Unfortunately we're keeping the tracking to a bare minimum so I don't believe that we can do that. We're trying to keep updates to the DB to a bare minimum so we don't have a new bottleneck.
Reasonable, again.
I do wonder, however, about your statistics on page impressions.. I know that when I'm active in a comment discussion (which is several times a day), I can go through hundreds of comment pages. At a half cent a page, that would add up quickly. I understand you're saying that most of your users just hit the home page and link off to whatever is new. I hope that you'll structure your advertising for that, and make the biggest, most obnoxious ads appear on the home page rather than on each comment viewed.
Interestingly, I note that when I click on a comment, the HREF # scrolls the page down so I can't even see the ads that are there currently. Surely this can't continue, or are you merely assuming your advertisers don't know that a lot of ad impressions you provide are never actually seen by the user?
We don't expect users to pay a half cent for every page. We expect that what will happen is that users will suppress ads on just articles or the homepage effectively costing a few cents a day.
Sweet reason and good sense. I'm happy.
Is there any way that you can display page view statistics even for pages we view with ads? As I said, right now I'm leaving ads enabled for everything. I'd kinda like to be able to know how many page views I'm getting with ads so that I can better judge whether I should turn the ads off for awhile and drain down my slashdot account and put more money in. Otherwise I can well imagine leaving all the ads on permanently, and just putting in money to my slashdot account at whatever interval makes me feel happy and that lets me feel like I'm contributing in a meaningful way.
Of course, X-10 pop-unders would take care of any such impulse for altruism pretty quickly.;-)
Of course, this begs the question as to whether Taco/OSDN will have to start restricting the RDF feeds and/or content scrapers so that people can't suck the editorial content from Slashdot to feed such a site.
I'm up on the paid site thing. In fact, I just paid for 10,000 page views.
Then I promptly went and turned all the ads back on. Why? Because a) the ads aren't annoying yet.. in fact, since they are no longer just the OSDN ones, they're actually more interesting these days, and b) because I don't think I can afford to pay a half cent for each page I view on the entire web.
And that's what we're really talking about, here, isn't it? If slashdot goes pay-per-click, why shouldn't every other web site do so as well? If every web site did so, and each was a half-penny per click, I imagine that would easily amount to 15 bucks a day for me based on my current browsing habits. That could be as much as $450 a month.
Psychologically, I'm happy to pay a lifetime subscription to Salon for only $50. And I have an all-you-can-eat yearly subscription to The Economist for only $125, which includes an actual magazine I get delivered to my home every week. I don't mind those because the expenditure is controlled, and the meter isn't continuously running at such a high rate as a half penny per page.
Does it really cost half a penny for each page impression served by Slashdot, Taco? Were the ads making anywhere near that much?
And I see that I'm getting new ads for each step in the posting process (preview, etc.). Does that mean that it could a couple of cents to post to slashdot?
That would be one way to tune out the trolls, I guess. <smirk>
I wrote the following to my senator, which captures the sense of my concerns far better than a 30 second phone call could ever do:
I am writing to you to express my opposition to any new legislation regarding digital copyright as sought by the film and recording industries. In particular, I am a computer programmer in vehement opposition to any legislation that would make it a crime to create or distribute any digital devices that do not include government mandated Digital Rights Management software. The point of such legislation, in conjunction with present law, is to attempt to give the film and recording industries absolute power over how anyone may view or use copyrighted material, without regard for the Supreme Court's finding in the Betamax case that citizens may engage in certain fair use practices.
Further, if such legislation were passed, there are grave dangers having to do with establishing a patent-based monopoly on the DRM standard, and grave dangers having to do with effectively outlawing any computer device that could be programmed by the user. The DMCA already makes it illegal to distribute tools to break protection, regardless of fair use. That is controversial, but manageable.. systems that don't touch protected material need not be affected, and would simply forego access to such content. An affirmative duty that all digital systems include DRM content controls would be incredibly far-reaching, and could be construed as banning any system (such as a PC running Linux) in which the user has complete control over the configuration and details of their system.
