I've used both and recommend them that way - if you want DVCS then use Git, if you want centralized then use SVN. Both are the top in their respective VCS markets. The only issue is if you use Windows - then Git just sucks as it does not integrate well at all, and as others have pointed out TortoiseGit doesn't really resolve that - it's a tooling problem and a problem in the design of Git. (One that could be probably be resolved by moving away from using so many Bash scripts and using a library for all the functionality like SVN does, which is why TSVN is just as first-class a citizen as the SVN command-line tools.)
And while I realize the they are very different - Git just drives me nuts, between things like having to stage files before a commit to it not tracking directory structures until there is something in it. (I like to setup the directory structure prior to putting things in it; often times I know I'll have something there but it may take a while to get to the point where I'll have something to put in it, yet I want the structure in the VCS before I get there.)
Really? Up here in Canadaland, the evidence required to get a warrant(aka R&PG and/or probable cause) require a chain of evidence when presented in court, because the defense must have full disclosure.
I believe it's the same in the US. Anything the Goverment uses to prosecute must be made available to the defense - classified or not.
The point I was making was not to state to create an unaltered copy of the original but get enough data on the variation of the copies to be able to mess up the watermark enough to render it useless. Random pick of formatting/wording in deviating sections from one of the N copies obtained at each case. The result may be that you have variants A, B and C as source and your scrambling causes it to look like variant K, so the buyer of variant K will be blamed until they figure out that they are chasing in the wrong direction.
Or by doing diffs on the text you might find the variations and be able to eliminate them, producing the original unvariated work and thus no watermark at all.
Except that Cogent earns sympathy for taking advantage of the real Tier 1's because the real Tier 1's only exist because of crony capitalistic efforts, and bullshit shenanigans that makes Cogent's shenanigans look like your dad's dick.
Quite small.
Tier 1's primarily exist because they were their first and had to invest in the market before there was really a market there to invest in. All Tier 1's have been around for a very long time (in the computer world). Cogents and others are relatively new to the market, so they simply don't have the infrastructure that the Tier 1's do. Could they build it? Yes; but for the moment they don't want to incur those very costly expenses.
As soon as someone takes on a monopoly (say telecoms, phama, oil, coal), those industries dump a pile of money on the GOP who go around talking about big government getting in the road of small businesses. The GOP faithful like this because it makes them feel like they have political power. It is all rather ironic.
And their unions dump a pile of money to the Dems as well.
Please, don't act like the GOP is the only party at fault. The Dems share just as much of the blame.
The other problem is that we get compared as the entire USA vs say Sweden. It's not really a fair comparison given geographic population distributions.
Not just that, but land-mass differences, etc.
Sweden is probably more comparable to say New Hampshire in some respects, California in others, and perhaps Oregon in others. Yet it's compared to the entirety of the US.
A more fair comparison would be US vs. Russia vs Europe vs Africa vs Mexico vs South America.
I've tried, when my wife and I first got married I was earning minimum wage back then I think it was around $5 or so an hour. We could barely feed ourselves, we got pregnant and I applied for financial aid, guess what supporting 2 people at the time with a 3rd on the way I was told I made too much money. Our grocery budget for a month $15.
I agree, the restrictions are insane, and actually encourage couples to not get married - you'd have gotten more support if the two of had divorced! (And they wonder why younger generations see no reason to get married.)
Kinda funny how all of the free-market lovers refuse to break up these telecom monopolies, or at the very least regulate them into being dumb pipes.
I love the free market; yet I also very much agree that the telecom industry needs a massive dose of breakup and regulation. For starters, Cable companies (and Fibre Optic services like FIOS and U-Verse) need to be forced to share their infrastructure.
Personally, I'd move it all to a single owner model - perhaps (even likely) owned by the localities - where companies have to lease them from the owner. So AT&T, Verizon, Comcast, etc, won't have any ability to control the basic pricing or infrastructure - they'd all be forced to compete on the quality of their services instead of their infrastructure.
