Re:Does Perfume give you trouble?
on
Safe Cigarettes?
·
· Score: 1
I am a smoker, and I don't have allergies. One thing that makes my eyes smart though, is some perfumes, in certain concentrations, on certain people. Are you affected by perfume as well as cigarette smoke?
Yeah, I am. It's pretty rare for me to come across someone who is wearing enough perfume to cause a problem, but it does happen. Usually just a little congestion and eyes watering.
I'm not in favor of outlawing perfume in public, though. Most things like that are solved by basic courtesy (don't put on the whole bottle!) But there really is no amount of smoking that won't cause people like me immediate problems. Besides, even if there were no emperical evidence of direct harm from second-hand cigarette smoke, it still contains known toxins and carcinogens. Common sense dictates that you shouldn't blow those on unwilling people, especially given the way cigarette smoke lingers.
Re:Cause or Risk Factor? (warning pro-smoking)
on
Safe Cigarettes?
·
· Score: 1
You know I'm allergic to cats (and some dogs), but you don't see me complaining that there needs to be a ban on walking your dog, or letting your cat loose now do you? Nope, because I value personal freedoms a little more than the slight inconvenience of watery eyes and irritated sinuses.
I also have those allergies, and I don't advocate that either. However, if someone (in whatever manner) deliberately put concentrated pet dander into the air all around you whenever you went into public, it would cause you a lot more than watery eyes and irritated sinuses. You would probably call that assault. That's a much more analogous situation.
I, too, value personal freedom. It's just that I value my freedom to go places without being assaulted more than your freedom to assault me.
Ok, as I've already stated, I'm not claiming that his theories are all true. I'm simply claiming that third party labs have confirmed the excessive heat release
Replication of experiments claiming to demonstrate excess heat production in light water-Ni-K2CO3 electrolytic cells was found to produce an apparent excess heat of 11 W maximum, for 60 W electrical power into the cell. The 28 liter cell used in these verification tests was on loan from a private corporation whose own tests with similar cells are documented to produce 50 W steady excess heat for a continuous period exceeding hundreds of days. [...] the present data do admit efficient recombination of dissolved hydrogen-oxygen as an ordinary explanation.
The reproduction wasn't nearly as dramatic as the original claim, and there's a potential ordinary explanation. He's making an extraordinary claim, and I want to see extraordinary evidence: eliminate the ordinary explanation and make it produce as much energy as he says for as long as he says.
Re:Cause or Risk Factor? (warning pro-smoking)
on
Safe Cigarettes?
·
· Score: 2, Insightful
Clearly there are more smokers than there are people who are this sensitive to smoke. In your situation, I'd say you are the one who needs to be more understanding, not the smokers.
That is not clear at all. You hear more from the smokers because they're obnoxious assholes. When strangers smoke near me, I generally give them an unpleasant look and move away. They never know or care how much of a problem they're causing me. Do you smoke? Do you get those looks? Next time, ask why.
It is common to some have degree of problems caused by second-hand smoke. The grandparent's watery eyes and headache are similar to my symptoms, just not as severe. People don't speak up because smoking is accepted. That's changing in most areas.
If you want to see scientific studies that link second-hand smoke to short-term medical problems, just google "second-hand smoke asthma". First hit. I've also heard that smoking in the home is either the #1 or #2 risk factor for asthma in children. You can debate that, of course, since correlation is not causation. (Smoking is more common among the poor and in the South.) But there is evidence, and more is coming. For better or for worse, there will come a day when you can have your children taken away for smoking around them.
If I start having seizures whenever I see the color red, am I justified in asking everyone to stop wearing red for me and the other 3 people in the world with my affliction?
No, because you just made that up. I have real physiological problems that are much more common than you think. Do you really think you can blow known carcinogens and toxic chemicals into the air around you and not cause people problems?
A better analogy would be peanut butter allergies. Very similar, fairly rare. There are people around who will die if they eat peanut butter. They demand accurate ingredient lists so that they don't eat peanuts involuntarily. I demand that I don't get second-hand smoke forced on me. They aren't suggesting outlawing the sale or consumption of peanuts. I'm not suggesting banning smoking in your own home. If you want to kill yourself, that's fine. But don't involve me.
