Sounds great, where can I buy shares in Scientology?
Seriously though, now that Hubbard's dead (or frozen in a secret lab?), where does the money go, who gets it? Does it all get funnelled back into Scientology in some self perpetuating system that's out of control, or what?
Leaving our kids a polluted planet to grow up in? I solved that one long ago, I'm not having kids.
If we all do our part then the children won't suffer.:)
I think I've got a more efficient design for Ontro too. I call it Disposable Landfill - you don't even have to bother drinking the contents, you just buy it and throw it out! Hard to get much easyier than that, the disposable society will love it, and I'll make millions.
There are some real devils in the details, but it's looking very much like we might be able to make Disposable Landfill out of mostly recycled materials so it'll be environmentally friendly too.
(Don't get any smart ideas, I already have a patent pending)
Regulation of the software industry would merely result in software companies getting insurance and tapping the insurance premium onto the price of the software. The only extra testing that would occur would be to bring about an equalibrium in the price-of-extra-testing vs lower-premiums-for-better-track-record.
If there was a market for software with liability then vendors would make software with liability. That they don't (or that it's rare) indicates that for most people, cost is more important - and if you regulate then you remove that choice.
The examples of other industry (buildings, planes etc) are always flawed as they are regulated for safety reasons, not consumer protection reasons (software upon which lives depend is a different kettle of fish). But even then I'm personally glad to live in a country where patients can't sue doctors (they held are accountable by other means), the state in the US looks more ridiculous each day.
However, with all the continual moaning about buggy software, it does surprise me that there aren't more vendors out there saying 'Our software works, and you can hold to that'.
Hey, you're a software person - maybe there's an untapped market there...
simple code is obvious and you can read what it does, you can't however read from the code why the code was put there in the first place - why it doing things the way it is, what the intent behind the code is etc.
eg:
ShowMessageExclamation('You can't currently do this because...'); PostMessage(historyEraseCheckBox.Handle, WM_LBUTTONUP, X, Y);
Perfectly readable - it brings up a message to the user then posts a WM_LBUTTONUP to the historyErase tick box.
WTF does it post that WM_LBUTTONUP?
With comments your attention would be drawn to the fact that that that piece of code gets called from within a historyErase click, tick boxes fire their click events on a mousedown, you have just changed focus after the mousedown event but before the mouse button is released (by bringing up a message box on the mouseDown), and so the check box will never get a mouse up and end up in the wrong state until clicked on again.
There are other ways to do this, but then (without comments) you'd be left wondering why the code didn't just call ShowMessageExclamation() in the onClick event.
I was going to say that only trivial code doesn't need comments, but even the example I gave is trivial.
Radix(??) sort the first array into a hash table, then go through the second array and look up the hash of each element to see what elements of the first array it might match.
O(n) if your hash table is big enough that you won't get any double ups (assuming array values are unique), but still very fast if a table that big isn't possible.
While I have to grant that the guy made his point nicely, I must complain bitterly that he's seeded my brain with this question - now I want to know what the fastest algorithm would be, and I don't even have an application for it.
1) A key generator: Create your own personalized registration key. This was my favorite way to pirate and it usually doesnt take the professionals long to create a keygen either.
Not possible if the key system uses public key encyption.
2) A crack: completely disables the 'time checking' on the shareware by altering the binaries in some way
Many people have issues running executables from sources as trustworthy as warez sites.
You don't need to auth with a server to prevent keygens from being possible.
You can't write a keygen if the key system uses public key encryption to sign the timestamp (a few implementation details ignored here as you get the idea). You'd have to crack the software instead then - and who knows what that software crack will install on your system.
This kind of 'strong' key isn't appropriate in all situations due to its length - for instance where the user has to type it out (eg a key printed on the softwares CD), but in the case of shareware where it is emailed to you (or supplied on a web page) and the user can just cut it and paste it, it's fine.