It's time to realize that the internet Infrastructure really should belong to the people, and not be tied up in a wasteland of monopolies enforced by local governments - e.g. a small housing group should be able to get their own fibre optic line and split it among the group without having the county say "you can't do that because we signed this contract with company X and they have that sole right", that should all be illegal.
Start with Computer Engineering. It's rare that people who start on the finance side ever develop first-class CE skills, while good CE people have very little trouble picking up the necessary finance knowledge. Hell, a number of the high-frequency shops don't even have finance experience as a prerequisite, but you can bet every one of them wants rock-solid technical skills.
There are jobs that people really, really, really want to do for zero pay. Why wouldn't you allow them to make that decision for themselves?
Why don't we allow people to sell themselves into slavery?
Interns aren't allowed to walk away from a company that mistreats them? Modern slave owners are allowed to whip and kill their interns?
I did not know that.
Interns are generally required to complete a certain time period of work to qualify for course credits; they are also bound by employment laws and if terminating the internship early must comply like any other employee in the state where their internship is. For AT-WILL employment they could walk away at any time, but forfeit the course credit; for non-AT-WILL employment, they have to give notice as required by the law.
My internship cost me around $6000 between tuition and commuting. I was reimbursed for mileage spent on the job, but was otherwise unpaid (save for the odd case of Heineken the company president shared when we had to work late). It took up 50 hours a week, which I couldn't use to get a job to pay all of the above plus those little luxuries of life like food and rent. I basically financed everything on a credit card during my internship and a few minor freelance gigs I could land, which really hurt me financially for a couple of years.
The only one letting an intern be taken advantage of like that is the intern themselves. And yes, I didn't take internships that didn't pay - at the very least I had to cover food and expenses (namely gas) over the internship.
In being fair to the judge here, they aren't the first person to think this. Obama himself made similar statements, a manifesto of sorts, about internships needing to benefit only the student and basically be nothing but a burden to the employer.
I would hardly say that a paid internship is harmful to the employer. I never had an unpaid internship, and often made good money for the employer. At one position my "internship" was essentially a senior level position; and in the first month I enabled the company to bill out to customers the budget of they had for my internship. At most, I was doing the level of work of their normal staff, but at a lot cheaper rate than their normal staff. So yes, they saved money by hiring me for the internship over having a normal software position; and while I learned a lot, I also earned the funds I needed for college.
Well, the study was about browsers and not operating systems. I sort of assumed they tested all browsers on Windows in which case that is a controlled variable.
I'm sure they would sponsor a study looking at power usage of Linux vs. Windows vs. OS X. You probably won't see the results of their work unless it is in their favour, and you probably won't see the results of their work w.r.t. desktop Linux as it doesn't have the mindshare that would make it worthwhile to waste advertising space on (few know what Linux actually is, fewer would consider using it, and fewer still would consider using it, would make a decision based on power usage, and would trust a study funded by Microsoft to base a decision on). I say all this as a Linux/OS X user.
True, the study was about browsers on Windows. However, given that IE is integrated into Windows you need to also have a control variable for Windows too.
Time for a reality check: the incredibly vast majority of people don't give a shit about Other OS. It was a sucky move by Sony to remove it, no denying that, but it affected 0.001% of their user base and perhaps put off half of those if not less. They don't give a damn about a few angered/. posters who swore off Sony because they couldn't run Linux on their game consoles.
They might not at the moment, but that will have a big effect on PS4 sales, at which point they will. The question is, how big the effect will be.
Wouldn't that be a bit extreme? All the game asks you is to perform an update, or a simple check to make sure you don't have a rom loader installed. That isn't a problem to non pirates. I really don't see the problem with this, but even if there is a problem, if that's the only DRM PS4 has, I don't mind the nuisance.
