Last time I installed 95 and/or 98 with Samba (Sorry "Microsoft Sharing") nothing was shared at all by default. Yes, it is automatically bound to the TCP/IP of every networking device you have but that is a different issue.
Are you sure you're not talking about windows NT or 2000, which has so called "Administrative Shares" that cannot be disabled. They are accessible by mapping (or using the command "net") \\servertoview\C$ (of D$ for disk D: etc). You "only" need to know the Administrator password, well, that implies "if it's not empty", which fortunately is not the case on many many workstations.
That's only because "IT expert" nowadays means "I can use Word". Which is a real sad state, because it spoils it for every real IT expert out there. Anyways, when the DSL guy came over to our place, I showed him the prepared OpenBSD box and asked him if he wanted to try. He said: "no" plugged in his own laptop to test if the connection was up, then he showed to the connection worked. Finally he wished me good luck and left with a big smile on his face.
I really don't think most people care about spyware.
It depends. Most teenagers whom I told about spyware didn't care... most adults however look at you with disbelief and say "they can't really do that, do they?". One of my closest friends, who is a marketing major, started muttering: "but...but...that's illegal!" (or something in the lines).
Just try to explain it to anyone you see using Kazaa or other spyware infested software (yes, I know about Kazaa Lite), oh, and don't just say "it contains spyware". That doesn't say a damn think to them. Make it a little bit more vivid and make them think there sits a nasty little monster in their computer that sends everything they surf/do on their computer to a "Big Bad Corporation(tm)".
That's going to be one boring show...
All they'll do is show them their stamps colletion and their PDP-11 in the basement. Well, that's what I'd do. I'm probably missing something, but I cannot remember what. Of, well, back to my code...
I agree with you... What a lot people fail to understand is that most software developers on this world don't write software for the general public, they write expensive inhouse software that requires maintaining and several changes over the years (laws that change etc...) This is exactly how I pay the rent.
So before everyone starts ro panick and leaves the software bussiness: there is work, you just won't your software on the shelves. So creating software is a viable "bussiness model", just *not* for the general public.
One remark: Dolby has the right idea: come up with an idea, and license it until the end of time
Isn't this essentially what Microsoft tried to move to? I mean "subscription software". For the moment they backed up, but I know one kind of subscription software that is quite popular: think "Antivirus".
Or is there an antivirus out there, that just "sells" and gives it's updates out for the rest of your lifetime? It used to be that way, when virusses were still virussen (appended themselves to.exe,.com) and McAfee was free for presonal use.
dont think that the NSA permitted 128 bit encryption to be exported outside the states if they didnt have some backdoor to decrypt without brute-forcing.
That's why pharmacists have to study so long. It's not the chemisty and stuff like that that makes it hard, but all the courses of "handwriting decryption" and "graphology".;-)
How many times exactly do I have to tell you that I took that course years ago? So, okay, I'm wrong (yes, I did some googling), buy hey, let's talk about your knowlegde about this subject as soon as you are out of University and had years of mind-numbing Corpoate environments to endure.
Of course it would... Well, I don't know for XP, but Windows 2000 would work nice, provided you have at least some memory spare. 128Meg for the OS, 128Meg for the rest, and you're fine to my experience. Anything less makes it slow as molasses. Also you need to know what you are doing (which I am sure you do), but 95% of computer users don't.
I did my share of Windows installing, but I've never disabled any of the default services (okay, I disabloe Indexing) that come with Windows 2000 Pro... That's because most of them do not really say what they do (For example "Computer brower"? Eh? Yeah, that's Samba, so I need it). If I can reduce my memory usage from 90Meg idle to 40Meg idle I'd be most happy to. Could you point to a link how to do that with a prefect explanation why a certain service is not needed.
