Anything above writing your own heap allocator over raw OS calls is "a crutch". It's merely a matter of choosing the right crutch.
Even writing your own allocator could be considered a crutch. Using a the operating system? Pshaw! Real men talk to the metal directly. C is a crutch. So is assembly.
In short, I agree with your point completely. I've written microcode, when I had to. I've designed special purpose hardware (with the help of an EE to keep me from making too many dumb-programmer blunders) when it was needed. But 99.9% of the time I work with the strongest crutches I can find--though I tend more to think of them as power armor.
C programmers who think they're living on the cutting edge of something should look around more.
did you miss the requirement that there *must* be PHP?
And I think you may have even hinted why.. "Although there is a lot of PHP out there".
I didn't miss it, I challenged it. Just because someone says "I'd like to cut wood with a rusty axe--what's the best way to do it?" doesn't mean I'm under some obligation to suspend all judgement in answering. I'm perfectly free to advise him to use a different tool, even when he says the rusty axe is a "requirement."
He is of course free to disregard my suggestion, but it's irresponsible to just blindly accept requirements when asked for your opinion. PHP is rather insecure (by virtue of the fact that security seems to have been added as a feature rather than as a core design phylosophy). Ruby is, IMHO, a much better choice. Thus my post, honestly answering the question rather than slavishly endorsing its underlying assumptions.
I'd strongly recommend Ruby-on-Rails in addition to or even instead of PHP. Although there is a lot of PHP out there, I can't think of a reason to use it for a new project if you knew both PHP and Ruby, or if you knew neither and had to learn one or the other.
And I can think of a few strong reason to use Ruby: tainting, true object orrientation, no overflows,..
It goes further than that sometimes. I see a lot of stuff on/. that's moderated as troll or flamebait that really just boils down to a difference of opinion with the parent post.
It may not be as bad as it seems. Yes, the down modded post might "boil down" to a simple disagreement, but the mod is often based on what you "boiled away" to get at the core.
Flamebait is essentially a disagreement with one "side" of a discussion, but phrased in such a way at to invite/incite overly emotional rebuttal by that "side" and thus foster the conceptualization of the discussion as a fight between two intrinsically opposing sides.
A troll often poses as a simple disagreement, but the key is that it's a pose--the poster may or may not actually disagree, but the post is falsly cast (or cliped from a standard file); it is not an honest response or contribution to the discussion.
So yes, you are right, but I submit that posts should be modded on their flavor as well as their substance; someone making a perfectly valid point in foul and abusive language deserves to be modded down as much as someone who is offtopic (for example), if only to keep the discourse civil.
--MarkusQ
I can see you've never run a small company
on
Layoffs at OSDL
·
· Score: 5, Insightful
If you need to downsize, you get rid of non-critical personnel first. Receptionist are definitely in that category, especially since they are most easily replaced out of any staff members.
I can see you've never run a small company. If you had, you'd realize that a good (or bad) receptionist can make (or break) your business. Think about it--here is one person who typically talks to every employee several times a day, and most of your customers every week or two. The person who watchs who and what enter and leave, gets to see the unguarded moments, the body language, hear the idle gossip--in short, the best clue catcher you'll ever have.
I'm always amazed at the money people will pay consultants for clues they could have gotten in far less time just by asking the recptionist. Often, the receptionist is the only person in the whole outfit that sees the big picture.
I mean, come on! Anyone who's ever dealt with it knows that kudzu stops everything.
Why not say "kudzu stops house panting" or "kudzu stops lawn mowing" or "kudzu stops grocery shopping" or "kudzu stops carjackings" or...well, the point is, unless you fight back with a nuke-it-from-orbit mentality, kudzu stops everything.
Why isn't there a FEDERAL agency overseeing this issue? Where are the FEDERAL agents to review the code and hardware?
Vote for me! My first act will be to establish a FEDERAL department for electronic voting with completely Open Source software and open standards for hardware.
I'll make it so easy for every person in the US to vote me out of office.
I'll put together a system so that 4 years later, you can see me lose the election, state by state, hour by hour.
With no question that every vote is counted correctly.
You'd have my vote. But then, I was always a sucker for statesmen. I'd rather have an honest man I disagreed with than a smarmy bastard that said everything I wanted to hear.
The question is, at what point will people be so fed up with the politicians that honesty will become the best policy again? In other words, am I alone in my disgust? I think not. I was at a small party last week where it came out that none of the Republicans present had voted for Bush, and none of the Democrats had voted for Kerry. The main reason? Nobody really liked or trusted either one of them, and nobody was all that confident that the system by which one of them got chosen wasn't rigged.
