Indeed. It wasn't its technical merits that made the Linux kernel what it is today, it was copyleft. That's why copyleft is popular (the 'herd mentality', as you put it). I don't think there's a 'lowest common denominator' issue here really, unless you're referring to over-strict copyleft.
GPL doesn't stop anyone from releasing code MIT/Apache/BSD/etc, but the most common code is GPL, so if you want to contribute to what is popular, then you pay the price of not having truly free.
Right, but that's the point - copyleft'ed software can 'snowball' in a way copycenter'ed software can't, precisely because no-one is allowed to release a closed fork. Compare FreeBSD vs Linux.
I think it was Stallman who put it this way: if you ban slavery, the total amount of freedom increases. So it is with copyleft (at least that's the intention, but I'd say it's been shown to work pretty well).
What selfless person is going to do this for free? Even completely free and open source P2P software would be limited by the same types of agreements that Comcast and Verizon and friends have today - what company wants to risk being sued for contributory copyright infringement
Don't we already know the answer to this? BitTorrenters. I don't see that Comcast and Verizon have been all that effective against it, but I'm not very clued in on the specifics there.
The torrent website can be for-profit if it's hosted in a country that turns a blind eye. Most or all of the seeds can be non-profit, so it's all legal by your downloader-himself-has-done-nothing-illegal system.
I'm not sure it is possible to invent a whole class of property without complex rules
I suspect this is the case. Morality isn't simple. I doubt well-functioning law can be either.
I now have to know the details of how torrent works and the licensing conditions - explicit or statutory - of the software. You know what? That is too burdensome, and the fact that I can have my economic life basically ruined by this single act is unconscionable.
The punishments the RIAA/MPAA seek have a record of being grossly unfair, certainly. Also agree that miles-long Terms and Conditions are too much for most people to be reasonably expected to cope with.
Literally every person in my age demographic has violated copyright law in such a way. I cannot think of a better measure of the illegitimacy of a law.
Fair point, but you've not responded to my central argument: one cannot simply allow personal consumption to be exempt from copyright law without demolishing whole industries.
I think that copyright law is too complex to burden the average person with. I don't think it is ethical to apply a law to the common man that cannot be fully described by lawyers who have spent their entire career studying it.
What are you referring to here? I think most people know torrenting Hollywood movies and video games is against the law.
Either simplify it greatly or make it apply to commercial trade only. Keep in mind that this would still close entities like Napster and isoHunt, since they are definitely in the commercial realm.
And not-for-profit BitTorrent sites? Your it's fine if it's just personal use stance doesn't deal with how it could remain profitable to produce intellectual works which are targeted exclusively/near-exclusively at personal use, such as video games and movie DVDs.
Honestly I don't think it would change much at all, except the few poor souls who got burned at the stake by the xIAAs as an example to us all would still have their quality of life intact.
Disagree for reasons given above, but agree that the RIAA/MPAA can burn for some of the shit they've done.
The problem is that the game is never actually released to the public. Because the code is never made available, except in a few rare cases. So the 'contract' with the public that gave them copyright is now violated because we don't get ever get the content into the public domain.
I disagree. I think it's reasonable to view proprietary software's source-code as nothing more than an internal development aid used by the software-house, even though certainly there would be advantages to eventually making source-code public-domain.
How about games that require activation servers? How in the hell will you get to play the game in 75 years when said servers are long since dead?
There needs to be a repository where software code gets placed so that when it's copyright is expired it gets released to the public. Or something like that so we actually get the creators to honor their end of the deal.
Viewing video-games (and indeed DRM'ed software in general) as art, or at least of historical interest, this is indeed worrisome, but I don't agree that software producers should be forced to hand over their source-code to the government. For one, it breaks the elegant simplicity that you don't have to do anything special to have the copyrights to your work. Also, I'd never trust a government database of source-code not to be hacked, releasing (legally unprotected) trade-secrets prematurely.
Regarding software patents, I share your view (the prevailing view on Slashdot) that they're probably generally a Bad Thing. Here is the best counter-argument I know of. (In short, that one can't forbid software patents without forbidding hardware patents. How New Zealand deals with the issue, I'm not sure.)
I never understood this line of thought that but it's just for personal use so that makes it ok. It doesn't. The copyright owners have the right to charge for their film, including 'personal use' (indeed, this is almost the entire point of releasing on DVD). Today's copyright law rightly grants them this prerogative.
