Sure, I remember it. But great technology appearing somewhere outside of Xerox is the first sign that it will be popularized and we'll get cool stuff and Xerox will get to kick themselves. Again.
Bad DB: you lose couple of thousands of dolars, minimum.
try, "a couple hundred thousand," buying hardware and setting up a pilot. And by the time you are done, any salvage value is eaten up by calendar time, staff, and opportunity costs.
Since configuring a DB is an art,
Mind if I speak harshly for a minute? Am I the only fucking engineer on this website? Yes, it is a challenge to set up a good test of a database. That's why it is so fucking important to centralize the task and share some good benchmarks along with opensourcing the test conditions. The objections you raise are not objections: you have correctly identified what the benchmarking task *is*. That's the first step to solving it. Why can't you see that?
It's easy to hate a movie: we still allow reviews.
It's easier to suppress free speech: we still think it's a good idea to allow it.
It's easy to misconfigure database, but it's also a valid part of setting one up and valid to test. If there are ratings that are consistenly bad, people will turn to other ratings.
In any case, for those of us who believe in the free flow of information, and find competition to be the best way to achieve high quality high performance software and testing, my post concerned reasonable ways to circumvent the restrictions. You can feel free not to read the results.
And, if you stop to think for yourself, you'll realize that markets without information are not "free
markets"
read again what I wrote and see for yourself, I didn't say they were, though they *are* considered laissez-faire as I said. Actually, I never refer to "free markets". I refer to "free market theory" and "free market principles" but never "free markets". The proper terminology is more accurately "perfect markets", but "free as in markets" is more ingrained in the language.
Your objections fall into the category of "this could happen in the real world too" so they are valid things to test, and furthermore represent not insurmountable obstacles but simply engineering challenges, which, if you don't feel competent to solve, you should not be the one to work on.
They're not meaningless. Yes, they are open to manipulation and misinterpretation, and no "digest" can ever tell a whole story, but testing anything will always be just a sample.
What you said about DB benchmarking you could have said about any software. And by the same reasoning, you could never say any piece of software sucked because how could you know that further use or testing wouldn't reveal a configuration that would work better? But, of course, some software does suck, and you find out by trying it.
Perhaps you are noting that the biggest vendors and most expensive packages are all "pretty good" and that the differences between them are subtle. Well, benchmarks could be used to demonstrate that. Let a thousand benchmarks blossom. Experienced DB programmers and SAs are familiar with "standard-seeming" tasks that they've had trouble with in the past, or big tasks that are so common they simply need to work well. These things can be benchmarked, and some information is just better than no information at all. If the vendor feels the test is not accurate, let them give instructions on how to improve and publish better results.
the big DB companies all ban publishing benchmark results through terms in their license agreements. This is incredibly irritating because it undermines the free market principle of "perfect information" (markets with "asymetric information" are known to be ineffecient, something laissez-faire Randites never seem to learn). In pursuit of the public benefit of market efficiancy, wouldn't this be a way around these stupid rules:
put up a webserver with some CGIs that do some useful largescale things.
Implement the same things in several CGIs that communicate with different back-end databases.
Allow the public to come to the server and run and compare results (yes, you need some locking to stop them from interfering with one another).
If some member of the using public is a journalist, that journalist is free to publish the results because they are not party to the license.
You, a party to the license, are free to implement a website like this because it's just like any other website, albeit with a little extra redundancy.
The days of the
so-called Digerati are numbered (they will not be missed)
the digerati would certainly be missed were they to disappear, but I think think you parsed that sentence as "those days... will not be missed". I miss them.
It is a really good thing that more and more people are getting net access, just like it's a great thing that most people have telephones and TVs. There are a lot of benefits filtering up and down from universal access.
But it's also true that elitism can be a very good thing if you are one of the elite. It's nice not to get spam, it's nice to ask questions and get them answered by smart people, etc., i.e. the digerati are great digital neighbors. A lot of that has been swept away by the^H^H^H hoi polloi ("hoi" means "the" in Greek) as we see everywhere and especially on Slashdot. So, yes, universal access is good, and hordes of users supply the add revenue you need, but don't pretend there is something wrong with the digerati, they're great in some unique ways and we'll miss them in every way that they get harder to connect to.
consider it this way: the point you raise is a hypothetical. Therefore, it's not a theory... unless it in practice is shown to hold true. If you ever find yourself setting "theory" and "practice" against one another, stop.