Hollywood and the recording industry are asking for a tremendous amount of control over technology in the United States (and the world, through their lobbying of WIPO), and it is not at all clear that it is in the nation's interest that they be given it.
I'm writing this on the occasion of tomorrow's upcoming digital copyright hearings of the Senate Commerce Committee. I hope you will be extremely vigilant in assessing the industry's claims in this matter.
Well, I called my Senator, and the staffer who answered the phone was not terribly interested in my trying to explain my whole theory of danger in the hearings.. she was just hot to get me off the phone. Once she figured out that I was against a specific piece of legislation, that's all she wanted to hear.
Of course, that piece of legislation has not been introduced, and it may well have been significantly changed since Declan McCullagh got ahold of the draft, so I don't know if the staffer really had any idea what I was talking about.
So, I'd definitely commend people to call, but try and make up a pithy script ahead of time. I figure they'll give you about 30 seconds before they start looking at their watch, tops.;-)
The list is just of those Senators who are on the Senate Commerce Committee. There's no a priori assumption that any of these Senators (aside from Senator Hollings) actually support the SSSCA or anything like it.
But what would we say if we called? This is a hearing, there's no proposed legislation on the table here. Yes, I know and you know that Fritz has the SSSCA in his left pocket and Disney money in his right, but until legislation is brought up, I'm not sure what I could tell my Senator.
Do you think that we should object to these hearings?
So that ALL of the parsing/paining logic (as well as javascript) would behave EXACTLY as IE
With what IE specification?
Mozilla is shooting for the W3C specs, which have the virtue that they do exist. Mozilla actually does have a 'broken HTML compatibility mode', which it will use if a given HTML page doesn't specify a modern HTML DTD.
Re:.NET good, not evil
on
What is .NET?
·
· Score: 3, Interesting
Yes, Roger Sessions' article does make that point, and the guy running that JVM languages page is quite honorable to link to the criticism.
However, saying that the JVM isn't a suitable substrate because a lot of the language tools written on it are experimental is sort of a non sequitor.
So if you copy a GPLd work like that, there are only two possibilities - either you have accepted the license, or you are infringing on the copyright.
Well, that may mean that everyone copying GPL'ed works are infringing copyright, then. Or, a judge might rule that since RMS et al released their code into the wild, that the code is effectively public domain, because the effectively public domain release doesn't grant the right to restrict people's behavior without negotiating/signing specific license documents with individuals copying GPL'ed works.
I don't believe that this is the case, I'm just parroting what I've read elsewhere as being a concern. It's not unknown in the legal world for terms put on a contract or other document to be considered unenforceable due to other provisions of law. If parts of the GPL were ruled unenforceable, there could be a problem.
IANAL and I have six years of my life licensed under the GPL, FWIW.
If I have a piece of software in my hot little hands, and it's not governed by any licence at all, that means I can do what I want with it. Don't trick yourself. If any part of a license is invalid, the rights reserved by the producer always decrease and the rights assigned to the user always increase.
That's not what copyright law says. Sometime back they changed the copyright law (Boerne convention?) so that no filing or declaration of copyright was required to place an item under copyright. If you have something that you didn't write and it is not in the public domain and you don't have a license to copy it, then you would be in violation of copyright to do so.
Now, the real question is: can a COPYING file in a tar file count as a license, when the user doesn't need to sign a contract or indicate their affirmative agreement to the license terms in any way? This might get touched on by a revision of EULA practices, and the GPL might well be ruled to be a legally deficient mechanism. That would require some judge to decide that sharing of software must be either public domain or on some sort of contract basis, however, which seems unlikely to occur.
Softman was being bundled Adobe packages and breaking them up and selling them individually for a profit. This is something different than Joe User selling his copy of Windows 98 that he got with his computer that he doesn't use. This is a company basically ripping off Adobe.