I am Pro-Mac, but only because I actually switched about 10 months ago. I know a lot about Windows, including tweaking the registry. I've been in the DOS world for over 10 years, I have seem Win 1.0, up to what we have now. I actually have a very custom Windows 2000 installation which does not include "My Documents" nor "Program Files", it's spread over different partitions for added security (works find, didn't lose anything in over 5 years). It's neraly setup as a BSD machine (think/home,/root,/tmp,/usr) I also minimize memory usage, anything that "wants" to go into the taskbar gets disabled at once by deleting the appropriate keys in the registry. Stangely enough, I do not see how the Add/Remove tool helps. The only apps that I see there are the ones I actually installed and want to use, the "Remove Windows components" doesn't list a thing expect "Indexing Service" which gets promptly removed every time.
It's not that I diss windows. Four of my machines run Windows 2000, only one is a Mac, two others are OpenBSD, another pure Linux and most Windows 2000 PC's dual-boot Linux. Finally, let me tell you: my Windows machines run stable most of the time (the registry corruption was a first! Except of that never had any problem), because I know how to manage them. I have userrights in place, different policies, etc.... You really should not think that I don't know anything about windows. The point is: how can anyone without my experience run a Windows 2000 machine, without needing the sixmonthly reinstall?
Why exactly does my Mac run out of the Box and my Windows machines need all these tweaking unless I run a top-of-the line machine? That is what really astonished me hard while using the Mac.
I agree, computers are tools... But even in tools there is a difference in quality.
By the way, it is very strange, every time I start talking about macs on slashdot I get flamed into oblivion even though I just talk about my personal experience, isn't that what this forum is all about? Isn't this forum to be supposed pro-Unix?
Well, for older games I have my P120 laptop lying around (think the original Civilisation). The PPro is used for Word Processing, Excell, Access, Surfing, Eudora, Scanning, Photoshop, CD-burning, SetiAtHome, various games including The Sims, Halflife (including Opposing forces), Need for Speed (III and IV, my brother loves that). Granted not the latest games, but unless a game is over 5 years old it's not really *old*;-)
I also did my share of programming on that machine, ranging from C++ to Java. Works fine. Back in the day I even ran 3DSMax on it, but I suck at modelling. All in all, a nice palmares.
And in case you wonder: I love older computers.
Look, I didn't bring up statistics, only my personal experience. Great for you that XP runs on a 500Mhz/64Meg RAM machine, but unless you only surf, email and so some wordprocessing you won't go far with that. I have talked to a lot of people running XP, and most were not that extact about it even those who had 1GHz+ machines.
From my experience Windows 2000 works fine on slower CPU's (I have a PPro 200 and a an AMD K6-II running it here) but unless you go wild on the RAM you won't get anything special done. (The PPro has 256Meg RAM and the AMD has 384Meg RAM...both work fine).
My experience with XP is limited, I mostly stay with Windows 2000. OS X doesn't feel more sluggish than Windows XP on a low end machine (Luna enabled, that's what most users, use you know... but I think you are righ, people that use it are insane.). Anyway, when I bought my PC which is a P-III 800Mhz, Windows 2000 ran like a slug. Only after I upgraded from 128Meg RAM to 768Meg RAM the thing got usable. The UI in OS X is a bit heavy, yes, but you can disable a lot of stuff in order to let it go faster. Well different tastes I guess. I like OS X, but mainly because of the undeling Unix, I'm more in the terminal than anything else.;-)
Great... I know those Lifebooks, but they came out about a year ago. Most Fujitsu lifebooks before that were not that pretty. How long is the TiBook out? At *least* two years, if not longer.
(Disclaimer: I actually work for Fujitsu... I refused their laptop and bought me an Apple instead.... Yes, I am hated by management)
Strange, I just had to reinstall a W2K Pro machine because of registry corruption. How is that stable? (Okay, it had been running about 1,5 years continuously) Windows XP (anything) is bloated, I wouldn't dare to install it on anything less than a 1Gz CPU with at least 256Meg RAM (and that's barely enough to *do* anything).