Re:Cause or Risk Factor? (warning pro-smoking)
on
Safe Cigarettes?
·
· Score: 4, Interesting
I couldn't care less if it's totally harmless to me in the long term - in the short term it gives me what feels like an allergic reaction (stuffiness, watering eyes, lethargy) which isn't very pleasant. [...] But in enclosed public spaces, please refrain from it - those of us who don't smoke find it at best smelly, at worst, feeling a bit ill.
Some of us more than a bit ill. Second-hand smoke causes me real short-term problems. I've heard people say that cigarette smoke isn't an allergen; it's chemical sensitivity. Whatever you call it, the effect is the similar to - but more potent than - my other allergies. My eyes burn and water, I cough and sneeze, and my sinuses fill, which often leads to a sinus infection, which means a week of misery and $100 of antibiotics. Probably a day missed from work. I used to live in Iowa, so there were a number of places I just couldn't go - bars, bowling alleys, etc. I had sinus surgery to to prevent the infections, with disappointing results.
I've moved to California, where they have seen reason. I can go anywhere. I'm incredibly sensitive to smoke, so I do have occasional problems still. I'll smell smoke and see a coworker who walked past and is now 100 feet away; the smell lingers. If I smell it, I know it will cause me problems. If I don't smell it, it still might. (I once had a pretty bad reaction to a hotel room that smelled just fine to me. Never found out why.) I deliberately shun people who smell like smoke. If I am forced to be near them, I take a shower as soon as I get home.
Interesting. I'm at work now, so I shouldn't take the time to actually solve it until I get home. But it sounds like a problem you could solve by:
Statistics. It seems like if you are One with the Statistics, you could do it with pencil, paper, and a calculator just powerful enough to do factorials. (Unfortunately, I am not.)
Dynamic programming or recursion with memoization. (There are overlapping subproblems.)
Recursion. This would be slow, and I bet it's the way most people use.
The author is spot on in quite a few respects - engineering is more a test of endurance than intelligence.
That's definitely why I switched out of engineering (into CS and physics). Lots of meaningless homework. Cheating was rampant. It seemed that there were three classes of students:
The majority (say, 90%), collaborated on the assignments much more than the academic standards allowed. Lots of blatant copying. I'm not sure if the professors deliberately turned a blind eye or if they - somehow - didn't know. But the TAs definitely knew. I had more than one who rather blatantly told me to join this group.
Some (say, 5%) did all of the homework themselves, lived in the professors' offices, etc. Obviously dedicated students, but I'm not sure they necessarily got more understanding of the material from this approach. The homework was just a lot of work, not a lot of learning.
The other 5% or so did very little of the homework but were honest about it. I was in this group, and my grades reflected it. My test scores were well above average, but that's not enough.
In my physics classes[*], almost everyone was in the second category. Except that doing the work really did pay off. The homework was different: fewer, more difficult problems. You'd actually be missing something if you skipped half the work.
There were a lot of other differences that made the physics program more satisfying. It was a smaller college, so you got a lot more attention from the instructors. A tighter community. More pleasant all around.
Shortly before I started engineering, a crazed physics TA went on a shooting rampage through my campus, killing seven people before he turned the gun on himself.
I'm relieved to see from your resumé that you're a fellow Iowa alumnus. For a second I wondered if there were more stories like this out there...
[*] Physics for majors, that is. I took the first physics for engineer class, and it was a totally different experience. Lecture with 300 people. Multiple choice tests (five choices), mean score of 3/12.
> Sure it does: [...] fornications [...] fornication [...] fornication
Try reading sometime. Here's the part of the grandparent post that you missed:
The word for perversion was translated as 'fornication' which was then defined as 'premarital sex'. Some verses of the Bible are utterly absurd with this interpretation. Jesus says you can't divorce a woman except for fornication. Well... married women do 'Adultery', not fornication. (I.e., extra-marital sex, not pre-marital). The Bible has a word for Adultery, believe me. When you subsitute perversion instead, the verse actually makes sense.