How on earth so? Adobe got paid for however much software was in the bundle, at a price set by Adobe. If the purchaser turns around and sells the individual packages for more than (bundle price)/(items in bundle), then Adobe left money on the table by selling the bundle too cheaply. Whyfore should that be anyone's problem but Adobe?
I define those who view EULA's just pieces of paper and justification for stealing software to be no worse than the corner store robber last night, who made off with 25 dollars. But, you dear friend are robber more than 150 dollars+ each time. Understand this.
So who said that being in favor of the maintenance of the First Sale doctrine means that one is in favor of stealing software?
If Softman did not install the software that came bundled with his system at all, why shouldn't he be able to sell it to someone else? That's what's at issue here. It's like you saying that a K-Mart shopper who buys a CD, then turns around and resells it unopened for half-off is breaking the law.
Oh, I didn't realize that C++ compilers fell together by random chance, with no guiding intelligence! I'll bet Bjarne Stroustroup will be downright disappointed about that, eh? Think about what the hell you are saying.
Get a grip. You said the following:
So the mechanism that produces protein is an extremely precise machine that is itself made of protein. How did it originate? What produced the protein in the protein-synthetic apparatus? Did it originate with "random" proteins. Ya, I know you've got it all figured out in your little mind, but the problem is that your "random" proteins do not exist. Oh, I know, they once existed, but now they are all gone. Isn't that too bad. The story always seems to be the same, doesn't it?
He was responding to this by giving you an analagous circumstance, where a (simpler) mechanism existed that allowed for bootstrapping of the system. When the more complex system built on top of the simpler became complete enough to build itself (the C++ compiler in his example) the earlier scaffolding was no longer needed.
This is also the primary refutation of Michael Behe's "Darwin's Black Box" thesis, which I have indeed read in full.
No one suggests that the complex protein, DNA, and RNA system spontaneously self-assembled in a single step. Creationists who throw around impossibly large numbers to show the Vast improbability of that happening are indeed correct. The only way that such a phenomenon could be explained without saying "god did it" is to accept one of the following:
Impossibly good luck (god, again)
A more Vast amount of time and space than we believe the Earth had to spontaneously hit upon the full modern replicator system. (seeds from space, perhaps).
The existence of simpler replicating systems that were amenable to evolution over time.
It's that last possibility that most scientists working in the field believe to be the most likely. All you need to trigger the evolutionary feedback loop is some configuration of chemicals that can make it more likely that a simliar configuration of chemicals will be produced in the immediate vicinity. Once you've got that, then any change that makes the replication process more efficient will 'catch', and life is off to the races. Eventually, the more refined replication process may lose the pieces that were needed at the beginning, and wa-la, you've got a C++ compiler (RNA, DNA, proteins) that can compile itself (reproduce) without keeping the vestigal pieces that made the production of the compiler a simpler task.
That's complex, surely, but if you have water and energy and carbon compounds and a billion years of time, who's to say it's impossible? The most primitive, fragile form of life imaginable would still be life so long as any kind of replication, no matter how slow, fragile, or inefficient, was there.
I'm fairly certain that what he meant to ask, using your own analogy, was "Where did the students come from?" Biology is happy saying "We need this much time for our theory to work out," whereas Physics says "Nature HAS to have had a beginning." Anything else is spiritual.
No, he asked how information arose from mutation. My answer was that it didn't, that the information comes from the environment as nature rewards or punishes the random mutation.
Now, you and I know that evolutionary theory says nothing about the origin of the universe, matter, etc., but that's okay. Scientists are perfectly willing to say "I don't know, and I don't know how I ever could know" for such ultimate questions. That doesn't mean that we shouldn't trace the chains of evidence as far back as we can.
The idea that random mutations can turn a functioning gene into another functioning gene (with no fatal in between states) makes exactly as much sense as the idea the random bit mutations can turn a functioning method into a new working method with a different function (without core dumping in the process).