Your position boils down to copyright law should only protect intellectual works aimed at the non-'personal' level of consumption. i.e. that CAD/CAM software should get copyright protection, but Hollywood movies should not.
I don't buy it.
(To be clear, I'd certainly be in favour of reduced copyright duration (15 years would be generous enough, in my opinion), but I don't see any merit to the argument that copyright violation on the personal level shouldn't be considered a copyright violation.)
It's analogous though. I understand the risk in my linking to the civil rights movement - didn't want to be pushing at the if you don't agree with me you're racist fallacy.
You're too generous - I've just read a bit about it and played about with it a little; I'm certainly no D master:P
Regarding frameworks/libraries/maturity/practical power, I think you might underestimate what Mono is capable of - I'm no expert, but I don't think you'll have any trouble doing web-dev work with it.
For D, I believe there are libraries out there for doing web-dev work. This one might do, but looks a bit on the low-level side. For serious production code I'd certainly recommend C# over D (ASP.NET is already out there in the 'real world'), but I'm sure it could be done with D.
Hm, should check how D is in the area of multi threading.
I don't know much about this, but I do know that D2 has some thread-safety features built into the language itself: 'global' variables are now by default actually merely 'thread-global'. You have to use a special keyword if you want real process-wide globals. I can see where they're coming from, but I'm not sure I like it.
But bottom line I don't like many of the design decisions in D
There are certainly seem to be some warts - I get the impression the const system is a bit broken.
like having no true MI
I've never been bitten by not having MI, but maybe that's because I learnt Java before I learnt C++. The arguments for not having MI are fairly persuasive imo: simpler for the compiler, generally less error-prone for the user, and MI is rarely the right solution anyway.
I never quite got my head around Scala's solution (it looked a lot like MI to me), but people seem to like it.
(with language versus platform I mean the broad spectrum of libraries that are focused on special purposes in Java, like JPA or the Java Rest and Soap support via annotations etc. the huge amount of third party frameworks like spring or Aspect/J or hibernate)
Ah, the advantages of bytecode and reflection. Sounds almost like Aspect Oriented Programming is what you're really getting at here. There's not much in the way of AOP for D, from what Google tells me.
My boss/familyare strong Republicans and they know I'm not. And you know what? Nobody really cares.
Existence does not imply universality. Other bosses/families might be, you know, different from yours.
Open voting will eliminate every current avenue of fraud and introduce *no* new avenue of fraud. It really is that simple.
I don't buy it. Fundamentally, if it is possible for a third party to know who you voted for, it is possible for them to incentivise a particular vote/non-vote. I can see that an 'open voting' system would help combat other voting issues (being able to check that one's vote has been properly registered would be a nice feature), but the downside I've pointed out is not a trivial one.
Why do you hate the USA?
Oh shut up. Does this look like a right-wing talk-show to you?
Firstly, I've already said I'm not American; you won't get far by questioning my American patriotism.
Also, you seem to be forgetting that America abandoned the 'open voting' system over a century ago - if we wish to indulge the logic used by talk-show hosts to pander to their moronic viewers, we could well conclude it is you who in fact 'hates America' for disapproving of their current-day voting system.
No, my point was that no-one should know which way you voted, precisely because of how many people are likely to be putting some sort of pressure on you. The poll workers, therefore, should not know which way you voted.
We'be had multiple presidential elections decided by vote error margins
This isn't an indication of voter apathy. A two-party contest is a predictable, indeed pretty much inevitable, outcome of the voting system.
and nobody wants to change the system (nobody being 3 insane Internet bloggers, and nobody else)
I wonder what the real numbers are on this issue. People that genuinely love the Democratic Party or the Republican Party (rather than just thinking in 'lesser of two evils' terms) would likely want the current system to stay even if they fully understood what it's doing.
I suspect you're right, though, that not nearly enough voters know/care about the voting system and its effects.
And why are you so ashamed of your vote? Quit voting for Satan, and you won't have to be ashamed.
Seriously? You've totally missed the point. 'Voting for Satan' is subjective, and voter pressure is a real issue. If your boss/family is strongly republican, it could put you in a difficult position if they found out you voted Democrat, no?
So you like to suggest that I suggest to the amazon CTO that we do the next system upgrades in D ?