The problem namely being discussed here was Sony's update to the PS3 that removed a feature of the PS3 that people wanted - OtherOS, which allowed running Linux directly in a supported manner. The update completely removed that functionality for everyone regardless of whether it was rooted or not, and that was all done after the PS3 was sold. Of course, IIRC Sony lost in court a class action against them over it, or at least settled in a favorable manner to PS3 owners (e.g. partial refunds or something). The concern is that they'll do something similar with PS4, even if they don't do so right out of the gate.
Dell, HP, IBM, and BCE are all windows shops, or at least were when I worked for them, and all of them have *well* over 10,000 employees. I still work for BCE. We've got 65,000 employees in Canada, and they're all on Windows. (well, some of the field repair technicians are being switched to Android, but that's still *well* over 10,000 Windows licenses).
FYI - HP and IBM at least have looked at replacing desktops with Linux. IBM has even gone to the extent of doing it with a sizeable chunk of their work force, probably to show governments it can be done. So just because they're a big employer doesn't mean that everyone there uses Windows, or that those that are not are a statistical blip.
The summary does a disservice by comparing the power usage to that required to make a cup of tea. If you read the article, it says "Laptops use about 14.7 Watts when idling. Firing up a browser adds another one or two Watts to this, depending which sites are browsed, and which browser you are using, Fraunhofer found. On desktop PCs, browsers add the same amount to the energy draw – but the baseline is around 37.8W."
The question remaining, however, is how much of that is hidden by running Windows? In other words, do the same thing with Linux (GNOME, KDE, etc) and Mac OSX. Does it really make a difference? Or is Windows merely hiding the extra usage by always having that usage included in the OS? Of course, don't expect Microsoft to fund such a study.
The increased resource requirements DO in fact do more with your computer. It means apps get developed in man-weeks or man-months rather than years or decades. It means we have a standard set of libraries shared between apps.
Not necessarily. If you use less efficient code, structures, memory usages, etc. then no you will not. You'll just do the same thing with more resources, and you may even spend more time building it.
Shared libraries do not necessarily mean extra code/resources at run-time. You may, for instance, use static linker level libraries that simply allow modules to be shared; or you may make additional DLLs that do have run-time requirements. But you have to determine which is the best use for the application usage, not just blindly use one or the other.
When RAM is under $10/GB, I'd much rather spend a couple of gigabytes on the OS platform to get apps that actually get released in a timely manner, include more features and plain do cool stuff that I want to do than bitch that they use more ram than my router built in 1992.
Now you're supposing that adding RAM is (i) a possible, and (ii) feasible. Others have already mentioned cases where the systems (even new systems) are already maxed out. RAM is cheap enough now that I tend to max out the RAM when I get a new computer as well. So no, you cannot necessarily just add more RAM.
And when you start writing data intensive applications, more RAM won't help you as you need every bit of it you can to start with. A database not written to be mindful of the resources it uses will not be very useful to other applications except with very small data sets.
I presently work on a near real-time measurement system. Brand new processor boards (even with Core2 Duos) have a max of 1 or 2 GB or RAM. My software has to be resource friendly. If we start swapping to disk (e.g. SWAP file or SWAP partition), it will cause the system to lose data which is unacceptable to our customers. And yes, I went from near zero code to having a deliverable system in less than 9 months, though some of the interfaces were crude it did its job very well.
I've written file servers in the past, and it's no different. If you're not mindful of resources then things can quickly get out of hand. We didn't expect the customer to transfer 1000 files in a day; but we tested for 10,000; the server was expected to be up for months and early testing determined that some lack of mindfulness of resources would crash the server side at just a few hundred at first. As we fixed things to be more mindful we eventually passed the 10,0000 test without any issues - you couldn't tell the software had 10,000 files transferred as we had a zero-sum of resources used after each transfer. We were a little more lenient on the client-side, but it was just as good in the end.
Point is, you have to look at what the software is doing, and making sure it works correctly for its use case and that it uses the resources properly and efficiently for those use cases.
By programmer efficiency, I meant programmer hrs vs. real world problems solved. As I mentioned earlier, it is a trade-off. If you want an OS written in 2/3 of the resources for the same actual problem set, you will be waiting 2-3x as long for it or more. Or it will get "too hard" and never come out. Ever.