There is a distinction between "hypothesis" and "practice", but where "theory" and "practice" deviate, there is suddenly no theory. "Theory" means "those ideas which we think we understand about reality because they match reality."
I think the distinction that's actually being drawn is between academics who understand what the word theory means and use it, and practitioners who never really grasped it but just recall that they last heard the word when they were in school.
Yes, do explain those things. I'm interested in knowing how you feel they apply in this context. It's been four days now and I've forgotten
most of the discussion attached to this post.
screw that, are you kidding? hit the "parent" link and you can read the context yourself. Then, if you truly do not see the relevance of basic statistics applied to his original anecdotal argument, I'd be most happy to edify you.
And gee, for someone who is so picky about analogies, you sure apply a looser standard to your own arguments: I never suggested
that "parents should be absolved of responsibility."
This is another insult/accusation pair
I do see on rereading how I interpreted what you wrote differently than you intended. But what I wrote was hardly an insult, and nor was it particularly an accusation unless suggesting that someone's argument is weak is an "accusation." Anyway, after reading this long-winded fulmination, not nearly the insult that you deserve. My original point was on target, and my followup to you was both concise and accurate. You continually stray from the thread, this time with your meta-personal attack claiming personal attack. If you can stick to the subject ("video games and violence and ratings and responsibility") I'd be happy to prove to your satisfaction that I'm both smarter and better educated than you.
kinda funny typo:
TH: Well, it went away for a while but itís popped microprocessor again recently.
apparently, the transcriber initially used the abbreviation "up" for "micro"-"processor" (there being no "mu" key on the keyboard). Then, a slip of the finger during the query-replace and we get "popped microprocessor"
Analogies are for illustrating ideas. I'm sure you got the idea I was illustrating. Analogies are never exactly the same as you [yawn] pointed out. Instead, you're supposed to go back and see if the idea applies to the original. It did. Do I need to explain statistical significance to you, or the importance of control groups? His argument was based on his own idea of the effect on him. That's stupid. Are you defending it?
And gee, for someone who is so picky about analogies, you sure apply a looser standard to your own arguments: I never suggested that "parents should be absolved of responsibility." However, I will suggest here that I believe that the children of neglectful parents should be protected by society at large as best we reasonably can. And even conscientious parents can't watch their children at all times.
It's common sense and the post I was criticizing contained none.
It's not MY job to be enforcing morality on your kid.
check with your employer's policy: it might very well be exactly your job. I suppose you think you should follow only the rules you want to.
get off your lazy ass
and if it was your job and somebody got off their lazy ass and came down and got you fired, we would no doubt have to listen to you whining on slashdot about it.
I've played Doom, Quake, Quake 2, Half-Life, Unreal, and Wolfenstein 3D since I was 10. Now, does this make me a
psychopathic killer? Did it encourage aggressive behavior?
I've driven drunk a bunch of times and till now I've never killed anybody, or even had an accident... so, drunk driving should no doubt be legal for me.
Women are better in space also because during self-gratification--c'mon, obviously important on long flights--hey, now cut it out, we're grownups planning a mission here, we have to talk about this stuff--as I was saying,
women conserve momentum by not firing a small(ish) rocket and
without the rocket there's no pesky
rocket fuel to dispose of
not too shabby? did you ever a play a lunar lander from that period? that's outstanding! of course, I read the article and it doesn't say anywhere how many plays he had at it.
the most famous PDP-8 emulator ran on the PDP-10. Because the PDP-8 was such a small machine (8bit!) there were only 4096 possible instructions including all combinations
of target addresses. So, the emulator implemented each opcode and operand combination as its own essentially hard-coded procedure. In this way, the emulator on a newer larger architecture was able to run programs faster than the original.
because it was a 36bit machine, it was particularly good for LISP, or better than the alternatives. Because you could divide a 36bit word into two 18-bit pointers, you could implement a LISP cons cell in one word, and thus handle cons cells fairly quickly. The same hack with competing 32bit machines, 16bit ob references just didn't provide a big enough data space.