"Fornication" is an English word, never found in the original text. This may shock you, but language can be ambiguous. Thus, translations can be wrong. This is why Muslims consider only Arabic versions of the Qur'an to be correct. Sometimes I wish Christians did the same.
I wouldn't take any part of this article seriously. I quote:
According to FBI headquarters, the war against smut is 'one of the top priorities' of Attorney General Gonazalez and FBI Director Robert Meuller. Although law enforcement agencies have always been aggressive when it comes to prosecuting exploitative child pornographers, this new initiative is unique in that it targets Internet pornography featuring consenting adults
I thought this was just the submitter's mistake, but it is actually that way in the article. I shouldn't be surprised; Ars Technica should stick to their stupid overclocking articles.
First off- developers are far less likely to comment in wierd formats, or update comments in wierd formats
I disagree. The farther away the documentation is from the source code, the less likely it is to get written or updated. The best-maintained documentation is inline, like Javadocs, Python docstrings, pod, or doxygen comments. When you make a change to the source, the documentation you're invalidating is right there. It takes almost deliberate blindness to not keep it in sync.
Second best is a wiki. It's not right next to the code, but it's so easy to change. Also great if you have documentation that you share with other teams. (QA, other development groups, etc.) Wikis let them easily change things without digging through your source code.
(even when they do, its easy to typo and break the generator).
Then make the doc processing part of the normal build process. Especially if you're an automatic builder like Cruise Control. Just make that one of the steps. If it fails, Cruise Control sends out an email to everyone with the error message and the changes since the last successful build.
Secondly, for good documentation you need more than what an autogenerator can do anyway- good documentatition has examples, use cases, warnings and gotchas, etc.
There's no such thing as a documentation autogenerator. Just tools that reformat, add links and context, etc. You still have to write the words, and you still get to choose what they are. Having examples, use cases, warnings and gotchas is orthogonal to the software you use.
but usually the data isn't important enough (like a slashdot post) to really care if a couple words get chopped off by accident
That is exactly the MySQL attitude. The problem is that my data tend to be important, or I wouldn't be putting them in a RDBMS. I'd rather the database assume the usual, safe thing - data should not silently truncated. If I want to throw away my data, I'll do it explicitly.
And filling in NOT NULLS is also handy for the same reason.... if you REALLY need the logic to reject things like blank inputs in web forms then you should be doing that in your application logic anyway....
What I REALLY need is for the database to pay attention to the constraints I specify. If I didn't care if these fields were blank, I wouldn't have said "NOT NULL".
I am by no means sticking up for non-conformity... just trying to relay the idea that in the context that mysql is usually used, these small quirks don't have a large impact.
That is not the context in which mysql.com is claiming their database is usually used. They claim they have a real database.
Ahh, but you see, no MySQL partisan came into this forum and posted that garbage.
I fail to understand the logic behind avoiding solid products because someone who likes it annoyed you on slashdot. If I made decisions like that, I wouldn't use:
PostgreSQL
MySQL
Oracle
Microsoft SQL
Linux
Windows
*BSD
OS X
Perl
Python
Ruby
C++
Java
...
In fact, I can't think of a single significant piece of software I would use. I'd have to get a new job.
You stop coming up with whole new APIs and practically new OSes just so you can have your little relational database world, and you work with the kernel to get the minimum basic support needed into the filesystem for ACID.
I don't think you understand how major an undertaking you're proposing.
Getting ACID right is hard. Let's say I'm a RDBMS developer. I have two choices:
Write the code in userspace in a way that can be deployed on existing systems easily. It's a lot of code. It'll crash a lot while I'm developing it. Security holes might be discovered in it. That's okay. It's a user process in its own little sandbox.
Work with kernel developers for every major OS and standardization committees to push through an entirely new API for ACID. Develop new filesystems that support it. (Yes, the actual storage would have to change.) Degrade the performance of the rest of the system, which doesn't need these constraints. Watch the whole system go down repeatedly because so much code is going into kernelspace. Have more remote root holes than Windows. Watch no one actually install it because they don't want to reformat their system for my database.