I thought the idea of only accepting keys less than 30 days old was quite a good one actually (one of those really damn obvious in hindsight things), sure the determined user can roll back their clock (if the key was provided with a date) but it adds two more hoops a pirating user has to jump through without affecting a legitimate user - calling up the server to auth the key is a Very Bad Idea, you get a tonne of paranoid users telling their friends and posting everywhere that you are spyware because their zonealarms went off (found that one out the hard way too;)).
The old joke, "if builders built buildings the way programmers wrote programs..." is still frighteningly true
Just a few notes:
Knuth also said software is hard. It could well be the case that software is harder than engineering - most commercial development is not forging any new ground at all, so there's no scapegoat there.
Regarding programmers taking pride in their work, I might have the pride in my work to write a cheque to anyone who found an error if I didn't have to keep a commerically viable pace - I say 'might' because I've never been in a situation where I can take as long as I want to complete something properly (I think I would, but I don't know). Knuth (being an academic) might have this opportunity, but I don't know anything about that, he might not.
The other problem is that I have pride in my work in the good sense, and pride in the bad sense - Most 'bugs' reported in my code will be partly someone elses fault (eg error in the OS API, or a bug in some 3rd party software) and I have great difficulty swallowing my pride and accepting responsibilty when someone else's code is partly involved. You need to be able to totally accept responsibility if you're going to start writing cheques.
You only have to make the source of GPL code available to whoever you make the program available to. If you're only using the program internally then you don't have to make the code available externally.
GPL covers only the code, it doesn't cover any of the data, graphics, resources, etc. Presumably your program will contain non-GPLable tryable-for-treason stuff like this, making your full working program non-distributable even if the code is (ala Quake).
If it really is classified Secret under US Code Umptifratz, or whatever, you have to ask yourself... do RMS or the FSF have Umptifratz clearance? Mum's the word.
If this stuff really concerns you, get a lawyer 'cause we don't know shit.
I have heaps of innovative ideas (eg how many commerical products do you see that make you think 'ha, I thought of that 5 years ago') and my common sense does not tell me to sell them or make a profit, common sense tells me you can't sell an idea without first putting the hard work into it to make it a product - and I have far more ideas than I can ever in a lifetime bring to fruition.
Innovative ideas are a time a dozen, the people who are willing to put in the hard labour on those ideas are the ones who get the profit, and rightly so - having an idea is no work at all, it just happens.
There are a ton of innovative ideas to revolutionize or just plain improve OS's out there (for free), but what hope is there of revolutions when even minor progress like losing the case sensitivity can't be achieved (in the entire history of computing, have case-sensitive file systems or command lines produced anything other than user errors or dodgy code?).
When linux has yet to even lose the case-sensitivity (and probably can't due to numerous things with their roots in decades old computing), I have to admit that Microsoft completely replacing the whole paradigm (hate that word but my vocab sucks) of their file system impresses me greatly and makes me think maybe one day we will get out of the OS rut we are stuck in, but I digress...
ShouldExist.org exists as an anti-patent site, and as a place to air your ideas so that hopefully someone will implement them (Tho I must admit I've been slack and have yet to used it for something serious).
He said it earned its position in the marketplace through illegal practices, he didn't say it's monopoly was illegal, and microsoft wouldn't be in the position it's in if it hadn't leveraged its monopoly illegally.
However I'll say it (And I'm a windows user, no less), even Microsoft's actual monopoly is earned (or at least strengthened) through illegal practices - requiring that OEMs only sell MS-and-only-MS machines, with the price of OEM non-compliance being to make all their windows machines cost so much that they will lose out to compeditor OEMs who are in bed with the MS-and-only-MS concept is very anticompeditive, and hopefully very, very illegal.
The code that checks the key can be cracked, but it's pretty trivial to create a key scheme that can't be cracked.
Just encrypt something (eg the user's name) with public key encryption and put the public key in the software.
And who would trust their machine to a Windows XP that had an invalid key but its authentication code tampered with? That kinda of thing could very easily come back and bite you.
The reason this happens is Ebay knows no distributor/retailer boundries.