But they weren't all that concerned by the possibility that it was rigged because they didn't like either major party candidate.
So here's an idea that just might work (for either party). Run somebody honest and forthright, somebody who will speak his or her mind plainly and clearly regardless of what the day's audience wants to hear. And then see if people don't start caring about the honesty of the system.
Who knows, somebody that followed this course might get re-elected because people liked them, instead of because they were less dispicable than their opponent.
--MarkusQ
P.S. We could even dust off that old concept of "leadership"...
As I'm sure many in the IT world could attest to, anytime the average person is presented with a procedure that is slightly more complex, the error rate jumps. Thus, you get a net increase in your error.
The problem is, we aren't as worried about error as we are about fraud or other abuse of the system. Passwords (or any system of authentication) make the process slightly more complex and thus (by your reasoning) increase the error rate. But they are worth it because they reduce the chance of unathorized persons abusing the system.
Would you use an ATM that didn't require a PIN (or, for that matter, a card)? Suppose you could just walk up to it, enter your name, and withdraw money from your account. Simple, and much less chance of error, right?
What amazes me is that people realize that a bad guy would go to considerable length to be able to filch a few hundred dollars out of their bank accounts, or even to wrest control of some third world bannana republic, but very few people accept that somebody might want to exercise undue influence over the richest and most powerful nation in the world. I guess you'd have to be some sort of conspiracy theory nut job to think that control of the US would be a prize worth cheating a little bit for.
Rather, the voting method seems to affect the relative turnout of different voter demographies.
I'm not sure what this means--how in the heck could the voting method affect the relative turnout?
Do people really think this way?
Hmmm, let's see...I have the ability to help select the next leader of the free world--literally to take part in a descision that could mean life or death for tens of thousands of people.
But...I'd have to use a #2 pencil.
*sigh*
I guess I'll skip it this time.
A much more likely explanation: some voting methods allow for more people to vote (faster throughput) and--if they are disproportionately deployed with respect to precinct demographics--you see what looks like a bias in voter turnout but really is election-rigging via selective resource allocation.
The people didn't stay home based on the voting method--they turned out, but they saw the long lines (which were caused by the voting method) and decided they they didn't have time to vote. In several races (e.g. WA Gov, OH Pres), well crafted manipulation of this effect would have been enough to turn the race.
That's the most sensible interpretation. In the world of conspiracy blogs, the line "[the researchers] found no evidence for systematic fraud by testing several potential indicators" will be interpreted by many as either "the researchers were paid off" or "the cover-up of theft is so complete and devious, even the researchers didn't see anything."
You missed the most obvious: some researchers found problem, others didn't, and only the ones that didn't find problems get air time. This is a time honoured tradition (cf. everything from tabaco studies in the 1950's to WMD in the past few years) for spreading disinformation.
He still doesn't seem to understand what he did wrong.
He doesn't understand what he did wrong when he published the article, and I'll bet dollars to doughnuts he doesn't yet realize what he did wrong in the interview.
I can't imagine anyone with even a shred of a clue, when giving an interview that is almost certain to be linked to by slashdot, giving blanket permision like he did:
Besides, talking about personal home numbers, you can find my home number listed in the white pages and my home address is listed there as well. If a reporter wants to call me up at home or in my office, they can look up my phone number and address and show up at my door and ring my bell. I will come out and take a picture with the reporter if he or she wants to take my picture. Any reporter is welcome to my home as well as my office. We are not anonymous or private when it relates to our professional lives.
*Wince* What was he thinking?
--MarkusQ
P.S. For the record, if I ever get interviewed and slashdot gets hold of it, I want all of you yahoos to stay the heck away from my house. My number is not listed and I will not pose for photos with you.
this reminds me of an essay I read by a second year physics student that nanotechnology would allow us to run 10GHz computers for 10 years off a watch battery. it's BS but you don't need to look at the technology to see that, it's just basic thermodynamics:
law 1. you can't win
law 2. you can't break even.
law 3. you can't get out of the game.
You need to reread the paper, or rethink your objection to it.
The point isn't that you can run a traditional 10GHz processor off a watch battery. The point is that there are systems permitted by the laws of physics (although we have no idea how to manufacture them at this point), which are capable of doing the same computations as a present-day 10GHz processor but produce far less heat in the process.
Do you know what 99.99...% of the power of a modern processor winds up as? Waste heat! Just as a large fraction of what you put into a horse is spent on things other than moving your cart forward, a the vast majority of the power used by a processor is wasted. Just as it is quite reasonable to try to design something that is more efficent than a horse at moving a load forward, it makes perfect sense to try to design things that compute more efficiently. And since we only care about the abstract result, we are better off making it smaller.