...and I'd still need to develop a database on top of it. The other thing you asked for, SQL, is not trivial either.
Read Codd's 12 rules to truly understand what a RDBMS does for you. (Google around a bit; that link doesn't describe them well.)
*rolls eyes* It is not 'information'. It is awful corporatese that tells me almost nothing. If you had bothered to understand it yourself and posted stuff about your personal experiences or understanding of how Postgress SQL is better than MySQL, I'd be all ears
You mean testimonials like this?
"Business Objects has always provided companies with an open business intelligence platform that offers fast and easy access to a wide variety of applications and data sources. With MySQL Network, organizations can more easily leverage their MySQL database technology with our industry leading business intelligence platform, BusinessObjects XI. Our integrated solution helps customers extend business intelligence across their entire organization reliably, easily, and affordably."
That's what I call awful corporatese. It involves impressive-sounding but generic words and phrases like "leverage", "industry-leading", and "extend business intelligence". It communicates no information.
Read the PostgreSQL blurb again. It may have frightened you with its professional writing, but it was full of well-defined technical terms and specific information. Very few of the things written there apply to MySQL.
I think relational databases in general are evil. The only thing good about them is SQL their support of transactions. And don't tell me that's all they are. If that's all they are, then why can't I 'ls' a table, cat a row, or insert something by creating a file?
That's all they are. Their support for transactions (ACID, to be precise) would not be possible if you could do those things.
How could you possibly guarantee atomicity by appending records to files? There's no way to ensure that both or neither of two operations happen.
How could you guarantee consistency? There's no way to enforce constraints on filesystem activity.
How could you guarantee isolation? You can't make things happen at the same time.
Durability's the only one you could do. A simple fdatasync should be enough.
So DJB's tools perfectly implementing the standard isn't a benefit because other people decided to do it wrong?
No, DJB's tools perfectly implementing a standard isn't a benefit because the designers of the operating system decided on a completely different standard. There's little difference between a correct implementation of an incompatible standard and an incorrect implementation of the standard you're using.
I'm not familiar enough with UTC and TAI to know the distinction, but I do know there is one, and I do know you misunderstood the parent post.
Even if the mission was completely flawless, they would still find something to carp about.
And would they have been wrong? When you say that something is unsafe, you're saying there's a significant unnecessary risk of someone being hurt or killed. If the supposedly unsafe thing happens and no one is hurt, that does not mean it was safe. If everything appears to have happened perfectly, that does not mean it was safe.
"Unsafe" can't be confirmed by someone being hurt (the risk might have been necessary) or by no one being hurt (the point is that they could have been hurt). It's not a direct observable; it's a conclusion.
For specific, falsifiable claims and recommendations, you need to read their report. But I'm sure you did so before challenging the conclusions of a panel of experts. What was wrong with it?
Will it help your company attract excellent developers? (This is a specialization of your "Will it create create good relations to someone the company cares about?") Arguably, yes - people like to work places they know OSS is respected and where they know interesting work is being done.
Will it improve the quality of your code? Patches are one way. Pride is another - people polish stuff much more if other people will see it.
[The Workplace Shell] is a truly object-oriented desktop design that's still superior - a decade later - to anything Windows has to offer.
What does this even mean? Object-oriented is a programming model. Are you talking about their implementation? Because as a user, I don't care about that. I can think of very few useful extensions to desktop interfaces - Tortoise(CVS|SVN) comes to mind. I can't even think of a #2. Thus, with no reason to interact with their code, I don't care what model it uses. If you had the ability to extend the desktop in any way, what would you do with it?
If you're not talking about the implementation, how does an "object-oriented" desktop behave differently?
I hear this every time OS/2 is mentioned, but I used the Workplace Shell and wasn't impressed. My memory is hazy, but I think there was some really stupid stuff - you had to select icons with the left mouse button then drag them with the right, or something. Awkward and counterintuitive.
