination, you don't have anything like enough atoms or seconds in the known universe to even approach reasonable odds of a successful outcome.
I could flip a coin a couple billion times (assuming infinite time and patience).
Unfortunately, you don't have infinite time, only about 1E17 seconds (if you're a universe; about 2E9 if you're a human). You also don't have an unlimited number of coins, so parallelism won't help (enough; or put it this way, it falls short by many hundreds of orders of magnitude). You don't even have time to complete enough trials, let alone make decrees based on the results.
Slashdot Mirror
If testing as a webserver, I think putting a pair of $500 whiteboxes up against the IIS monster server would make the point nicely, no need to go all out here.
If testing as a firewall, by all means use the WRT54G, it'll hammer the competition in bang for the buck.
Now how about the rest of Microsoft's false advertising?
8 minutes from /. posting to digital smoking crater. (-:
./; within about 1/2 an hour the link was grinding (./ was much smaller in those days, this is not my first ./ ID) and within two hours it died completely (pings went into hyperspace). For three days.
I remember putting an article up on a 64kb (yes, b not B) link some years back, then it got found and posted to
So heavy was the traffic that taking the webserver down didn't make a noticeable difference. Even if they'd been able to get through, the DNS queries alone would have been enough to smash the link flat. Think "trying to fill a thimble from a wide-open firehose".
WayBack or Google's cache either. Reminds me of Scrat and the glacier.
I remember a story from mainframe days where the sysop had carefully explained how to log in on a decwriter, feed labels, and run the mailing-list-printing program to some blonde several times, then shortly afterwards he noticed that she'd logged on, but wasn't running the program. Remembering the mistake she always made, he found the terminal's location on a site map, found the nearest telephone extension, rang up, waited a looong time for her to answer, then said "Hi, you need to do such-and-such" and was instantly rewarded with a piercing scream 'coz Ms Blonde was totally freaked out that someone knew here every move (cue twilight zone music) without being there.
I got similar results with a girlfriend of old. I rang her prior to going out together for the evening, and she took a long time answering. Since she lived in a one-bedroom flat at the time, the likely reasons for the delay were either being on the 'loo or in the shower, so I took a punt and greeted her with "Hi Gail! Ooh! You're all wet!" Human nature being what it is, she didn't stop to think "how could he see me down the 'phone", just squealed and dropped the handset.
This was from the days when mobiles were expensive analogue briefcase-sized monsters, so I couldn't have been peeking through a window.
It must be the drugs.
Why not go whole hog if you're going to use a broken keyboard?
...dear Enter key: you will hereinafter transmit "Enter, 'echo', lock screen, lock keyboard, 'give my account J.ACT', sleep 1, reset Enter key to default, unlock screen, Enter, unlock keyboard". Sent to the console.
Shiny! (-:
...then the one true acronym would be apropros again, with swap-over-USB.
Obviously, the system is broken.
Chances of getting it fixed through conventional paths? Zippo. You then have three choices: live with it, break the law or break the system.
Case of been there, done that?
What keeps you out of sunny Georgia? Got a 15yo pregnant? Prefer mud on your dipstick? What?
I like this idea. Should drive the price of big Flash cards through the floor as well. (-:
Also, small differences can be important, CF lightning bolt vs lightning bug, and being unable to cross a chasm in two smaller jumps rather than one.
That and a few other trinkets to let ${FAVOURITEGAME} run anywhere without the WINE crew having to constantly play BlackBox to try and figure out what bizarre new tack MS've taken with their code this month.
Microsoft's problem has always been over control. Billy boy is buying up patents hand over fist for the public good. No? I didn't really think so, either. When I see MS stop doing that and start seriously opening up stuff like DirectX that allows others access to large markets they've fenced off, then I'll start trusting them.
Start trusting, mind you, not bare my soul. They've been screwing the market over for more than a decade, and they won't turn that much corporate inertia on a dime.
Try mine (non-IE users, tell your browser to lie about its ID if you want to see what he sees).
There you go. Naked women, oil within easy reach, what more could you ask? (-:
Preferred form, tick. All modules, debatable - are libraries for your gonzoscript included? Interface definitions, probably not. Scripts, neither here nor there. But your gonzoscript compiler/runtime or interpreter is not "normally distributed" with anyone's "major components", so technically you'd have to distribute that as well. To get a grey area, you'd have to have achieved a reasonable "market" penetration with gonzoscript - some hope of having it decreed "normally distributed
Up to that point, you have an Open Source scripting example for gonzoscript - a trivial one, too - not a distinct Open Source program. Distributing examples with Visual BASIC didn't open source either the examples (go and read the copyrights on them) or VB.
In practical terms, an Open Source program written in VB is like the key to the toilet you get from some service stations, chained to a besser block. Yes, you have the key in your hand, but you're hardly able to pocket it or use it at will. Yes, the program can be argued to be fully Open Source because VB (or at least the runtime) can be reasonably argued to be "normally distributed" with the OS, and within the community of VB-capable platforms it functions as FOSS. But at some level it really is Clayton's FOSS and also at some level practically everybody realises this.
Of course, something like a dot-MSI maker ain't gonna be as widely useful running under Linux anyway, but even something as apparently silly as that means that someone compiling for MS-Windows using native-to-Linux tools can build an MSI using another native-to-Linux tool and push it across to his FTP server for his MS-Windows-using friends and testers to use.
There is indeed evidence of large-scale structure in the form of quantum redshifts for QSOs, but not of complexity. If there were high levels of complexity in the universe, we should be able to see precursors and aggregations of it at a scale perceptible to us. In short, if it's complex at one level, it is reasonable to expect complexity at all levels.
It's kind of like saying, "If you'll cede me the existence of unicorns, the evidence for elves looks quite good." (-:
Short story is, multiply out the odds of forming any useful organics given all necessary amino acids in any required amounts versus all "natural" combinations (ie, only trialling arrangements known to be chemically possible in order to eliminate accusations of stretching the odds). Ignoring the constraints of distance (molecules getting to each other to interact) and gross structure, and contamination, you don't have anything like enough atoms or seconds in the known universe to even approach reasonable odds of a successful outcome.