And that's essentially the same thing as what I was calling it. it's something that only comes with experience, and cannot be reduced linearly by adding more programmers. And as I noted, programmers that have better programmer efficiency also have the ability to make software that is more efficient and determine where to make the tradeoffs.
If you want to run wm2 or fluxbox or whatever, go for it. Other people have real world tasks they want to do and are prepared to spend say, 50 bucks on RAM and not have to fuck around wasting their time to do it.
Take the emulator Bochs for example. Everyone said 'it is slow because it is in software'. Couple guys came along and got it within 5% of vmware speed all in software and not using dynamic recompile. The guy doing it has lost interest, but he was fairly confident he could be faster than vmware eventually.
I remember when the speed-up for Bochs came, and it was primarily because the QEMU guys made a kernel module that allowed them to use the hardware to run some of the instructions instead of having to virtualize every single instruction. Now everyone has access to the "VM" instructions on Intel/AMD processors, so you can do that without the QEMU kernel module; but then you are limited to having a host CPU be the same CPU as (or at least a compatible variant of) the virtualized one.
And, FYI, Bochs is one of the few virtual machines that can emulate a full X86 and X86-64 computer with ANY computer architecture as its host because they emulate the entire computer.
Now, if they've gotten some additional stuff since that one, I don't know.
And I'm here to tell you that sloppy programming costs billions of real-live dollars in hardware annually.
And everyone cares about their own pocketbook.
Put that in your rapid-development pipe and smoke it.
Slashdot Mirror
Git won the DVCS war
True, and SVN won the Centralized VCS wars.
I've used both and recommend them that way - if you want DVCS then use Git, if you want centralized then use SVN. Both are the top in their respective VCS markets. The only issue is if you use Windows - then Git just sucks as it does not integrate well at all, and as others have pointed out TortoiseGit doesn't really resolve that - it's a tooling problem and a problem in the design of Git. (One that could be probably be resolved by moving away from using so many Bash scripts and using a library for all the functionality like SVN does, which is why TSVN is just as first-class a citizen as the SVN command-line tools.)
And while I realize the they are very different - Git just drives me nuts, between things like having to stage files before a commit to it not tracking directory structures until there is something in it. (I like to setup the directory structure prior to putting things in it; often times I know I'll have something there but it may take a while to get to the point where I'll have something to put in it, yet I want the structure in the VCS before I get there.)
git --bare init
Or even better...
.
svnadmin create
And that works on any platform - Windows or Linux/Unix/Mac; where git just doesn't do Windows very well (outside of Cygwin; the msys version is bleh).
Really? Up here in Canadaland, the evidence required to get a warrant(aka R&PG and/or probable cause) require a chain of evidence when presented in court, because the defense must have full disclosure.
I believe it's the same in the US. Anything the Goverment uses to prosecute must be made available to the defense - classified or not.
The point I was making was not to state to create an unaltered copy of the original but get enough data on the variation of the copies to be able to mess up the watermark enough to render it useless. Random pick of formatting/wording in deviating sections from one of the N copies obtained at each case. The result may be that you have variants A, B and C as source and your scrambling causes it to look like variant K, so the buyer of variant K will be blamed until they figure out that they are chasing in the wrong direction.
Or by doing diffs on the text you might find the variations and be able to eliminate them, producing the original unvariated work and thus no watermark at all.
Except that Cogent earns sympathy for taking advantage of the real Tier 1's because the real Tier 1's only exist because of crony capitalistic efforts, and bullshit shenanigans that makes Cogent's shenanigans look like your dad's dick.
Quite small.
Tier 1's primarily exist because they were their first and had to invest in the market before there was really a market there to invest in. All Tier 1's have been around for a very long time (in the computer world). Cogents and others are relatively new to the market, so they simply don't have the infrastructure that the Tier 1's do. Could they build it? Yes; but for the moment they don't want to incur those very costly expenses.