Thanks for taking the time to put some real numbers to this. You've demonstrated that the overall population's proportion does not change, despite families stopping when they have a girl.
But what cpotoso was saying is that an average family's percentage of girls does change based on this selection. Lets calculate a single family's expected proportion of girls (using this selection algorithm):
1/2 * 1 + 1/4 * 1/2 + 1/8 * 1/3 + 1/16 * 1/4 +... = somewhere above 68%. (The probability of each sequence occuring times the proportion of girls in that sequence, yielding the expected proportion of girls.)
(This is \Sigma_i=1^\inf (\frac{1}{2^i}) (\frac{1}{i}). I'm too tired to figure out what the series actually converges to now, but it's clearly non-decreasing, and the first four terms are 68%. Thus, the result is well above 50%.)
Stopping when you get a girl indeed increases your proportion of girls, as common sense indicates.
These two seemingly incompatible statements can be reconciled when you realize that the families with a lower proportion of girls have more children overall. Thus they are weighted higher in the aggregate.
> > I can sell compiled binaries of grep for a billion dollars each if I can find someone willing to pay that for them. The GPL allows it.
> Um, would you care to state where in the license this is explicitly allowed? Because I can cite where it's explicitly prohibited.
No, you can't. The GPL is explicitly designed to allow you to do this, as stated in the preamble:
When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things.
I suspect you're thinking of this paragraph:
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
But you're not required to give the source to anyone who doesn't have the binary, and you're not required to give the binary away.
I can sell compiled binaries of grep for a billion dollars each if I can find someone willing to pay that for them.
But of course you can't find someone willing to pay. The GPL permits you to sell software, but it also permits any of your customers to undercut you by selling it for a lower price or just giving it away. So if you write some GPLed software, you can probably sell it once. If someone else wrote it, you probably can't sell it at all.
So there are two realistic ways to make money off GPLed software:
Do custom development for a fee. Like a consulting business that develops niche software that probably only one company would ever use. They have more options in case you become unreasonable or go bankrupt, so presumably they'd be willing to pay more than for a proprietary solution.
Sell an aggregation (like Red Hat Enterprise Linux), documentation, or support. This is what those companies you mentioned - RedHat, SuSE, Lindows, etc. - do.
These are valid ways to make a living, but they'll never be as lucrative as Microsoft's business model. Namely, to write the software once, sell it over and over, and sell the extras separately.
I think it's getting worse. They're by no means the only violators; just the most flagrant.
I recently discovered that Acquisition (a popular Mac OS X gnutella client) is using GPLed Limewire code. It's not anywhere on the main Acquisition website, acquisitionx.com. The website implies the whole thing is written by David Watanabe. It's shareware with nag screens. Most users will never know there's Limewire code used. The most obvious place it's mentioned is the fine print of the "About" box.
There is some source available, at AcquisitionX.org. (There are no links from the other site. Found it through some googling.) This is the "core" of Acquisition, a modified version of LimeWire's core code. But the actual UI code is not released. The developer claims this is "full and complete compliance with the LimeWire GPL", but it's not. The key characteristic of the GPL is that you have to release applications that use GPLed libraries under the GPL themselves. He's following the terms of the LGPL instead. If they'd meant to release it under that license, they would have! He's profiting from other people's work without following their license or giving them proper credit.
(Sorry for making people click through. I deliberately have no links to either Acquisition site because I don't want to increase his PageRank.)
Sure you can start them concurrently! Just add strategic "&" signs after the commands that start them up and they will start up in background processes of their own.
And your system won't work anymore, either. The order is important, and if you just add a "&" to the end of the commands, you lose control of that. Conventional init satisfies dependencies by prefixing the scripts with numbers indicating when they are to be launched - S10network, S12syslog, etc - and doing so sequentially. But this is overly conservative; the whole system doesn't start until a single service does, even if nothing depends on it. It also means you have to maintain a total order instead of a DAG, which sucks.
There was also an IBM article called Boot Linux faster on replacing initscripts with make. (This is less weird than it seems. Correctly resolving parallel dependencies is what make was designed to do.)