"Nobody ever got fired for choosing Java", right? Fair point, but you asked What other language/platform is out there that could rival Java, and I submit that the C#/CLI language/platform pair, and the D language, could. They're by no means a million miles away from Java; if we look at the broader spectrum of programming languages, the three are pretty close.
There are so many D programmers?
Again, fair point: D ranks at 37 on TIOBE, where Java is in second place to C. It depends what you mean by could rival Java, really; if you are asking for a language with considerable pre-existing adoption, that's of course quite different from the technical side, which is what I was looking at.
And we already have a D compiler for the AS400 and the IBM Mainframes?
For funky platforms, Java will beat D, of course. For the major server/desktop platforms (Lin/Win/Mac), D should work (though it generally focuses on Windows), but yes, Java certainly wins on implementation maturity.
And it gives us over Java... what exactly? A servlet framework?
It gives you a language which rivals Java. Basing my decision purely on language preference, I'd choose D over Java.
D is a language, not a platform.
It's a little fuzzy what this means - D doesn't have a security model like Java does, sure, but as I understand it, CLI doesn't have one either. Compilation is fully ahead-of-time, also. This doesn't rule out D from doing some of the things Java is used for, though.
I'm not American, but as I understand it poll workers don't know who specifically you're voting for. That's the way it is in the UK, at least. The reason for this is precisely to avoid having to trust them not to pressure you.
You're right that ultimately, yes, you have to trust the system, but regarding the specific issue of voter-pressuring, you don't want anyone else to know who you voted for, whether it be poll workers, family and friends, your boss, etc.
What would be even better is if all these manufacturers got together and worked out a common API. Then developers wouldn't have to write a different app for each device. It can't really be that hard.
Web standards tend to be out-paced by tectonic plates, but I'm sure we'll see one for 'native app' development eventually.
I haven't looked at the Firefox OS APIs, but I imagine the existing web standards (HTML5) cover most of what a developer could want, at least in theory.
Sounds to me like Ubuntu Phone, Windows Phone, and Firefox OS are all using web technologies in their 'native' toolkits.
What makes Firefox OS any different, such that you stand a reasonable chance of being able to reuse some of your code in your web project only with Firefox OS?
Yeah, so? Someone tries to buy or pressure my vote, and the FBI will hear about it, if they are even operating now. I hear they take that stuff seriously.
This is a non-answer. If 'the system' could be trusted, we wouldn't be having this discussion.
In assuming that the 'powers that be' are themselves opposed to such pressure (the way the FBI presumably are), you ignore the real possibility that it might be them who are putting pressure on voters. I know nothing of Azerbaijan's politics, but it seems that's what's going on here. Their elections are broken. They have no FBI.
This is a Who will guard the guards? question, to which the answer is certainly not No problem, I trust the guards.
No, you're claiming, and now reinforcing your claim, that there is a grand divide between high level and assembly level programming. Call it 'programming microcontrollers' if you insist on your own straw man.
No, TheRaven64 was saying that writing assembly code for a microcontroller is quite different from writing assembly code for a modern CPU.
The examples Raven64 gave of the differences are valid: modern CPUs have FPUs and several layers of cache, where microcontrollers do not.
Unless the police are so corrupt that they commit more crime than they solve/deter, I don't see many places volunteering for this science.
Also, it's actually not scientific: you're failing to control the other variables. If you can guarantee the job situation won't change, etc, then you might have something.
who wouldn't dream of being robbers themselves or allowing some particular group robbers, or being part of the local mafia
Well, no, not if they want to keep their part-time security job.
Unlike a police officer, they personally would be hired by the specific people they are there to protect. If anything that makes corruption less likely, surely.
Slashdot Mirror
GPL has done great things
Indeed. It wasn't its technical merits that made the Linux kernel what it is today, it was copyleft. That's why copyleft is popular (the 'herd mentality', as you put it). I don't think there's a 'lowest common denominator' issue here really, unless you're referring to over-strict copyleft.
GPL doesn't stop anyone from releasing code MIT/Apache/BSD/etc, but the most common code is GPL, so if you want to contribute to what is popular, then you pay the price of not having truly free.
Right, but that's the point - copyleft'ed software can 'snowball' in a way copycenter'ed software can't, precisely because no-one is allowed to release a closed fork. Compare FreeBSD vs Linux.
I think it was Stallman who put it this way: if you ban slavery, the total amount of freedom increases. So it is with copyleft (at least that's the intention, but I'd say it's been shown to work pretty well).