Before you lable something "BS" it's best to make sure you understand it.
As a counter example, how about the four color theorem?
Case in point--the idea can just as well be explained by assigning numbers or any other sort of label to the nodes, and was in fact finaly proved via a computer program which (one would hope) could be understood without reliance on a syntax-highlighting editor.
Or, for that matter, dancing hamsters.
I have on my shelf two books that discribe the proof, one in the abstract and the other in gory detail. Both are simple black-and-white, non-flash math books.
Yes! Exactly!
I just droped my last foe to add you to my friends list.
I can't recall who said it first, but I agree with the adage "any idea that can't be convincingly explained without using several different colours has a serious flaw in it somewhere."
The only thing I could find that even sort of suport your claim was
RMS's thing about not wanting the backend to drift into becoming an LGPL black-box (thus chilling the development of new GPL'd front ends).
I can live with changes that shorten the time spans for the copyright (in fact, I'm in favour of something on the order of ten years), but I do not support any that would (even marginally) shorten the time spans for my life.
I wasn't talking about you per se (though I can see how it might have been read that way); posting a google link is mostly harmless. But the people claiming this is "established"? Or a reason to abondon real research? C'mon.
Next they modify the data you receive to influence your opinion.
Oh yeah? I just Googled for this very topic and there is absolutly no proof of that sort of thing. Ever. To Anyone. You'd think that if it were true somebody would have blogged about it. So you must just be parinoid.
You'll get your wish. It isn't how much money that either side has that will matter in the long run; it's how fast they spend it. SCO/Microsoft/et. al. are not at equilibrium with their environment; they have to pay layers, PR firms, programers, astroturfers, accountants, etc. to work for them, or else they just *pop* go away.
The F/OSS movement is more like Cubs fans. The Cubs don't pay people to suport them. It's the same with Free/Open Source Software suporters. In the vast majority of cases, no one pays them to do it--in fact, in most cases it costs them (in time, if nothing else) to be "a F/OSS advocate." So their burn rate is...zero.
The more the fight heats up, the worse it gets. More people pitch in on the F/OSS side, while on the other side expensses rise and revenues fall.
In the long run, who's going to run out of money first?
As the parent poster pointed out, there is no perfect defense. But we all know that the best defense is a good offense--you need to beat these creeps at their own game.
Do you remember the story about the woman who had her purse snatched twice in the same train station, so she started carrying about a purse filled with dog poop? It only got stolen one more time. The thief learned quickly! And I say, that's the way to do it!
Strke first! Strike hard! Trash your own credit to the point that anybody that claims to be you is pretty much guarenteed to have all their stuff taken from them within a week. The only way to stop identity theft is to make it so painful for the thief that they can't take it any more and cry uncle (note: if you think about it this also explains what's been going on in the movie/music industries lately).
Slashdot Mirror
Anything above writing your own heap allocator over raw OS calls is "a crutch". It's merely a matter of choosing the right crutch.
Even writing your own allocator could be considered a crutch. Using a the operating system? Pshaw! Real men talk to the metal directly. C is a crutch. So is assembly.
In short, I agree with your point completely. I've written microcode, when I had to. I've designed special purpose hardware (with the help of an EE to keep me from making too many dumb-programmer blunders) when it was needed. But 99.9% of the time I work with the strongest crutches I can find--though I tend more to think of them as power armor.
C programmers who think they're living on the cutting edge of something should look around more.
--MarkusQ
I didn't miss it, I challenged it. Just because someone says "I'd like to cut wood with a rusty axe--what's the best way to do it?" doesn't mean I'm under some obligation to suspend all judgement in answering. I'm perfectly free to advise him to use a different tool, even when he says the rusty axe is a "requirement."
He is of course free to disregard my suggestion, but it's irresponsible to just blindly accept requirements when asked for your opinion. PHP is rather insecure (by virtue of the fact that security seems to have been added as a feature rather than as a core design phylosophy). Ruby is, IMHO, a much better choice. Thus my post, honestly answering the question rather than slavishly endorsing its underlying assumptions.
--MarkusQ
And in the next release, they'll replace it with "Our".
--MarkusQ
I'd strongly recommend Ruby-on-Rails in addition to or even instead of PHP. Although there is a lot of PHP out there, I can't think of a reason to use it for a new project if you knew both PHP and Ruby, or if you knew neither and had to learn one or the other.
And I can think of a few strong reason to use Ruby: tainting, true object orrientation, no overflows,..