So Apple's not the only one to see a problem here and replace it. I don't know why the major Linux distributions haven't yet; the init system really sucks. Hopefully they'll take a look at launchd, which I believe is open source.
Interesting. I'd never seen that way before, but it's similar to the self-pipe trick (which is also portable).
sigsafe's advantage is that it works with a lot more syscalls than just the file polling ones, and (hopefully) your code is simpler. The porting's not as bad as you'd think, either. There's no hope of it running on a new system without writing new code, but there's not a lot of code to write. I've got it working on a bunch of systems.
Slashdot Mirror
Yeah, I am. It's pretty rare for me to come across someone who is wearing enough perfume to cause a problem, but it does happen. Usually just a little congestion and eyes watering.
I'm not in favor of outlawing perfume in public, though. Most things like that are solved by basic courtesy (don't put on the whole bottle!) But there really is no amount of smoking that won't cause people like me immediate problems. Besides, even if there were no emperical evidence of direct harm from second-hand cigarette smoke, it still contains known toxins and carcinogens. Common sense dictates that you shouldn't blow those on unwilling people, especially given the way cigarette smoke lingers.
I also have those allergies, and I don't advocate that either. However, if someone (in whatever manner) deliberately put concentrated pet dander into the air all around you whenever you went into public, it would cause you a lot more than watery eyes and irritated sinuses. You would probably call that assault. That's a much more analogous situation.
I, too, value personal freedom. It's just that I value my freedom to go places without being assaulted more than your freedom to assault me.
Okay. I looked up one of those results, Replication of the Apparent Excess Heat Effect in a Light Water-Potassium Carbonate-Nickel Electrolytic Cell. You have to pay for the entire article, but the abstract is freely available. I quote:
The reproduction wasn't nearly as dramatic as the original claim, and there's a potential ordinary explanation. He's making an extraordinary claim, and I want to see extraordinary evidence: eliminate the ordinary explanation and make it produce as much energy as he says for as long as he says.
That is not clear at all. You hear more from the smokers because they're obnoxious assholes. When strangers smoke near me, I generally give them an unpleasant look and move away. They never know or care how much of a problem they're causing me. Do you smoke? Do you get those looks? Next time, ask why.
It is common to some have degree of problems caused by second-hand smoke. The grandparent's watery eyes and headache are similar to my symptoms, just not as severe. People don't speak up because smoking is accepted. That's changing in most areas.
If you want to see scientific studies that link second-hand smoke to short-term medical problems, just google "second-hand smoke asthma". First hit. I've also heard that smoking in the home is either the #1 or #2 risk factor for asthma in children. You can debate that, of course, since correlation is not causation. (Smoking is more common among the poor and in the South.) But there is evidence, and more is coming. For better or for worse, there will come a day when you can have your children taken away for smoking around them.
If I start having seizures whenever I see the color red, am I justified in asking everyone to stop wearing red for me and the other 3 people in the world with my affliction?
No, because you just made that up. I have real physiological problems that are much more common than you think. Do you really think you can blow known carcinogens and toxic chemicals into the air around you and not cause people problems?
A better analogy would be peanut butter allergies. Very similar, fairly rare. There are people around who will die if they eat peanut butter. They demand accurate ingredient lists so that they don't eat peanuts involuntarily. I demand that I don't get second-hand smoke forced on me. They aren't suggesting outlawing the sale or consumption of peanuts. I'm not suggesting banning smoking in your own home. If you want to kill yourself, that's fine. But don't involve me.
Some of us more than a bit ill. Second-hand smoke causes me real short-term problems. I've heard people say that cigarette smoke isn't an allergen; it's chemical sensitivity. Whatever you call it, the effect is the similar to - but more potent than - my other allergies. My eyes burn and water, I cough and sneeze, and my sinuses fill, which often leads to a sinus infection, which means a week of misery and $100 of antibiotics. Probably a day missed from work. I used to live in Iowa, so there were a number of places I just couldn't go - bars, bowling alleys, etc. I had sinus surgery to to prevent the infections, with disappointing results.