And their unions dump a pile of money to the Dems as well.
Look up how much money unions donate compared to big corps. (face-palm.)
Yeah, it's not really that different; and they've been known to sway elections a lot more (though that's been waning in recent elections).
As soon as someone takes on a monopoly (say telecoms, phama, oil, coal), those industries dump a pile of money on the GOP who go around talking about big government getting in the road of small businesses. The GOP faithful like this because it makes them feel like they have political power. It is all rather ironic.
And their unions dump a pile of money to the Dems as well.
Please, don't act like the GOP is the only party at fault. The Dems share just as much of the blame.
The other problem is that we get compared as the entire USA vs say Sweden. It's not really a fair comparison given geographic population distributions.
Not just that, but land-mass differences, etc.
Sweden is probably more comparable to say New Hampshire in some respects, California in others, and perhaps Oregon in others. Yet it's compared to the entirety of the US.
A more fair comparison would be US vs. Russia vs Europe vs Africa vs Mexico vs South America.
I've tried, when my wife and I first got married I was earning minimum wage back then I think it was around $5 or so an hour. We could barely feed ourselves, we got pregnant and I applied for financial aid, guess what supporting 2 people at the time with a 3rd on the way I was told I made too much money. Our grocery budget for a month $15.
I agree, the restrictions are insane, and actually encourage couples to not get married - you'd have gotten more support if the two of had divorced! (And they wonder why younger generations see no reason to get married.)
Kinda funny how all of the free-market lovers refuse to break up these telecom monopolies, or at the very least regulate them into being dumb pipes.
I love the free market; yet I also very much agree that the telecom industry needs a massive dose of breakup and regulation. For starters, Cable companies (and Fibre Optic services like FIOS and U-Verse) need to be forced to share their infrastructure.
Personally, I'd move it all to a single owner model - perhaps (even likely) owned by the localities - where companies have to lease them from the owner. So AT&T, Verizon, Comcast, etc, won't have any ability to control the basic pricing or infrastructure - they'd all be forced to compete on the quality of their services instead of their infrastructure.
It's time to realize that the internet Infrastructure really should belong to the people, and not be tied up in a wasteland of monopolies enforced by local governments - e.g. a small housing group should be able to get their own fibre optic line and split it among the group without having the county say "you can't do that because we signed this contract with company X and they have that sole right", that should all be illegal.
Start with Computer Engineering. It's rare that people who start on the finance side ever develop first-class CE skills, while good CE people have very little trouble picking up the necessary finance knowledge. Hell, a number of the high-frequency shops don't even have finance experience as a prerequisite, but you can bet every one of them wants rock-solid technical skills.
FTFY
Your niche is not typical.
And you completely missed the point
There are jobs that people really, really, really want to do for zero pay. Why wouldn't you allow them to make that decision for themselves?
Why don't we allow people to sell themselves into slavery?
Interns aren't allowed to walk away from a company that mistreats them? Modern slave owners are allowed to whip and kill their interns?
I did not know that.
Interns are generally required to complete a certain time period of work to qualify for course credits; they are also bound by employment laws and if terminating the internship early must comply like any other employee in the state where their internship is. For AT-WILL employment they could walk away at any time, but forfeit the course credit; for non-AT-WILL employment, they have to give notice as required by the law.
My internship cost me around $6000 between tuition and commuting. I was reimbursed for mileage spent on the job, but was otherwise unpaid (save for the odd case of Heineken the company president shared when we had to work late). It took up 50 hours a week, which I couldn't use to get a job to pay all of the above plus those little luxuries of life like food and rent. I basically financed everything on a credit card during my internship and a few minor freelance gigs I could land, which really hurt me financially for a couple of years.
The only one letting an intern be taken advantage of like that is the intern themselves. And yes, I didn't take internships that didn't pay - at the very least I had to cover food and expenses (namely gas) over the internship.
In being fair to the judge here, they aren't the first person to think this. Obama himself made similar statements, a manifesto of sorts, about internships needing to benefit only the student and basically be nothing but a burden to the employer.