The PDP-10 had its registers mapped into memory as the first 16 words starting with 0. But, being registers, they were really fast. So the text editor TECO (daddy to EMACS) would "compile" a search command into a tiny little program stored in the registers. Executing the code in the fast registers made for really quick (for the time) text searching.
Re:how fast were these things?
on
PDP-10 Revival
·
· Score: 1
your definitely write! only loosers use MHz. the one true measure of speed is the BogoMIP!
the PDP-11, a 16-bit byte addressable machine, is very definitely not a PDP-10, a 36-bit word addressible architecture.
they have almost nothing in common, except that Digitals OS's had some similarities, and as precursors to CP/M, can lay claim to being precursors to MS-DOS.
I think you are quibbling. What you said was, "granting him use of this mark"; "mark" is a term of art and means "trademark"
He was not granted use of that call sign as a trademark because only the PTO can do that. Yes, the work "mark" has other meanings, but when you are discussing trademarks, substituting in those other meanings will not help you to arrive at a sound legal argument.
because once you are granted the rights of a TM holder, you have the right to go after people. But you can't get the TM rights without engaging in commerce.
Slashdot Mirror
actually, I did a search, and you know what came up first? this page at the same AT&T group saying that Xerox uses this system.
Sure, I remember it. But great technology appearing somewhere outside of Xerox is the first sign that it will be popularized and we'll get cool stuff and Xerox will get to kick themselves. Again.
try, "a couple hundred thousand," buying hardware and setting up a pilot. And by the time you are done, any salvage value is eaten up by calendar time, staff, and opportunity costs.
Since configuring a DB is an art,
Mind if I speak harshly for a minute? Am I the only fucking engineer on this website? Yes, it is a challenge to set up a good test of a database. That's why it is so fucking important to centralize the task and share some good benchmarks along with opensourcing the test conditions. The objections you raise are not objections: you have correctly identified what the benchmarking task *is*. That's the first step to solving it. Why can't you see that?
It's easy to hate a movie: we still allow reviews.
It's easier to suppress free speech: we still think it's a good idea to allow it.
It's easy to misconfigure database, but it's also a valid part of setting one up and valid to test. If there are ratings that are consistenly bad, people will turn to other ratings.
In any case, for those of us who believe in the free flow of information, and find competition to be the best way to achieve high quality high performance software and testing, my post concerned reasonable ways to circumvent the restrictions. You can feel free not to read the results.
read again what I wrote and see for yourself, I didn't say they were, though they *are* considered laissez-faire as I said. Actually, I never refer to "free markets". I refer to "free market theory" and "free market principles" but never "free markets". The proper terminology is more accurately "perfect markets", but "free as in markets" is more ingrained in the language.
Your objections fall into the category of "this could happen in the real world too" so they are valid things to test, and furthermore represent not insurmountable obstacles but simply engineering challenges, which, if you don't feel competent to solve, you should not be the one to work on.
They're not meaningless. Yes, they are open to manipulation and misinterpretation, and no "digest" can ever tell a whole story, but testing anything will always be just a sample.
What you said about DB benchmarking you could have said about any software. And by the same reasoning, you could never say any piece of software sucked because how could you know that further use or testing wouldn't reveal a configuration that would work better? But, of course, some software does suck, and you find out by trying it.
Perhaps you are noting that the biggest vendors and most expensive packages are all "pretty good" and that the differences between them are subtle. Well, benchmarks could be used to demonstrate that. Let a thousand benchmarks blossom. Experienced DB programmers and SAs are familiar with "standard-seeming" tasks that they've had trouble with in the past, or big tasks that are so common they simply need to work well. These things can be benchmarked, and some information is just better than no information at all. If the vendor feels the test is not accurate, let them give instructions on how to improve and publish better results.
Slash should do it.
the digerati would certainly be missed were they to disappear, but I think think you parsed that sentence as "those days ... will not be missed". I miss them.
It is a really good thing that more and more people are getting net access, just like it's a great thing that most people have telephones and TVs. There are a lot of benefits filtering up and down from universal access.