Strangely enough Mac OS X runs fine on my 600Mhz iBook with 384Meg RAM. About 112Meg of that is used, and I have open Mail, two browsers (Chimera and Mozilla), Terminal, SetiAtHome, CPU Monitor, Memory Monitor and Stickies.
From experience I know that a W2K Pro installation boots up and idles at 95Meg RAM. Throw in some apps and you're memory I fast filled.
So what exactly is "bloat" to you?
No, that is not what I say... The missing word in BOTH statements is "deterministic".
Is this that hard to comprehend? It's one word that is missing that makes the statement false, and that is it.
The point is: We KNOW that NP-hard problems are solvable in polynomial time, just NOT in a deterministic way.
The question here is not at all if NP=P, since NP is defined as "Solvable non-deterministically in Polynomial time" and P is defined as "Solvable deterministically in polynomial time". *sigh* Hence: both techniques solve in Polynomial time only the way of processing is drastically different in order to archieve a solution.
Insightful? I don't think so. People are by now used to ignore ads because they are everywhere. Now, marketers need to be much more creative in order to get noticed...and doing extravagant (or even illegal) things will get attention in many ways. The main action stirs up dust, with a result that local newspapers will report it (or even non-local). Remember: there is no such thing as bad publicity.
Besides, with a little luck these rollerblading people with butterfly wings are cute girl. And who never fantasized about cute elfs... Oh, wait... that would be just me... did I say that out loud?
Well it is after all you that said "we simply don't know whether NP-hard problems can be solved in polynomial time or not.", which in my eyes is completely equivalent to saying "NP problems cannot be solved in polynomial time". As you said, you implied determinism, but to someone who didn't have computer theory classes this non-obvious. Not everyone around here has a computer science degree.
You are right that not every problem is parallelizable. Your example doesn't really "solve" anything, it just displays a list of "hello" encrypted with 2**1000 different keys. But even then it is synchronizable, let each of your computer start the encryption 1ms apart and send it to a queue: the results will come in nicely sequential and you have your result. This would take very very very long (since 1ms * 2**^1000 seconds = very very long), but it would be Polynomial time. Polynomial time does not imply "in our lifetime". Note also that you would only need 86400000 computers, since after 86400000ms the first one is done and can start the next calculation.
This is solving a problem: namely finding the key of a known encrypted message. And very easy to solve when computing non-deterministically. Just let each of your nodes do exactly one calculation. Things like DNA-Computing and Quantum computing are based on this philosophy: try everything at the same time and filter out the result.
I have to admit that my thinking was more in the lines of the classics like Travelling Salesman, or the one with the boolean expression where you have to fill in booleans in order to get it true with a give set of operators.
Finally, since all NP-hard problems can be reformulated as another NP-hard problem, it is should be possible to transform any non-synchronizable (ahem!) NP-hard problem into a synchronizable NP-problem. Not sure about that, as said... long time ago I had that course.
I'm pretty sure that my computer theory professor would have deducted points if I said "NP problems cannot be solved in polynomial time". It's some years ago I had that course, so times might have changed.
Now, please explain *why* those statement are wrong for parallelism and non-determinism (I know that non-determinism != parallelism, but it's a way to *illustrate*). That's the way I used to remember the difference between P and NP, worked fine you know. Of course, I'm willing to learn if you know so much better.
we simply don't know whether NP-hard problems can be solved in polynomial time or not.
And that is wrong too.... Because NP means "Non-Deterministically in Polynomial time". This means you *can* solve the problem in polynomial time, but that you need to approach the problem by attacking it in a parallel way. Look at it this way: if you have a problem that needs 2^1000 different calculation and each of those calculations lasts 1 day. Given you have 2^1000 computers, you can solve the problem in 1 day, which is *definately* polynomial (it's constant!)
Luxembourg (Europe) here.... I got modpoints on my main account and I modded the post up +1, Funny. That's the least it deserved. No, I normally don't do this.