--MarkusQ
It may not be as bad as it seems. Yes, the down modded post might "boil down" to a simple disagreement, but the mod is often based on what you "boiled away" to get at the core.
Flamebait is essentially a disagreement with one "side" of a discussion, but phrased in such a way at to invite/incite overly emotional rebuttal by that "side" and thus foster the conceptualization of the discussion as a fight between two intrinsically opposing sides.
A troll often poses as a simple disagreement, but the key is that it's a pose--the poster may or may not actually disagree, but the post is falsly cast (or cliped from a standard file); it is not an honest response or contribution to the discussion.
So yes, you are right, but I submit that posts should be modded on their flavor as well as their substance; someone making a perfectly valid point in foul and abusive language deserves to be modded down as much as someone who is offtopic (for example), if only to keep the discourse civil.
--MarkusQ
I can see you've never run a small company. If you had, you'd realize that a good (or bad) receptionist can make (or break) your business. Think about it--here is one person who typically talks to every employee several times a day, and most of your customers every week or two. The person who watchs who and what enter and leave, gets to see the unguarded moments, the body language, hear the idle gossip--in short, the best clue catcher you'll ever have.
I'm always amazed at the money people will pay consultants for clues they could have gotten in far less time just by asking the recptionist. Often, the receptionist is the only person in the whole outfit that sees the big picture.
--MarkusQ
I mean, come on! Anyone who's ever dealt with it knows that kudzu stops everything.
Why not say "kudzu stops house panting" or "kudzu stops lawn mowing" or "kudzu stops grocery shopping" or "kudzu stops carjackings" or...well, the point is, unless you fight back with a nuke-it-from-orbit mentality, kudzu stops everything.
What? They ate it? Eeew.
--MarkusQ
You'd have my vote. But then, I was always a sucker for statesmen. I'd rather have an honest man I disagreed with than a smarmy bastard that said everything I wanted to hear.
The question is, at what point will people be so fed up with the politicians that honesty will become the best policy again? In other words, am I alone in my disgust? I think not. I was at a small party last week where it came out that none of the Republicans present had voted for Bush, and none of the Democrats had voted for Kerry. The main reason? Nobody really liked or trusted either one of them, and nobody was all that confident that the system by which one of them got chosen wasn't rigged.
But they weren't all that concerned by the possibility that it was rigged because they didn't like either major party candidate.
So here's an idea that just might work (for either party). Run somebody honest and forthright, somebody who will speak his or her mind plainly and clearly regardless of what the day's audience wants to hear. And then see if people don't start caring about the honesty of the system.
Who knows, somebody that followed this course might get re-elected because people liked them, instead of because they were less dispicable than their opponent.
--MarkusQ
P.S. We could even dust off that old concept of "leadership"...
The problem is, we aren't as worried about error as we are about fraud or other abuse of the system. Passwords (or any system of authentication) make the process slightly more complex and thus (by your reasoning) increase the error rate. But they are worth it because they reduce the chance of unathorized persons abusing the system.
Would you use an ATM that didn't require a PIN (or, for that matter, a card)? Suppose you could just walk up to it, enter your name, and withdraw money from your account. Simple, and much less chance of error, right?
What amazes me is that people realize that a bad guy would go to considerable length to be able to filch a few hundred dollars out of their bank accounts, or even to wrest control of some third world bannana republic, but very few people accept that somebody might want to exercise undue influence over the richest and most powerful nation in the world. I guess you'd have to be some sort of conspiracy theory nut job to think that control of the US would be a prize worth cheating a little bit for.
--MarkusQ
I'm not sure what this means--how in the heck could the voting method affect the relative turnout?
Do people really think this way?
A much more likely explanation: some voting methods allow for more people to vote (faster throughput) and--if they are disproportionately deployed with respect to precinct demographics--you see what looks like a bias in voter turnout but really is election-rigging via selective resource allocation.The people didn't stay home based on the voting method--they turned out, but they saw the long lines (which were caused by the voting method) and decided they they didn't have time to vote. In several races (e.g. WA Gov, OH Pres), well crafted manipulation of this effect would have been enough to turn the race.
--MarkusQ
You missed the most obvious: some researchers found problem, others didn't, and only the ones that didn't find problems get air time. This is a time honoured tradition (cf. everything from tabaco studies in the 1950's to WMD in the past few years) for spreading disinformation.
--MarkusQ
He doesn't understand what he did wrong when he published the article, and I'll bet dollars to doughnuts he doesn't yet realize what he did wrong in the interview.