Never heard of genetic algorithms? They do precisely as you suggest. The fact is, the vast majority of code is not evolved in the sense of vast amounts of mostly faithful replication strewn with the occasional mutation and a population big enough that it can withstand genetic failures without threatening the entire population.
Nature is extremely subtle. One of the things that all living organisms have in common is that their genetic mechanisms have proven to be amenable to some mutation. An organism that was so finely tuned and so brittle that *any* change in its genome would be fatal would be a strong rebuff to evolutionary theory. The fact is that organisms are just not that fragile. Your code is, my code is, Bill Gates' code is, but none of us developed our code under the same conditions that nature developed life on the planet.
The theory is that the genetic mechanism has itself been selected for evolvability. Why else would we have diploid gene pairs? Why else would DNA have the base pair redundancy? Why else would the genome have 'trigger points' which can make for large body changes with small mutations?
Have you ever thought about how stupid you will feel as you are burning in hell?
But what about these classics:
Have you ever thought about how stupid you will feel as Anubis weighs your heart and finds you wanting after your death?
Have you ever thought about how stupid you will feel as you are reincarnated as a cockroach for your arrogance against the gods?
Have you ever thought about how stupid you will feel when you learn after your death that L. Ron Hubbard was right?
Have you ever thought about how stupid you will feel as Allah takes a dump on you thrice daily after you die?
Have you ever thought about how stupid you will feel as you play bridge with Zeus in Olympus and he laughs at you for not believing in him?
Have you ever thought about how stupid you will feel as you are welcomed to heaven for being kind, compassionate, thrifty, brave, trustworthy, scientifically literate, clean, and honest?
etc.?
Why should anyone believe in life after death? Where's the evidence? Where's the logic? Do computer programs go to hell or heaven when they stop being executed?
Slashdot Mirror
Unfortunately we're keeping the tracking to a bare minimum so I don't believe that we can do that. We're trying to keep updates to the DB to a bare minimum so we don't have a new bottleneck.
Reasonable, again.
I do wonder, however, about your statistics on page impressions.. I know that when I'm active in a comment discussion (which is several times a day), I can go through hundreds of comment pages. At a half cent a page, that would add up quickly. I understand you're saying that most of your users just hit the home page and link off to whatever is new. I hope that you'll structure your advertising for that, and make the biggest, most obnoxious ads appear on the home page rather than on each comment viewed.
Interestingly, I note that when I click on a comment, the HREF # scrolls the page down so I can't even see the ads that are there currently. Surely this can't continue, or are you merely assuming your advertisers don't know that a lot of ad impressions you provide are never actually seen by the user?
I'm not _worried_ about it, I just want to know how pervasive its use is.
And yeah, I know what gzip is. I just wasn't aware that some (many?) web servers were autonegotiating its use behind the scenes.
We don't expect users to pay a half cent for every page. We expect that what will happen is that users will suppress ads on just articles or the homepage effectively costing a few cents a day.
Sweet reason and good sense. I'm happy.
Is there any way that you can display page view statistics even for pages we view with ads? As I said, right now I'm leaving ads enabled for everything. I'd kinda like to be able to know how many page views I'm getting with ads so that I can better judge whether I should turn the ads off for awhile and drain down my slashdot account and put more money in. Otherwise I can well imagine leaving all the ads on permanently, and just putting in money to my slashdot account at whatever interval makes me feel happy and that lets me feel like I'm contributing in a meaningful way.
Of course, X-10 pop-unders would take care of any such impulse for altruism pretty quickly. ;-)
And that would be different from this proposal how?
I've never heard of mod_gzip before. Which browsers support it? How can we tell if we have downloaded a gzip'ed page?
Of course, this begs the question as to whether Taco/OSDN will have to start restricting the RDF feeds and/or content scrapers so that people can't suck the editorial content from Slashdot to feed such a site.