Re:The slashdot crowd runs Windows most of the tim
on
ECCp-109 Solved
·
· Score: 1
I know, I'm guilty myself (though I finally bought a second harddisk to put Linux on, instead of giving it a meager 4Gig). At home I either use W2K with Phoenis/Mozilla or my iBook with Chimera/Mozilla, servers are of course BSD. At work, well, no choice there... W2K on IE6, a shame...
On my own personal webserver I see quite some Macs, I once even had a Sun machine visiting, and one or two Linux machines. Yes, it is linked from slashdot (but not from this account).
Slashdot Mirror
Last time I installed 95 and/or 98 with Samba (Sorry "Microsoft Sharing") nothing was shared at all by default. Yes, it is automatically bound to the TCP/IP of every networking device you have but that is a different issue.
Are you sure you're not talking about windows NT or 2000, which has so called "Administrative Shares" that cannot be disabled. They are accessible by mapping (or using the command "net") \\servertoview\C$ (of D$ for disk D: etc). You "only" need to know the Administrator password, well, that implies "if it's not empty", which fortunately is not the case on many many workstations.
Mod up Funny! Click the link... (no fear, it's not goatse.cx) I'm still laughing.
So when do I talk to your lawyer?
Very appropriate sig on the topic by the way. And an addenum to the sig: "show a man slashdot and he is lost forever".
That's only because "IT expert" nowadays means "I can use Word". Which is a real sad state, because it spoils it for every real IT expert out there.
Anyways, when the DSL guy came over to our place, I showed him the prepared OpenBSD box and asked him if he wanted to try. He said: "no" plugged in his own laptop to test if the connection was up, then he showed to the connection worked. Finally he wished me good luck and left with a big smile on his face.
It depends. Most teenagers whom I told about spyware didn't care... most adults however look at you with disbelief and say "they can't really do that, do they?". One of my closest friends, who is a marketing major, started muttering: "but...but...that's illegal!" (or something in the lines).
Just try to explain it to anyone you see using Kazaa or other spyware infested software (yes, I know about Kazaa Lite), oh, and don't just say "it contains spyware". That doesn't say a damn think to them. Make it a little bit more vivid and make them think there sits a nasty little monster in their computer that sends everything they surf/do on their computer to a "Big Bad Corporation(tm)".
That's going to be one boring show...
All they'll do is show them their stamps colletion and their PDP-11 in the basement. Well, that's what I'd do. I'm probably missing something, but I cannot remember what. Of, well, back to my code...
So before everyone starts ro panick and leaves the software bussiness: there is work, you just won't your software on the shelves. So creating software is a viable "bussiness model", just *not* for the general public.
One remark:
Dolby has the right idea: come up with an idea, and license it until the end of time
Isn't this essentially what Microsoft tried to move to? I mean "subscription software". For the moment they backed up, but I know one kind of subscription software that is quite popular: think "Antivirus". .exe, .com) and McAfee was free for presonal use.
Or is there an antivirus out there, that just "sells" and gives it's updates out for the rest of your lifetime? It used to be that way, when virusses were still virussen (appended themselves to
See if we care... Greetings from Europe!
That's why pharmacists have to study so long. It's not the chemisty and stuff like that that makes it hard, but all the courses of "handwriting decryption" and "graphology". ;-)
You know, you forget stuff while aging....
Of course it would... Well, I don't know for XP, but Windows 2000 would work nice, provided you have at least some memory spare. 128Meg for the OS, 128Meg for the rest, and you're fine to my experience. Anything less makes it slow as molasses.
Also you need to know what you are doing (which I am sure you do), but 95% of computer users don't.