I've moved to California, where they have seen reason. I can go anywhere. I'm incredibly sensitive to smoke, so I do have occasional problems still. I'll smell smoke and see a coworker who walked past and is now 100 feet away; the smell lingers. If I smell it, I know it will cause me problems. If I don't smell it, it still might. (I once had a pretty bad reaction to a hotel room that smelled just fine to me. Never found out why.) I deliberately shun people who smell like smoke. If I am forced to be near them, I take a shower as soon as I get home.
That's definitely why I switched out of engineering (into CS and physics). Lots of meaningless homework. Cheating was rampant. It seemed that there were three classes of students:
In my physics classes[*], almost everyone was in the second category. Except that doing the work really did pay off. The homework was different: fewer, more difficult problems. You'd actually be missing something if you skipped half the work.
There were a lot of other differences that made the physics program more satisfying. It was a smaller college, so you got a lot more attention from the instructors. A tighter community. More pleasant all around.
Shortly before I started engineering, a crazed physics TA went on a shooting rampage through my campus, killing seven people before he turned the gun on himself.
I'm relieved to see from your resumé that you're a fellow Iowa alumnus. For a second I wondered if there were more stories like this out there...
[*] Physics for majors, that is. I took the first physics for engineer class, and it was a totally different experience. Lecture with 300 people. Multiple choice tests (five choices), mean score of 3/12.
> Sure it does: [...] fornications [...] fornication [...] fornication
Try reading sometime. Here's the part of the grandparent post that you missed:
"Fornication" is an English word, never found in the original text. This may shock you, but language can be ambiguous. Thus, translations can be wrong. This is why Muslims consider only Arabic versions of the Qur'an to be correct. Sometimes I wish Christians did the same.
It's Attorney General Gonzalez and Director Robert Mueller .
I thought this was just the submitter's mistake, but it is actually that way in the article. I shouldn't be surprised; Ars Technica should stick to their stupid overclocking articles.
I disagree. The farther away the documentation is from the source code, the less likely it is to get written or updated. The best-maintained documentation is inline, like Javadocs, Python docstrings, pod, or doxygen comments. When you make a change to the source, the documentation you're invalidating is right there. It takes almost deliberate blindness to not keep it in sync.
Second best is a wiki. It's not right next to the code, but it's so easy to change. Also great if you have documentation that you share with other teams. (QA, other development groups, etc.) Wikis let them easily change things without digging through your source code.
(even when they do, its easy to typo and break the generator).
Then make the doc processing part of the normal build process. Especially if you're an automatic builder like Cruise Control. Just make that one of the steps. If it fails, Cruise Control sends out an email to everyone with the error message and the changes since the last successful build.
Secondly, for good documentation you need more than what an autogenerator can do anyway- good documentatition has examples, use cases, warnings and gotchas, etc.
There's no such thing as a documentation autogenerator. Just tools that reformat, add links and context, etc. You still have to write the words, and you still get to choose what they are. Having examples, use cases, warnings and gotchas is orthogonal to the software you use.
That is exactly the MySQL attitude. The problem is that my data tend to be important, or I wouldn't be putting them in a RDBMS. I'd rather the database assume the usual, safe thing - data should not silently truncated. If I want to throw away my data, I'll do it explicitly.
And filling in NOT NULLS is also handy for the same reason.... if you REALLY need the logic to reject things like blank inputs in web forms then you should be doing that in your application logic anyway....
What I REALLY need is for the database to pay attention to the constraints I specify. If I didn't care if these fields were blank, I wouldn't have said "NOT NULL".
I am by no means sticking up for non-conformity... just trying to relay the idea that in the context that mysql is usually used, these small quirks don't have a large impact.
That is not the context in which mysql.com is claiming their database is usually used. They claim they have a real database.
I fail to understand the logic behind avoiding solid products because someone who likes it annoyed you on slashdot. If I made decisions like that, I wouldn't use:
In fact, I can't think of a single significant piece of software I would use. I'd have to get a new job.
I don't think you understand how major an undertaking you're proposing.