I'm up on the paid site thing. In fact, I just paid for 10,000 page views.
Then I promptly went and turned all the ads back on. Why? Because a) the ads aren't annoying yet.. in fact, since they are no longer just the OSDN ones, they're actually more interesting these days, and b) because I don't think I can afford to pay a half cent for each page I view on the entire web.
And that's what we're really talking about, here, isn't it? If slashdot goes pay-per-click, why shouldn't every other web site do so as well? If every web site did so, and each was a half-penny per click, I imagine that would easily amount to 15 bucks a day for me based on my current browsing habits. That could be as much as $450 a month.
Psychologically, I'm happy to pay a lifetime subscription to Salon for only $50. And I have an all-you-can-eat yearly subscription to The Economist for only $125, which includes an actual magazine I get delivered to my home every week. I don't mind those because the expenditure is controlled, and the meter isn't continuously running at such a high rate as a half penny per page.
Does it really cost half a penny for each page impression served by Slashdot, Taco? Were the ads making anywhere near that much?
And I see that I'm getting new ads for each step in the posting process (preview, etc.). Does that mean that it could a couple of cents to post to slashdot?
That would be one way to tune out the trolls, I guess. <smirk>
Now, what about our profit sharing? ;-)
Sigh.
This would be a lot more funny if Jack Valenti wasn't selling the Senate Commerce Committee precisely this line of reasoning right now.
Having worked in a member's office, they are swamped with mail and usually won't read the letter except to categorize it as "SSSCA - No"
Well, poop. I read that they produce a correspondance report for the member.. hopefully there'll be at least a chance that it gets through.
I guess I'll just have to wait until something is actually introduced and then try to put the smack down on all of my representatives about it.
Thanks for the inside dope.
I wrote the following to my senator, which captures the sense of my concerns far better than a 30 second phone call could ever do:
I am writing to you to express my opposition to any new legislation regarding digital copyright as sought by the film and recording industries. In particular, I am a computer programmer in vehement opposition to any legislation that would make it a crime to create or distribute any digital devices that do not include government mandated Digital Rights Management software. The point of such legislation, in conjunction with present law, is to attempt to give the film and recording industries absolute power over how anyone may view or use copyrighted material, without regard for the Supreme Court's finding in the Betamax case that citizens may engage in certain fair use practices.
Further, if such legislation were passed, there are grave dangers having to do with establishing a patent-based monopoly on the DRM standard, and grave dangers having to do with effectively outlawing any computer device that could be programmed by the user. The DMCA already makes it illegal to distribute tools to break protection, regardless of fair use. That is controversial, but manageable.. systems that don't touch protected material need not be affected, and would simply forego access to such content. An affirmative duty that all digital systems include DRM content controls would be incredibly far-reaching, and could be construed as banning any system (such as a PC running Linux) in which the user has complete control over the configuration and details of their system.
Hollywood and the recording industry are asking for a tremendous amount of control over technology in the United States (and the world, through their lobbying of WIPO), and it is not at all clear that it is in the nation's interest that they be given it.
I'm writing this on the occasion of tomorrow's upcoming digital copyright hearings of the Senate Commerce Committee. I hope you will be extremely vigilant in assessing the industry's claims in this matter.
Thank you sincerely,
Me
City, State
Well, I called my Senator, and the staffer who answered the phone was not terribly interested in my trying to explain my whole theory of danger in the hearings.. she was just hot to get me off the phone. Once she figured out that I was against a specific piece of legislation, that's all she wanted to hear.
Of course, that piece of legislation has not been introduced, and it may well have been significantly changed since Declan McCullagh got ahold of the draft, so I don't know if the staffer really had any idea what I was talking about.
So, I'd definitely commend people to call, but try and make up a pithy script ahead of time. I figure they'll give you about 30 seconds before they start looking at their watch, tops. ;-)
The list is just of those Senators who are on the Senate Commerce Committee. There's no a priori assumption that any of these Senators (aside from Senator Hollings) actually support the SSSCA or anything like it.