I did my share of Windows installing, but I've never disabled any of the default services (okay, I disabloe Indexing) that come with Windows 2000 Pro... That's because most of them do not really say what they do (For example "Computer brower"? Eh? Yeah, that's Samba, so I need it). If I can reduce my memory usage from 90Meg idle to 40Meg idle I'd be most happy to. Could you point to a link how to do that with a prefect explanation why a certain service is not needed. /home, /root, /tmp, /usr) I also minimize memory usage, anything that "wants" to go into the taskbar gets disabled at once by deleting the appropriate keys in the registry. Stangely enough, I do not see how the Add/Remove tool helps. The only apps that I see there are the ones I actually installed and want to use, the "Remove Windows components" doesn't list a thing expect "Indexing Service" which gets promptly removed every time.
It's not that I diss windows. Four of my machines run Windows 2000, only one is a Mac, two others are OpenBSD, another pure Linux and most Windows 2000 PC's dual-boot Linux. Finally, let me tell you: my Windows machines run stable most of the time (the registry corruption was a first! Except of that never had any problem), because I know how to manage them. I have userrights in place, different policies, etc.... You really should not think that I don't know anything about windows. The point is: how can anyone without my experience run a Windows 2000 machine, without needing the sixmonthly reinstall?
I am Pro-Mac, but only because I actually switched about 10 months ago. I know a lot about Windows, including tweaking the registry. I've been in the DOS world for over 10 years, I have seem Win 1.0, up to what we have now. I actually have a very custom Windows 2000 installation which does not include "My Documents" nor "Program Files", it's spread over different partitions for added security (works find, didn't lose anything in over 5 years). It's neraly setup as a BSD machine (think
Why exactly does my Mac run out of the Box and my Windows machines need all these tweaking unless I run a top-of-the line machine? That is what really astonished me hard while using the Mac.
I agree, computers are tools... But even in tools there is a difference in quality.
By the way, it is very strange, every time I start talking about macs on slashdot I get flamed into oblivion even though I just talk about my personal experience, isn't that what this forum is all about? Isn't this forum to be supposed pro-Unix?
Well, for older games I have my P120 laptop lying around (think the original Civilisation). The PPro is used for Word Processing, Excell, Access, Surfing, Eudora, Scanning, Photoshop, CD-burning, SetiAtHome, various games including The Sims, Halflife (including Opposing forces), Need for Speed (III and IV, my brother loves that). Granted not the latest games, but unless a game is over 5 years old it's not really *old* ;-)
I also did my share of programming on that machine, ranging from C++ to Java. Works fine. Back in the day I even ran 3DSMax on it, but I suck at modelling. All in all, a nice palmares.
And in case you wonder: I love older computers.
Look, I didn't bring up statistics, only my personal experience. Great for you that XP runs on a 500Mhz/64Meg RAM machine, but unless you only surf, email and so some wordprocessing you won't go far with that. I have talked to a lot of people running XP, and most were not that extact about it even those who had 1GHz+ machines.
From my experience Windows 2000 works fine on slower CPU's (I have a PPro 200 and a an AMD K6-II running it here) but unless you go wild on the RAM you won't get anything special done. (The PPro has 256Meg RAM and the AMD has 384Meg RAM...both work fine).
My experience with XP is limited, I mostly stay with Windows 2000. OS X doesn't feel more sluggish than Windows XP on a low end machine (Luna enabled, that's what most users, use you know... but I think you are righ, people that use it are insane.). Anyway, when I bought my PC which is a P-III 800Mhz, Windows 2000 ran like a slug. Only after I upgraded from 128Meg RAM to 768Meg RAM the thing got usable. ;-)
The UI in OS X is a bit heavy, yes, but you can disable a lot of stuff in order to let it go faster.
Well different tastes I guess. I like OS X, but mainly because of the undeling Unix, I'm more in the terminal than anything else.
Great... I know those Lifebooks, but they came out about a year ago. Most Fujitsu lifebooks before that were not that pretty. How long is the TiBook out? At *least* two years, if not longer.