I would hardly say that a paid internship is harmful to the employer. I never had an unpaid internship, and often made good money for the employer. At one position my "internship" was essentially a senior level position; and in the first month I enabled the company to bill out to customers the budget of they had for my internship. At most, I was doing the level of work of their normal staff, but at a lot cheaper rate than their normal staff. So yes, they saved money by hiring me for the internship over having a normal software position; and while I learned a lot, I also earned the funds I needed for college.
some schools make you pay for the credits so work for free and pay to get credit for it.
Some schools require the companies to pay their interns too.
Well, the study was about browsers and not operating systems. I sort of assumed they tested all browsers on Windows in which case that is a controlled variable.
I'm sure they would sponsor a study looking at power usage of Linux vs. Windows vs. OS X. You probably won't see the results of their work unless it is in their favour, and you probably won't see the results of their work w.r.t. desktop Linux as it doesn't have the mindshare that would make it worthwhile to waste advertising space on (few know what Linux actually is, fewer would consider using it, and fewer still would consider using it, would make a decision based on power usage, and would trust a study funded by Microsoft to base a decision on). I say all this as a Linux/OS X user.
True, the study was about browsers on Windows. However, given that IE is integrated into Windows you need to also have a control variable for Windows too.
Time for a reality check: the incredibly vast majority of people don't give a shit about Other OS. It was a sucky move by Sony to remove it, no denying that, but it affected 0.001% of their user base and perhaps put off half of those if not less. They don't give a damn about a few angered /. posters who swore off Sony because they couldn't run Linux on their game consoles.
They might not at the moment, but that will have a big effect on PS4 sales, at which point they will. The question is, how big the effect will be.
Wouldn't that be a bit extreme? All the game asks you is to perform an update, or a simple check to make sure you don't have a rom loader installed. That isn't a problem to non pirates. I really don't see the problem with this, but even if there is a problem, if that's the only DRM PS4 has, I don't mind the nuisance.
The problem namely being discussed here was Sony's update to the PS3 that removed a feature of the PS3 that people wanted - OtherOS, which allowed running Linux directly in a supported manner. The update completely removed that functionality for everyone regardless of whether it was rooted or not, and that was all done after the PS3 was sold. Of course, IIRC Sony lost in court a class action against them over it, or at least settled in a favorable manner to PS3 owners (e.g. partial refunds or something). The concern is that they'll do something similar with PS4, even if they don't do so right out of the gate.
Let me guess -- I.E. is the only browser which respects that timer setting?
Given that it would be a setting for the Microsoft JavaScript Engine that only IE uses, you're right.
Dell, HP, IBM, and BCE are all windows shops, or at least were when I worked for them, and all of them have *well* over 10,000 employees. I still work for BCE. We've got 65,000 employees in Canada, and they're all on Windows. (well, some of the field repair technicians are being switched to Android, but that's still *well* over 10,000 Windows licenses).
FYI - HP and IBM at least have looked at replacing desktops with Linux. IBM has even gone to the extent of doing it with a sizeable chunk of their work force, probably to show governments it can be done. So just because they're a big employer doesn't mean that everyone there uses Windows, or that those that are not are a statistical blip.
The summary does a disservice by comparing the power usage to that required to make a cup of tea. If you read the article, it says "Laptops use about 14.7 Watts when idling. Firing up a browser adds another one or two Watts to this, depending which sites are browsed, and which browser you are using, Fraunhofer found. On desktop PCs, browsers add the same amount to the energy draw – but the baseline is around 37.8W."
The question remaining, however, is how much of that is hidden by running Windows? In other words, do the same thing with Linux (GNOME, KDE, etc) and Mac OSX. Does it really make a difference? Or is Windows merely hiding the extra usage by always having that usage included in the OS? Of course, don't expect Microsoft to fund such a study.