Ebay has everything you can buy, you can spend 3 hours surfing around stores looking for what you want and not find it, or a few days walking around stores looking for what you want, or just go to Ebay and buy it (and pay the premium).
With Ebay you don't need to find the shop before you can find the item - you just look straight for the item. It doesn't matter if it's brand new or years out of print, it doesn't matter if it's rare, it doesn't matter if its not sold in any stores in your country (a reason I occasionally use it for new stuff).
A retail equivalent of Ebay, offering the wares of *all* retail stores would achieve the same sort of thing for retail goods - you could concertrate on what you want rather than who will have the best chance of stocking something like what you want. But since I doubt the retail industry will ever get its shit together that well, shops are stuck with using ebay (which is designed for auction, not retail)
People's use of Ebay (and paying above retail) isn't always stupidity (I myself have emailed sellers after a lost auction and bought the same item at retail price, and I've probably paid above retail on other auctions), it's often just convienience/lazyness.
To block SuperCookies requires changing an obscure option in WMP which is barely documented.
That is highly misleading at best, and complete bollox at worst.
Now I'm someone who will cherily click past a click-through license agreement without reading it, but Microsoft still managed to draw my attention to the existance of this ID, then told me what benifits it gave, and then how to disable it (which I did).
(They didn't mention the supercookie privacy bug tho:))
When you install WMP7 it brings up a Privacy Policy dialog (and those words immediately make anyone who would actually care [about web pages being able to collate info about them etc] decide 'this is something I should read') which explains pretty much in bullet points every aspect of WMP that might violate your privacy, what advantge you get by having it on, and how you can turn it off (including the Content Rights Management). You then have to tick an "I have read the privacy policy" checkbox before you can continue the install.
In that sense "an obscure option in WMP which is barely documented" is complete bollox. However, I imagine it's possible (now or soon) that you could buy a machine preconfigured from the store with WMP7, and not be provided with any information, or warning.
Windows2000 (SP2) comes bundled with a much earlier version of WMP so no worries there, but I've not looked at XP.
My question for anyone who has bothered to read this far...
(I'll word the same question it 3 different ways)
Is this just a bug, or would the only way to fix this bug defeat the entire purpose of the ID? / Can this feature exist without the side-effect? / Is it a side-effect or just the other side of a double edged sword?
Yes, it appears many people here seem to think the challenge is finding a P=NP solution.
We pretty much know that P != NP, we just can't prove it, the challenge is proving P != NP, not finding a P=NP solution.
Similarly, we know there are an infinite number of primes, just that nobody has been able to prove it.
Saying things like "So if (when) the time comes" with regard to finding a P=NP solution is somewhat akin to saying if (when) we can finally prove that the primes are finite...
Sure, it's somewhat fun to speculate on what the social implications might be, but why stop with P=NP, why don't we have an ask slashdot about how you would go public if (when) you discover the fountain of youth?
The scary part is that many people (including myself) have posted many, MANY messages to USENET, not realizing that 20 years later those same messages would be staring us back in the face.
Many tech employers do a web search of candidates they are considering hiring...
Don't worry about it.
Many people say that, but unless you included your full name in the 'From' field and you have a remarkably uncommon name, those 20 year old posts will never be seen by any HR guys.
Even I would have a tough time finding my own posts - my name is not unique, you cannot search usenet by geographic location of the poster, and free persistant email servers are a relatively new invention, before which my email address would change whenever I changed job/uni/course/isp/underwear etc.
Given a manned brute force search of usenet posts containing part or all of your name, some of your old email addresses may be discovered and confirmed, but there's no way to automate it - How many weeks can an HR guy get paid to investigate one of the many prospective employees?
If you're not a socially inept nerd, good for you. A sizable bunch of people on/. are, however.
Still, a legion of socially inept nerds armed with linux-powered tricorders that scan for single women who like Lego is probably a good start
(unless of course you are a single woman and you like lego).
This still doesn't seem like a good answer.