But it's also true that elitism can be a very good thing if you are one of the elite. It's nice not to get spam, it's nice to ask questions and get them answered by smart people, etc., i.e. the digerati are great digital neighbors. A lot of that has been swept away by the^H^H^H hoi polloi ("hoi" means "the" in Greek) as we see everywhere and especially on Slashdot. So, yes, universal access is good, and hordes of users supply the add revenue you need, but don't pretend there is something wrong with the digerati, they're great in some unique ways and we'll miss them in every way that they get harder to connect to.
consider it this way: the point you raise is a hypothetical. Therefore, it's not a theory... unless it in practice is shown to hold true. If you ever find yourself setting "theory" and "practice" against one another, stop.
I think the distinction that's actually being drawn is between academics who understand what the word theory means and use it, and practitioners who never really grasped it but just recall that they last heard the word when they were in school.
screw that, are you kidding? hit the "parent" link and you can read the context yourself. Then, if you truly do not see the relevance of basic statistics applied to his original anecdotal argument, I'd be most happy to edify you.
This is another insult/accusation pairI do see on rereading how I interpreted what you wrote differently than you intended. But what I wrote was hardly an insult, and nor was it particularly an accusation unless suggesting that someone's argument is weak is an "accusation." Anyway, after reading this long-winded fulmination, not nearly the insult that you deserve. My original point was on target, and my followup to you was both concise and accurate. You continually stray from the thread, this time with your meta-personal attack claiming personal attack. If you can stick to the subject ("video games and violence and ratings and responsibility") I'd be happy to prove to your satisfaction that I'm both smarter and better educated than you.
apparently, the transcriber initially used the abbreviation "up" for "micro"-"processor" (there being no "mu" key on the keyboard). Then, a slip of the finger during the query-replace and we get "popped microprocessor"
rsh is not a shell itself, it provides access to a "remote shell". In the same way, a secure shell would be any shell that you got access to securely.
Which leads to my favorite new name for the open source version: what it does is "opens shell" or, OpensSH. There, everybody happy now?
And gee, for someone who is so picky about analogies, you sure apply a looser standard to your own arguments: I never suggested that "parents should be absolved of responsibility." However, I will suggest here that I believe that the children of neglectful parents should be protected by society at large as best we reasonably can. And even conscientious parents can't watch their children at all times. It's common sense and the post I was criticizing contained none.
check with your employer's policy: it might very well be exactly your job. I suppose you think you should follow only the rules you want to.
get off your lazy ass
and if it was your job and somebody got off their lazy ass and came down and got you fired, we would no doubt have to listen to you whining on slashdot about it.
I've played Doom, Quake, Quake 2, Half-Life, Unreal, and Wolfenstein 3D since I was 10. Now, does this make me a psychopathic killer? Did it encourage aggressive behavior?
I've driven drunk a bunch of times and till now I've never killed anybody, or even had an accident... so, drunk driving should no doubt be legal for me.
the US has PG-13 in addition to PG, and NC-17 in addition to R.
not too shabby? did you ever a play a lunar lander from that period? that's outstanding! of course, I read the article and it doesn't say anywhere how many plays he had at it.
that is a hack.
because it was a 36bit machine, it was particularly good for LISP, or better than the alternatives. Because you could divide a 36bit word into two 18-bit pointers, you could implement a LISP cons cell in one word, and thus handle cons cells fairly quickly. The same hack with competing 32bit machines, 16bit ob references just didn't provide a big enough data space.
The PDP-10 had its registers mapped into memory as the first 16 words starting with 0. But, being registers, they were really fast. So the text editor TECO (daddy to EMACS) would "compile" a search command into a tiny little program stored in the registers. Executing the code in the fast registers made for really quick (for the time) text searching.
:)
they have almost nothing in common, except that Digitals OS's had some similarities, and as precursors to CP/M, can lay claim to being precursors to MS-DOS.
I think you are quibbling. What you said was, "granting him use of this mark"; "mark" is a term of art and means "trademark" He was not granted use of that call sign as a trademark because only the PTO can do that. Yes, the work "mark" has other meanings, but when you are discussing trademarks, substituting in those other meanings will not help you to arrive at a sound legal argument.
because once you are granted the rights of a TM holder, you have the right to go after people. But you can't get the TM rights without engaging in commerce.