What selfless person is going to do this for free? Even completely free and open source P2P software would be limited by the same types of agreements that Comcast and Verizon and friends have today - what company wants to risk being sued for contributory copyright infringement
Don't we already know the answer to this? BitTorrenters. I don't see that Comcast and Verizon have been all that effective against it, but I'm not very clued in on the specifics there.
The torrent website can be for-profit if it's hosted in a country that turns a blind eye. Most or all of the seeds can be non-profit, so it's all legal by your downloader-himself-has-done-nothing-illegal system.
I'm not sure it is possible to invent a whole class of property without complex rules
I suspect this is the case. Morality isn't simple. I doubt well-functioning law can be either.
I now have to know the details of how torrent works and the licensing conditions - explicit or statutory - of the software. You know what? That is too burdensome, and the fact that I can have my economic life basically ruined by this single act is unconscionable.
The punishments the RIAA/MPAA seek have a record of being grossly unfair, certainly. Also agree that miles-long Terms and Conditions are too much for most people to be reasonably expected to cope with.
Literally every person in my age demographic has violated copyright law in such a way. I cannot think of a better measure of the illegitimacy of a law.
Fair point, but you've not responded to my central argument: one cannot simply allow personal consumption to be exempt from copyright law without demolishing whole industries.
This distinction is actually part of the law.
Is that with regard to degree of punishment, or does it actually allow more expansive actions compared to non-personal use?
I believe patent law doesn't apply to personal works, but I don't imagine there's any such dramatic exclusion in copyright law.
I think that copyright law is too complex to burden the average person with. I don't think it is ethical to apply a law to the common man that cannot be fully described by lawyers who have spent their entire career studying it.
What are you referring to here? I think most people know torrenting Hollywood movies and video games is against the law.
Either simplify it greatly or make it apply to commercial trade only. Keep in mind that this would still close entities like Napster and isoHunt, since they are definitely in the commercial realm.
And not-for-profit BitTorrent sites? Your it's fine if it's just personal use stance doesn't deal with how it could remain profitable to produce intellectual works which are targeted exclusively/near-exclusively at personal use, such as video games and movie DVDs.
Honestly I don't think it would change much at all, except the few poor souls who got burned at the stake by the xIAAs as an example to us all would still have their quality of life intact.
Disagree for reasons given above, but agree that the RIAA/MPAA can burn for some of the shit they've done.
The problem is that the game is never actually released to the public. Because the code is never made available, except in a few rare cases. So the 'contract' with the public that gave them copyright is now violated because we don't get ever get the content into the public domain.
I disagree. I think it's reasonable to view proprietary software's source-code as nothing more than an internal development aid used by the software-house, even though certainly there would be advantages to eventually making source-code public-domain.
How about games that require activation servers? How in the hell will you get to play the game in 75 years when said servers are long since dead?
There needs to be a repository where software code gets placed so that when it's copyright is expired it gets released to the public. Or something like that so we actually get the creators to honor their end of the deal.
Viewing video-games (and indeed DRM'ed software in general) as art, or at least of historical interest, this is indeed worrisome, but I don't agree that software producers should be forced to hand over their source-code to the government. For one, it breaks the elegant simplicity that you don't have to do anything special to have the copyrights to your work. Also, I'd never trust a government database of source-code not to be hacked, releasing (legally unprotected) trade-secrets prematurely.
Regarding software patents, I share your view (the prevailing view on Slashdot) that they're probably generally a Bad Thing. Here is the best counter-argument I know of. (In short, that one can't forbid software patents without forbidding hardware patents. How New Zealand deals with the issue, I'm not sure.)
I never understood this line of thought that but it's just for personal use so that makes it ok. It doesn't. The copyright owners have the right to charge for their film, including 'personal use' (indeed, this is almost the entire point of releasing on DVD). Today's copyright law rightly grants them this prerogative.
Your position boils down to copyright law should only protect intellectual works aimed at the non-'personal' level of consumption. i.e. that CAD/CAM software should get copyright protection, but Hollywood movies should not.
I don't buy it.
(To be clear, I'd certainly be in favour of reduced copyright duration (15 years would be generous enough, in my opinion), but I don't see any merit to the argument that copyright violation on the personal level shouldn't be considered a copyright violation.)