You're right, it really should be a mindstorm powered tricorder built from lego and developed on linux.
Nuclear isn't a renewable energy source, and I assume your trolling when you say nice and clean - "venting the exhaust" into underground dumps instead of the atmosphere doesn't mean it's clean.
Fuel cells are also extremely efficient, nuclear reactors are very inefficient.
Geeks love fuel cells because you just don't get more elegant, fission reactors are such an ugly hack.
Slashdot Mirror
Seriously though, now that Hubbard's dead (or frozen in a secret lab?), where does the money go, who gets it? Does it all get funnelled back into Scientology in some self perpetuating system that's out of control, or what?
PS good on you.
If we all do our part then the children won't suffer.
I think I've got a more efficient design for Ontro too. I call it Disposable Landfill - you don't even have to bother drinking the contents, you just buy it and throw it out! Hard to get much easyier than that, the disposable society will love it, and I'll make millions.
There are some real devils in the details, but it's looking very much like we might be able to make Disposable Landfill out of mostly recycled materials so it'll be environmentally friendly too.
(Don't get any smart ideas, I already have a patent pending)
And if we weren't already lazy enough, we now have disposable drink heaters to save us having to walk to the nearest microwave.
:)
Should we turn even more virgin resources into landfill for the sake of such trifleing convenience?
I think consumers will answer that, and it'll be a resounding YES!
Regulation of the software industry would merely result in software companies getting insurance and tapping the insurance premium onto the price of the software. The only extra testing that would occur would be to bring about an equalibrium in the price-of-extra-testing vs lower-premiums-for-better-track-record.
If there was a market for software with liability then vendors would make software with liability. That they don't (or that it's rare) indicates that for most people, cost is more important - and if you regulate then you remove that choice.
The examples of other industry (buildings, planes etc) are always flawed as they are regulated for safety reasons, not consumer protection reasons (software upon which lives depend is a different kettle of fish). But even then I'm personally glad to live in a country where patients can't sue doctors (they held are accountable by other means), the state in the US looks more ridiculous each day.
However, with all the continual moaning about buggy software, it does surprise me that there aren't more vendors out there saying 'Our software works, and you can hold to that'.
Hey, you're a software person - maybe there's an untapped market there...
eg:Perfectly readable - it brings up a message to the user then posts a WM_LBUTTONUP to the historyErase tick box.
WTF does it post that WM_LBUTTONUP?
With comments your attention would be drawn to the fact that that that piece of code gets called from within a historyErase click, tick boxes fire their click events on a mousedown, you have just changed focus after the mousedown event but before the mouse button is released (by bringing up a message box on the mouseDown), and so the check box will never get a mouse up and end up in the wrong state until clicked on again.
There are other ways to do this, but then (without comments) you'd be left wondering why the code didn't just call ShowMessageExclamation() in the onClick event.
I was going to say that only trivial code doesn't need comments, but even the example I gave is trivial.
Radix(??) sort the first array into a hash table, then go through the second array and look up the hash of each element to see what elements of the first array it might match.
O(n) if your hash table is big enough that you won't get any double ups (assuming array values are unique), but still very fast if a table that big isn't possible.
While I have to grant that the guy made his point nicely, I must complain bitterly that he's seeded my brain with this question - now I want to know what the fastest algorithm would be, and I don't even have an application for it.
Many people have issues running executables from sources as trustworthy as warez sites.
You don't need to auth with a server to prevent keygens from being possible.
;)).
You can't write a keygen if the key system uses public key encryption to sign the timestamp (a few implementation details ignored here as you get the idea). You'd have to crack the software instead then - and who knows what that software crack will install on your system.
This kind of 'strong' key isn't appropriate in all situations due to its length - for instance where the user has to type it out (eg a key printed on the softwares CD), but in the case of shareware where it is emailed to you (or supplied on a web page) and the user can just cut it and paste it, it's fine.