I can't imagine anyone with even a shred of a clue, when giving an interview that is almost certain to be linked to by slashdot, giving blanket permision like he did:
*Wince* What was he thinking?--MarkusQ
P.S. For the record, if I ever get interviewed and slashdot gets hold of it, I want all of you yahoos to stay the heck away from my house. My number is not listed and I will not pose for photos with you.
You need to reread the paper, or rethink your objection to it. The point isn't that you can run a traditional 10GHz processor off a watch battery. The point is that there are systems permitted by the laws of physics (although we have no idea how to manufacture them at this point), which are capable of doing the same computations as a present-day 10GHz processor but produce far less heat in the process.
Do you know what 99.99...% of the power of a modern processor winds up as? Waste heat! Just as a large fraction of what you put into a horse is spent on things other than moving your cart forward, a the vast majority of the power used by a processor is wasted. Just as it is quite reasonable to try to design something that is more efficent than a horse at moving a load forward, it makes perfect sense to try to design things that compute more efficiently. And since we only care about the abstract result, we are better off making it smaller.
Before you lable something "BS" it's best to make sure you understand it.
--MarkusQ
Case in point--the idea can just as well be explained by assigning numbers or any other sort of label to the nodes, and was in fact finaly proved via a computer program which (one would hope) could be understood without reliance on a syntax-highlighting editor.
Or, for that matter, dancing hamsters.
I have on my shelf two books that discribe the proof, one in the abstract and the other in gory detail. Both are simple black-and-white, non-flash math books.
--MarkusQ
Yes! Exactly! I just droped my last foe to add you to my friends list.
I can't recall who said it first, but I agree with the adage "any idea that can't be convincingly explained without using several different colours has a serious flaw in it somewhere."
--MarkusQ
My first thought was, gee, that's not been my experience--as I recall, although it was complicated, it was rather well documented. So I did a bit of Google and found that other people seemed to agree with me (i.e., they say things like "Furthermore, compared to the other compiler projects, GCC offered the most comprehensive documentation for backend porters." and so forth).
The only thing I could find that even sort of suport your claim was RMS's thing about not wanting the backend to drift into becoming an LGPL black-box (thus chilling the development of new GPL'd front ends).
So, do you care to back your claim up?
--MarkusQ
No one's going to kill authors to shorten terms.
Great. So we have to live with the knowledge that our work isn't worth killing for.
I tell you, it's a lose-lose proposition.
--MarkusQ
I can live with changes that shorten the time spans for the copyright (in fact, I'm in favour of something on the order of ten years), but I do not support any that would (even marginally) shorten the time spans for my life.
--MarkusQ
I wasn't talking about you per se (though I can see how it might have been read that way); posting a google link is mostly harmless. But the people claiming this is "established"? Or a reason to abondon real research? C'mon.
--MarkusQ
Great links. Basically various groups of people trying to sell something and various other groups trying to shut them down.
Not exactly "research."
--MarkusQ
Yes, but the 5th amendment is against the patriot act.
Which one do you think would win?
--MarkusQ
Oh yeah? I just Googled for this very topic and there is absolutly no proof of that sort of thing. Ever. To Anyone. You'd think that if it were true somebody would have blogged about it. So you must just be parinoid.
--MarkusQ
You'll get your wish. It isn't how much money that either side has that will matter in the long run; it's how fast they spend it. SCO/Microsoft/et. al. are not at equilibrium with their environment; they have to pay layers, PR firms, programers, astroturfers, accountants, etc. to work for them, or else they just *pop* go away.
The F/OSS movement is more like Cubs fans. The Cubs don't pay people to suport them. It's the same with Free/Open Source Software suporters. In the vast majority of cases, no one pays them to do it--in fact, in most cases it costs them (in time, if nothing else) to be "a F/OSS advocate." So their burn rate is...zero.
The more the fight heats up, the worse it gets. More people pitch in on the F/OSS side, while on the other side expensses rise and revenues fall. In the long run, who's going to run out of money first?
--MarkusQ
As the parent poster pointed out, there is no perfect defense. But we all know that the best defense is a good offense--you need to beat these creeps at their own game.
Do you remember the story about the woman who had her purse snatched twice in the same train station, so she started carrying about a purse filled with dog poop? It only got stolen one more time. The thief learned quickly! And I say, that's the way to do it!
Strke first! Strike hard! Trash your own credit to the point that anybody that claims to be you is pretty much guarenteed to have all their stuff taken from them within a week. The only way to stop identity theft is to make it so painful for the thief that they can't take it any more and cry uncle (note: if you think about it this also explains what's been going on in the movie/music industries lately).
--MarkusQ
I'm working on a Ruby extension on work hours too;
Nice! I can already think of several interesting uses for something like that!
--MarkusQ