But what would we say if we called? This is a hearing, there's no proposed legislation on the table here. Yes, I know and you know that Fritz has the SSSCA in his left pocket and Disney money in his right, but until legislation is brought up, I'm not sure what I could tell my Senator.
Do you think that we should object to these hearings?
So that ALL of the parsing/paining logic (as well as javascript) would behave EXACTLY as IE
With what IE specification?
Mozilla is shooting for the W3C specs, which have the virtue that they do exist. Mozilla actually does have a 'broken HTML compatibility mode', which it will use if a given HTML page doesn't specify a modern HTML DTD.
Yes, Roger Sessions' article does make that point, and the guy running that JVM languages page is quite honorable to link to the criticism.
However, saying that the JVM isn't a suitable substrate because a lot of the language tools written on it are experimental is sort of a non sequitor.
Or the utterly wonderful Aussie indie film, The Interview.
If you haven't see it, it's very, very, very much worth seeking out for a rental. The DVD is pretty spiff, too.
So if you copy a GPLd work like that, there are only two possibilities - either you have accepted the license, or you are infringing on the copyright.
Well, that may mean that everyone copying GPL'ed works are infringing copyright, then. Or, a judge might rule that since RMS et al released their code into the wild, that the code is effectively public domain, because the effectively public domain release doesn't grant the right to restrict people's behavior without negotiating/signing specific license documents with individuals copying GPL'ed works.
I don't believe that this is the case, I'm just parroting what I've read elsewhere as being a concern. It's not unknown in the legal world for terms put on a contract or other document to be considered unenforceable due to other provisions of law. If parts of the GPL were ruled unenforceable, there could be a problem.
IANAL and I have six years of my life licensed under the GPL, FWIW.
If I have a piece of software in my hot little hands, and it's not governed by any licence at all, that means I can do what I want with it. Don't trick yourself. If any part of a license is invalid, the rights reserved by the producer always decrease and the rights assigned to the user always increase.
That's not what copyright law says. Sometime back they changed the copyright law (Boerne convention?) so that no filing or declaration of copyright was required to place an item under copyright. If you have something that you didn't write and it is not in the public domain and you don't have a license to copy it, then you would be in violation of copyright to do so.
Now, the real question is: can a COPYING file in a tar file count as a license, when the user doesn't need to sign a contract or indicate their affirmative agreement to the license terms in any way? This might get touched on by a revision of EULA practices, and the GPL might well be ruled to be a legally deficient mechanism. That would require some judge to decide that sharing of software must be either public domain or on some sort of contract basis, however, which seems unlikely to occur.
Softman was being bundled Adobe packages and breaking them up and selling them individually for a profit. This is something different than Joe User selling his copy of Windows 98 that he got with his computer that he doesn't use. This is a company basically ripping off Adobe.
How on earth so? Adobe got paid for however much software was in the bundle, at a price set by Adobe. If the purchaser turns around and sells the individual packages for more than (bundle price)/(items in bundle), then Adobe left money on the table by selling the bundle too cheaply. Whyfore should that be anyone's problem but Adobe?
I define those who view EULA's just pieces of paper and justification for stealing software to be no worse than the corner store robber last night, who made off with 25 dollars. But, you dear friend are robber more than 150 dollars+ each time. Understand this.
So who said that being in favor of the maintenance of the First Sale doctrine means that one is in favor of stealing software?
If Softman did not install the software that came bundled with his system at all, why shouldn't he be able to sell it to someone else? That's what's at issue here. It's like you saying that a K-Mart shopper who buys a CD, then turns around and resells it unopened for half-off is breaking the law.
That can't be right, can it?
Of course, 5 seconds with Google turns up what looks like some pretty effective refutation of the Polonium Halo theory.