I thought the idea of only accepting keys less than 30 days old was quite a good one actually (one of those really damn obvious in hindsight things), sure the determined user can roll back their clock (if the key was provided with a date) but it adds two more hoops a pirating user has to jump through without affecting a legitimate user - calling up the server to auth the key is a Very Bad Idea, you get a tonne of paranoid users telling their friends and posting everywhere that you are spyware because their zonealarms went off (found that one out the hard way too
Just a few notes:
Knuth also said software is hard.
It could well be the case that software is harder than engineering - most commercial development is not forging any new ground at all, so there's no scapegoat there.
Regarding programmers taking pride in their work, I might have the pride in my work to write a cheque to anyone who found an error if I didn't have to keep a commerically viable pace - I say 'might' because I've never been in a situation where I can take as long as I want to complete something properly (I think I would, but I don't know). Knuth (being an academic) might have this opportunity, but I don't know anything about that, he might not.
The other problem is that I have pride in my work in the good sense, and pride in the bad sense - Most 'bugs' reported in my code will be partly someone elses fault (eg error in the OS API, or a bug in some 3rd party software) and I have great difficulty swallowing my pride and accepting responsibilty when someone else's code is partly involved. You need to be able to totally accept responsibility if you're going to start writing cheques.
You know, knowing you exist makes me sleep sounder at night.
In the efforts of keeping the discussion going.
I have heaps of innovative ideas (eg how many commerical products do you see that make you think 'ha, I thought of that 5 years ago') and my common sense does not tell me to sell them or make a profit, common sense tells me you can't sell an idea without first putting the hard work into it to make it a product - and I have far more ideas than I can ever in a lifetime bring to fruition.
Innovative ideas are a time a dozen, the people who are willing to put in the hard labour on those ideas are the ones who get the profit, and rightly so - having an idea is no work at all, it just happens.
There are a ton of innovative ideas to revolutionize or just plain improve OS's out there (for free), but what hope is there of revolutions when even minor progress like losing the case sensitivity can't be achieved (in the entire history of computing, have case-sensitive file systems or command lines produced anything other than user errors or dodgy code?).
When linux has yet to even lose the case-sensitivity (and probably can't due to numerous things with their roots in decades old computing), I have to admit that Microsoft completely replacing the whole paradigm (hate that word but my vocab sucks) of their file system impresses me greatly and makes me think maybe one day we will get out of the OS rut we are stuck in, but I digress...
ShouldExist.org exists as an anti-patent site, and as a place to air your ideas so that hopefully someone will implement them (Tho I must admit I've been slack and have yet to used it for something serious).
This story on shouldexist provides arguments finer than mine.
He said it earned its position in the marketplace through illegal practices, he didn't say it's monopoly was illegal, and microsoft wouldn't be in the position it's in if it hadn't leveraged its monopoly illegally.
However I'll say it (And I'm a windows user, no less), even Microsoft's actual monopoly is earned (or at least strengthened) through illegal practices - requiring that OEMs only sell MS-and-only-MS machines, with the price of OEM non-compliance being to make all their windows machines cost so much that they will lose out to compeditor OEMs who are in bed with the MS-and-only-MS concept is very anticompeditive, and hopefully very, very illegal.
Just encrypt something (eg the user's name) with public key encryption and put the public key in the software.
And who would trust their machine to a Windows XP that had an invalid key but its authentication code tampered with? That kinda of thing could very easily come back and bite you.
The reason this happens is Ebay knows no distributor/retailer boundries.
Ebay has everything you can buy, you can spend 3 hours surfing around stores looking for what you want and not find it, or a few days walking around stores looking for what you want, or just go to Ebay and buy it (and pay the premium).
With Ebay you don't need to find the shop before you can find the item - you just look straight for the item. It doesn't matter if it's brand new or years out of print, it doesn't matter if it's rare, it doesn't matter if its not sold in any stores in your country (a reason I occasionally use it for new stuff).