(Disclaimer: I actually work for Fujitsu... I refused their laptop and bought me an Apple instead.... Yes, I am hated by management)
Strange, I just had to reinstall a W2K Pro machine because of registry corruption. How is that stable? (Okay, it had been running about 1,5 years continuously) Windows XP (anything) is bloated, I wouldn't dare to install it on anything less than a 1Gz CPU with at least 256Meg RAM (and that's barely enough to *do* anything).
Strangely enough Mac OS X runs fine on my 600Mhz iBook with 384Meg RAM. About 112Meg of that is used, and I have open Mail, two browsers (Chimera and Mozilla), Terminal, SetiAtHome, CPU Monitor, Memory Monitor and Stickies.
From experience I know that a W2K Pro installation boots up and idles at 95Meg RAM. Throw in some apps and you're memory I fast filled.
So what exactly is "bloat" to you?
Is this that hard to comprehend? It's one word that is missing that makes the statement false, and that is it.
The point is: We KNOW that NP-hard problems are solvable in polynomial time, just NOT in a deterministic way.
The question here is not at all if NP=P, since NP is defined as "Solvable non-deterministically in Polynomial time" and P is defined as "Solvable deterministically in polynomial time". *sigh* Hence: both techniques solve in Polynomial time only the way of processing is drastically different in order to archieve a solution.
Besides, with a little luck these rollerblading people with butterfly wings are cute girl. And who never fantasized about cute elfs... Oh, wait... that would be just me... did I say that out loud?
Oh, thank you...now you fried my brain just by making me think of that!
You are right that not every problem is parallelizable. Your example doesn't really "solve" anything, it just displays a list of "hello" encrypted with 2**1000 different keys. But even then it is synchronizable, let each of your computer start the encryption 1ms apart and send it to a queue: the results will come in nicely sequential and you have your result. This would take very very very long (since 1ms * 2**^1000 seconds = very very long), but it would be Polynomial time. Polynomial time does not imply "in our lifetime". Note also that you would only need 86400000 computers, since after 86400000ms the first one is done and can start the next calculation.
A better example would have been this:
String unencryptedMessage="Hello World";
String encryptedMessgae="fqwdwezufzd";
for( verylong key=0; key < 2^1000; key++ ) {
if ( encrypt( key, unencryptedMesage).equals( encryptedMessgae ) ) { System.out.println( "The encryption key is: " + key ); }
}
This is solving a problem: namely finding the key of a known encrypted message. And very easy to solve when computing non-deterministically. Just let each of your nodes do exactly one calculation. Things like DNA-Computing and Quantum computing are based on this philosophy: try everything at the same time and filter out the result.
I have to admit that my thinking was more in the lines of the classics like Travelling Salesman, or the one with the boolean expression where you have to fill in booleans in order to get it true with a give set of operators.
Finally, since all NP-hard problems can be reformulated as another NP-hard problem, it is should be possible to transform any non-synchronizable (ahem!) NP-hard problem into a synchronizable NP-problem. Not sure about that, as said... long time ago I had that course.
Now, please explain *why* those statement are wrong for parallelism and non-determinism (I know that non-determinism != parallelism, but it's a way to *illustrate*). That's the way I used to remember the difference between P and NP, worked fine you know. Of course, I'm willing to learn if you know so much better.
And that is wrong too.... Because NP means "Non-Deterministically in Polynomial time". This means you *can* solve the problem in polynomial time, but that you need to approach the problem by attacking it in a parallel way. Look at it this way: if you have a problem that needs 2^1000 different calculation and each of those calculations lasts 1 day. Given you have 2^1000 computers, you can solve the problem in 1 day, which is *definately* polynomial (it's constant!)
Luxembourg (Europe) here.... I got modpoints on my main account and I modded the post up +1, Funny. That's the least it deserved. No, I normally don't do this.
On my own personal webserver I see quite some Macs, I once even had a Sun machine visiting, and one or two Linux machines. Yes, it is linked from slashdot (but not from this account).