Oh, I didn't realize that C++ compilers fell together by random chance, with no guiding intelligence! I'll bet Bjarne Stroustroup will be downright disappointed about that, eh? Think about what the hell you are saying.
Get a grip. You said the following:
So the mechanism that produces protein is an extremely precise machine that is itself made of protein. How did it originate? What produced the protein in the protein-synthetic apparatus? Did it originate with "random" proteins. Ya, I know you've got it all figured out in your little mind, but the problem is that your "random" proteins do not exist. Oh, I know, they once existed, but now they are all gone. Isn't that too bad. The story always seems to be the same, doesn't it?
He was responding to this by giving you an analagous circumstance, where a (simpler) mechanism existed that allowed for bootstrapping of the system. When the more complex system built on top of the simpler became complete enough to build itself (the C++ compiler in his example) the earlier scaffolding was no longer needed.
This is also the primary refutation of Michael Behe's "Darwin's Black Box" thesis, which I have indeed read in full.
No one suggests that the complex protein, DNA, and RNA system spontaneously self-assembled in a single step. Creationists who throw around impossibly large numbers to show the Vast improbability of that happening are indeed correct. The only way that such a phenomenon could be explained without saying "god did it" is to accept one of the following:
It's that last possibility that most scientists working in the field believe to be the most likely. All you need to trigger the evolutionary feedback loop is some configuration of chemicals that can make it more likely that a simliar configuration of chemicals will be produced in the immediate vicinity. Once you've got that, then any change that makes the replication process more efficient will 'catch', and life is off to the races. Eventually, the more refined replication process may lose the pieces that were needed at the beginning, and wa-la, you've got a C++ compiler (RNA, DNA, proteins) that can compile itself (reproduce) without keeping the vestigal pieces that made the production of the compiler a simpler task.
That's complex, surely, but if you have water and energy and carbon compounds and a billion years of time, who's to say it's impossible? The most primitive, fragile form of life imaginable would still be life so long as any kind of replication, no matter how slow, fragile, or inefficient, was there.
I'm fairly certain that what he meant to ask, using your own analogy, was "Where did the students come from?" Biology is happy saying "We need this much time for our theory to work out," whereas Physics says "Nature HAS to have had a beginning." Anything else is spiritual.
No, he asked how information arose from mutation. My answer was that it didn't, that the information comes from the environment as nature rewards or punishes the random mutation.
Now, you and I know that evolutionary theory says nothing about the origin of the universe, matter, etc., but that's okay. Scientists are perfectly willing to say "I don't know, and I don't know how I ever could know" for such ultimate questions. That doesn't mean that we shouldn't trace the chains of evidence as far back as we can.
The idea that random mutations can turn a functioning gene into another functioning gene (with no fatal in between states) makes exactly as much sense as the idea the random bit mutations can turn a functioning method into a new working method with a different function (without core dumping in the process).
Never heard of genetic algorithms? They do precisely as you suggest. The fact is, the vast majority of code is not evolved in the sense of vast amounts of mostly faithful replication strewn with the occasional mutation and a population big enough that it can withstand genetic failures without threatening the entire population.
Nature is extremely subtle. One of the things that all living organisms have in common is that their genetic mechanisms have proven to be amenable to some mutation. An organism that was so finely tuned and so brittle that *any* change in its genome would be fatal would be a strong rebuff to evolutionary theory. The fact is that organisms are just not that fragile. Your code is, my code is, Bill Gates' code is, but none of us developed our code under the same conditions that nature developed life on the planet.
The theory is that the genetic mechanism has itself been selected for evolvability. Why else would we have diploid gene pairs? Why else would DNA have the base pair redundancy? Why else would the genome have 'trigger points' which can make for large body changes with small mutations?
Have you ever thought about how stupid you will feel as you are burning in hell?
But what about these classics:
Why should anyone believe in life after death? Where's the evidence? Where's the logic? Do computer programs go to hell or heaven when they stop being executed?