More than one Negro was killed for attempting to vote, even with anonymous voting. So, does that prove that anonymous voting is unsafe?
Fair point - if you profile people, you can pressure some of them not to even enter the booth. Another case of 'it works in a stable society'.
You have some fair points, but I'm not sure I agree with your assessment of the risk of vote-selling.
It's analogous though. I understand the risk in my linking to the civil rights movement - didn't want to be pushing at the if you don't agree with me you're racist fallacy.
you really don't want to see what happens when a substantial portion of society gets extremely angry
Not to Godwin the discussion or anything, but, uhh....
You definitely know a lot about D.
You're too generous - I've just read a bit about it and played about with it a little; I'm certainly no D master :P
Regarding frameworks/libraries/maturity/practical power, I think you might underestimate what Mono is capable of - I'm no expert, but I don't think you'll have any trouble doing web-dev work with it.
For D, I believe there are libraries out there for doing web-dev work. This one might do, but looks a bit on the low-level side. For serious production code I'd certainly recommend C# over D (ASP.NET is already out there in the 'real world'), but I'm sure it could be done with D.
Hm, should check how D is in the area of multi threading.
I don't know much about this, but I do know that D2 has some thread-safety features built into the language itself: 'global' variables are now by default actually merely 'thread-global'. You have to use a special keyword if you want real process-wide globals. I can see where they're coming from, but I'm not sure I like it.
But bottom line I don't like many of the design decisions in D
There are certainly seem to be some warts - I get the impression the const system is a bit broken.
like having no true MI
I've never been bitten by not having MI, but maybe that's because I learnt Java before I learnt C++. The arguments for not having MI are fairly persuasive imo: simpler for the compiler, generally less error-prone for the user, and MI is rarely the right solution anyway.
I never quite got my head around Scala's solution (it looked a lot like MI to me), but people seem to like it.
(with language versus platform I mean the broad spectrum of libraries that are focused on special purposes in Java, like JPA or the Java Rest and Soap support via annotations etc. the huge amount of third party frameworks like spring or Aspect/J or hibernate)
Ah, the advantages of bytecode and reflection. Sounds almost like Aspect Oriented Programming is what you're really getting at here. There's not much in the way of AOP for D, from what Google tells me.
My boss/familyare strong Republicans and they know I'm not. And you know what? Nobody really cares.
Existence does not imply universality. Other bosses/families might be, you know, different from yours.
Open voting will eliminate every current avenue of fraud and introduce *no* new avenue of fraud. It really is that simple.
I don't buy it. Fundamentally, if it is possible for a third party to know who you voted for, it is possible for them to incentivise a particular vote/non-vote. I can see that an 'open voting' system would help combat other voting issues (being able to check that one's vote has been properly registered would be a nice feature), but the downside I've pointed out is not a trivial one.
Why do you hate the USA?
Oh shut up. Does this look like a right-wing talk-show to you?
Firstly, I've already said I'm not American; you won't get far by questioning my American patriotism.
Also, you seem to be forgetting that America abandoned the 'open voting' system over a century ago - if we wish to indulge the logic used by talk-show hosts to pander to their moronic viewers, we could well conclude it is you who in fact 'hates America' for disapproving of their current-day voting system.
But I'm sure neither of us are that stupid.
Then don't tell them.
No, my point was that no-one should know which way you voted, precisely because of how many people are likely to be putting some sort of pressure on you. The poll workers, therefore, should not know which way you voted.
We'be had multiple presidential elections decided by vote error margins
This isn't an indication of voter apathy. A two-party contest is a predictable, indeed pretty much inevitable, outcome of the voting system.
and nobody wants to change the system (nobody being 3 insane Internet bloggers, and nobody else)
I wonder what the real numbers are on this issue. People that genuinely love the Democratic Party or the Republican Party (rather than just thinking in 'lesser of two evils' terms) would likely want the current system to stay even if they fully understood what it's doing.
I suspect you're right, though, that not nearly enough voters know/care about the voting system and its effects.
And why are you so ashamed of your vote? Quit voting for Satan, and you won't have to be ashamed.
Seriously? You've totally missed the point. 'Voting for Satan' is subjective, and voter pressure is a real issue. If your boss/family is strongly republican, it could put you in a difficult position if they found out you voted Democrat, no?
So you like to suggest that I suggest to the amazon CTO that we do the next system upgrades in D ?