A retail equivalent of Ebay, offering the wares of *all* retail stores would achieve the same sort of thing for retail goods - you could concertrate on what you want rather than who will have the best chance of stocking something like what you want. But since I doubt the retail industry will ever get its shit together that well, shops are stuck with using ebay (which is designed for auction, not retail)
People's use of Ebay (and paying above retail) isn't always stupidity (I myself have emailed sellers after a lost auction and bought the same item at retail price, and I've probably paid above retail on other auctions), it's often just convienience/lazyness.
Well, I take it all back then. I guess unless you upgrade to WMP 7 then it is going to be rather obsure.
Now I'm someone who will cherily click past a click-through license agreement without reading it, but Microsoft still managed to draw my attention to the existance of this ID, then told me what benifits it gave, and then how to disable it (which I did).
(They didn't mention the supercookie privacy bug tho
When you install WMP7 it brings up a Privacy Policy dialog (and those words immediately make anyone who would actually care [about web pages being able to collate info about them etc] decide 'this is something I should read') which explains pretty much in bullet points every aspect of WMP that might violate your privacy, what advantge you get by having it on, and how you can turn it off (including the Content Rights Management). You then have to tick an "I have read the privacy policy" checkbox before you can continue the install.
In that sense "an obscure option in WMP which is barely documented" is complete bollox. However, I imagine it's possible (now or soon) that you could buy a machine preconfigured from the store with WMP7, and not be provided with any information, or warning.
Windows2000 (SP2) comes bundled with a much earlier version of WMP so no worries there, but I've not looked at XP.
My question for anyone who has bothered to read this far...
(I'll word the same question it 3 different ways)
Is this just a bug, or would the only way to fix this bug defeat the entire purpose of the ID? / Can this feature exist without the side-effect? / Is it a side-effect or just the other side of a double edged sword?
My mistake, I meant infinitely many twin primes.
That's a neat proof tho.
Yup, that was the question I meant.
Yes, it appears many people here seem to think the challenge is finding a P=NP solution.
We pretty much know that P != NP, we just can't prove it, the challenge is proving P != NP, not finding a P=NP solution.
Similarly, we know there are an infinite number of primes, just that nobody has been able to prove it.
Saying things like "So if (when) the time comes" with regard to finding a P=NP solution is somewhat akin to saying if (when) we can finally prove that the primes are finite...
Sure, it's somewhat fun to speculate on what the social implications might be, but why stop with P=NP, why don't we have an ask slashdot about how you would go public if (when) you discover the fountain of youth?
Many people say that, but unless you included your full name in the 'From' field and you have a remarkably uncommon name, those 20 year old posts will never be seen by any HR guys.
Even I would have a tough time finding my own posts - my name is not unique, you cannot search usenet by geographic location of the poster, and free persistant email servers are a relatively new invention, before which my email address would change whenever I changed job/uni/course/isp/underwear etc.
Given a manned brute force search of usenet posts containing part or all of your name, some of your old email addresses may be discovered and confirmed, but there's no way to automate it - How many weeks can an HR guy get paid to investigate one of the many prospective employees?
(unless of course you are a single woman and you like lego).
You're right, it really should be a mindstorm powered tricorder built from lego and developed on linux.
What's wrong with divX becoming entrenched?
DivX and Open-DivX are encode/decode compatible.
Open-DivX may be patent encumbered, but it sounds like this new format is too.
DivX walks all over mpeg1 and mpeg2 (mpeg2 basically being mpeg1 with more options), so the sooner mpeg1 & 2 is unentrenched the better.
The only other popular codec that's up to spec is Quicktime Sorenson, and it's hard to get any more proprietry than that.
Nuclear isn't a renewable energy source, and I assume your trolling when you say nice and clean - "venting the exhaust" into underground dumps instead of the atmosphere doesn't mean it's clean.
Fuel cells are also extremely efficient, nuclear reactors are very inefficient.
Geeks love fuel cells because you just don't get more elegant, fission reactors are such an ugly hack.
Fuel cells are most efficient when running on straight hydrogen, which may have it's own safety issues.