The increased resource requirements DO in fact do more with your computer. It means apps get developed in man-weeks or man-months rather than years or decades. It means we have a standard set of libraries shared between apps.
Not necessarily. If you use less efficient code, structures, memory usages, etc. then no you will not. You'll just do the same thing with more resources, and you may even spend more time building it.
Shared libraries do not necessarily mean extra code/resources at run-time. You may, for instance, use static linker level libraries that simply allow modules to be shared; or you may make additional DLLs that do have run-time requirements. But you have to determine which is the best use for the application usage, not just blindly use one or the other.
When RAM is under $10/GB, I'd much rather spend a couple of gigabytes on the OS platform to get apps that actually get released in a timely manner, include more features and plain do cool stuff that I want to do than bitch that they use more ram than my router built in 1992.
Now you're supposing that adding RAM is (i) a possible, and (ii) feasible. Others have already mentioned cases where the systems (even new systems) are already maxed out. RAM is cheap enough now that I tend to max out the RAM when I get a new computer as well. So no, you cannot necessarily just add more RAM.
And when you start writing data intensive applications, more RAM won't help you as you need every bit of it you can to start with. A database not written to be mindful of the resources it uses will not be very useful to other applications except with very small data sets.
I presently work on a near real-time measurement system. Brand new processor boards (even with Core2 Duos) have a max of 1 or 2 GB or RAM. My software has to be resource friendly. If we start swapping to disk (e.g. SWAP file or SWAP partition), it will cause the system to lose data which is unacceptable to our customers. And yes, I went from near zero code to having a deliverable system in less than 9 months, though some of the interfaces were crude it did its job very well.
I've written file servers in the past, and it's no different. If you're not mindful of resources then things can quickly get out of hand. We didn't expect the customer to transfer 1000 files in a day; but we tested for 10,000; the server was expected to be up for months and early testing determined that some lack of mindfulness of resources would crash the server side at just a few hundred at first. As we fixed things to be more mindful we eventually passed the 10,0000 test without any issues - you couldn't tell the software had 10,000 files transferred as we had a zero-sum of resources used after each transfer. We were a little more lenient on the client-side, but it was just as good in the end.
Point is, you have to look at what the software is doing, and making sure it works correctly for its use case and that it uses the resources properly and efficiently for those use cases.
By programmer efficiency, I meant programmer hrs vs. real world problems solved. As I mentioned earlier, it is a trade-off. If you want an OS written in 2/3 of the resources for the same actual problem set, you will be waiting 2-3x as long for it or more. Or it will get "too hard" and never come out. Ever.
And that's essentially the same thing as what I was calling it. it's something that only comes with experience, and cannot be reduced linearly by adding more programmers. And as I noted, programmers that have better programmer efficiency also have the ability to make software that is more efficient and determine where to make the tradeoffs.
If you want to run wm2 or fluxbox or whatever, go for it. Other people have real world tasks they want to do and are prepared to spend say, 50 bucks on RAM and not have to fuck around wasting their time to do it.
Take the emulator Bochs for example. Everyone said 'it is slow because it is in software'. Couple guys came along and got it within 5% of vmware speed all in software and not using dynamic recompile. The guy doing it has lost interest, but he was fairly confident he could be faster than vmware eventually.
I remember when the speed-up for Bochs came, and it was primarily because the QEMU guys made a kernel module that allowed them to use the hardware to run some of the instructions instead of having to virtualize every single instruction. Now everyone has access to the "VM" instructions on Intel/AMD processors, so you can do that without the QEMU kernel module; but then you are limited to having a host CPU be the same CPU as (or at least a compatible variant of) the virtualized one.
And, FYI, Bochs is one of the few virtual machines that can emulate a full X86 and X86-64 computer with ANY computer architecture as its host because they emulate the entire computer.
Now, if they've gotten some additional stuff since that one, I don't know.
And I'm here to tell you that sloppy programming costs billions of real-live dollars in hardware annually.
And everyone cares about their own pocketbook.
Put that in your rapid-development pipe and smoke it.
Well said.