"Nobody ever got fired for choosing Java", right? Fair point, but you asked What other language/platform is out there that could rival Java, and I submit that the C#/CLI language/platform pair, and the D language, could. They're by no means a million miles away from Java; if we look at the broader spectrum of programming languages, the three are pretty close.
There are so many D programmers?
Again, fair point: D ranks at 37 on TIOBE, where Java is in second place to C. It depends what you mean by could rival Java, really; if you are asking for a language with considerable pre-existing adoption, that's of course quite different from the technical side, which is what I was looking at.
And we already have a D compiler for the AS400 and the IBM Mainframes?
For funky platforms, Java will beat D, of course. For the major server/desktop platforms (Lin/Win/Mac), D should work (though it generally focuses on Windows), but yes, Java certainly wins on implementation maturity.
And it gives us over Java ... what exactly? A servlet framework?
It gives you a language which rivals Java. Basing my decision purely on language preference, I'd choose D over Java.
D is a language, not a platform.
It's a little fuzzy what this means - D doesn't have a security model like Java does, sure, but as I understand it, CLI doesn't have one either. Compilation is fully ahead-of-time, also. This doesn't rule out D from doing some of the things Java is used for, though.
I'm not American, but as I understand it poll workers don't know who specifically you're voting for. That's the way it is in the UK, at least. The reason for this is precisely to avoid having to trust them not to pressure you.
You're right that ultimately, yes, you have to trust the system, but regarding the specific issue of voter-pressuring, you don't want anyone else to know who you voted for, whether it be poll workers, family and friends, your boss, etc.
What other language/platform is out there that could rival Java? AFAIK none!
Depending on what you're after, C# (yes, with Mono, it's actually pretty good) or D might work for you.
What would be even better is if all these manufacturers got together and worked out a common API. Then developers wouldn't have to write a different app for each device. It can't really be that hard.
Web standards tend to be out-paced by tectonic plates, but I'm sure we'll see one for 'native app' development eventually.
I haven't looked at the Firefox OS APIs, but I imagine the existing web standards (HTML5) cover most of what a developer could want, at least in theory.
You can get a decent little brand new plane for about the same price as a loaded mid sized car these days ($30K-50).
In this case, the plane was leased by Flying Fox Aviation, based at Bagby Airfield near Thirsk in North Yorkshire, to Sandtoft Airfield and Flying School .
Sounds to me like Ubuntu Phone, Windows Phone, and Firefox OS are all using web technologies in their 'native' toolkits.
What makes Firefox OS any different, such that you stand a reasonable chance of being able to reuse some of your code in your web project only with Firefox OS?
breaking lease contracts with people that have already purchased rights to conduct business
Please explain. Non-American here, sorry if I've missed something obvious.
block scenic viewpoints on highways to prevent you from viewing national treasures
Again, please explain.
Yeah, so? Someone tries to buy or pressure my vote, and the FBI will hear about it, if they are even operating now. I hear they take that stuff seriously.
This is a non-answer. If 'the system' could be trusted, we wouldn't be having this discussion.
In assuming that the 'powers that be' are themselves opposed to such pressure (the way the FBI presumably are), you ignore the real possibility that it might be them who are putting pressure on voters. I know nothing of Azerbaijan's politics, but it seems that's what's going on here. Their elections are broken. They have no FBI.
This is a Who will guard the guards? question, to which the answer is certainly not No problem, I trust the guards.
No, you're claiming, and now reinforcing your claim, that there is a grand divide between high level and assembly level programming. Call it 'programming microcontrollers' if you insist on your own straw man.
No, TheRaven64 was saying that writing assembly code for a microcontroller is quite different from writing assembly code for a modern CPU.
The examples Raven64 gave of the differences are valid: modern CPUs have FPUs and several layers of cache, where microcontrollers do not.
Unless the police are so corrupt that they commit more crime than they solve/deter, I don't see many places volunteering for this science.
Also, it's actually not scientific: you're failing to control the other variables. If you can guarantee the job situation won't change, etc, then you might have something.
who wouldn't dream of being robbers themselves or allowing some particular group robbers, or being part of the local mafia
Well, no, not if they want to keep their part-time security job.
Unlike a police officer, they personally would be hired by the specific people they are there to protect. If anything that makes corruption less likely, surely.
I dare say buyers/suppliers are more fickle than you're giving them credit for.