Getting ACID right is hard. Let's say I'm a RDBMS developer. I have two choices:
Read Codd's 12 rules to truly understand what a RDBMS does for you. (Google around a bit; that link doesn't describe them well.)
You mean testimonials like this?
That's what I call awful corporatese. It involves impressive-sounding but generic words and phrases like "leverage", "industry-leading", and "extend business intelligence". It communicates no information.
Read the PostgreSQL blurb again. It may have frightened you with its professional writing, but it was full of well-defined technical terms and specific information. Very few of the things written there apply to MySQL.
That's all they are. Their support for transactions (ACID, to be precise) would not be possible if you could do those things.
How could you possibly guarantee atomicity by appending records to files? There's no way to ensure that both or neither of two operations happen.
How could you guarantee consistency? There's no way to enforce constraints on filesystem activity.
How could you guarantee isolation? You can't make things happen at the same time.
Durability's the only one you could do. A simple fdatasync should be enough.
No, DJB's tools perfectly implementing a standard isn't a benefit because the designers of the operating system decided on a completely different standard. There's little difference between a correct implementation of an incompatible standard and an incorrect implementation of the standard you're using.
I'm not familiar enough with UTC and TAI to know the distinction, but I do know there is one, and I do know you misunderstood the parent post.
And would they have been wrong? When you say that something is unsafe, you're saying there's a significant unnecessary risk of someone being hurt or killed. If the supposedly unsafe thing happens and no one is hurt, that does not mean it was safe. If everything appears to have happened perfectly, that does not mean it was safe.
"Unsafe" can't be confirmed by someone being hurt (the risk might have been necessary) or by no one being hurt (the point is that they could have been hurt). It's not a direct observable; it's a conclusion.
For specific, falsifiable claims and recommendations, you need to read their report. But I'm sure you did so before challenging the conclusions of a panel of experts. What was wrong with it?
What does this even mean? Object-oriented is a programming model. Are you talking about their implementation? Because as a user, I don't care about that. I can think of very few useful extensions to desktop interfaces - Tortoise(CVS|SVN) comes to mind. I can't even think of a #2. Thus, with no reason to interact with their code, I don't care what model it uses. If you had the ability to extend the desktop in any way, what would you do with it?
If you're not talking about the implementation, how does an "object-oriented" desktop behave differently?
I hear this every time OS/2 is mentioned, but I used the Workplace Shell and wasn't impressed. My memory is hazy, but I think there was some really stupid stuff - you had to select icons with the left mouse button then drag them with the right, or something. Awkward and counterintuitive.
But what cpotoso was saying is that an average family's percentage of girls does change based on this selection. Lets calculate a single family's expected proportion of girls (using this selection algorithm):
1/2 * 1 + 1/4 * 1/2 + 1/8 * 1/3 + 1/16 * 1/4 + ... = somewhere above 68%. (The probability of each sequence occuring times the proportion of girls in that sequence, yielding the expected proportion of girls.)
(This is \Sigma_i=1^\inf (\frac{1}{2^i}) (\frac{1}{i}). I'm too tired to figure out what the series actually converges to now, but it's clearly non-decreasing, and the first four terms are 68%. Thus, the result is well above 50%.)
Stopping when you get a girl indeed increases your proportion of girls, as common sense indicates.
These two seemingly incompatible statements can be reconciled when you realize that the families with a lower proportion of girls have more children overall. Thus they are weighted higher in the aggregate.
> Um, would you care to state where in the license this is explicitly allowed? Because I can cite where it's explicitly prohibited.
No, you can't. The GPL is explicitly designed to allow you to do this, as stated in the preamble:
I suspect you're thinking of this paragraph:
But you're not required to give the source to anyone who doesn't have the binary, and you're not required to give the binary away.
But of course you can't find someone willing to pay. The GPL permits you to sell software, but it also permits any of your customers to undercut you by selling it for a lower price or just giving it away. So if you write some GPLed software, you can probably sell it once. If someone else wrote it, you probably can't sell it at all.
So there are two realistic ways to make money off GPLed software:
These are valid ways to make a living, but they'll never be as lucrative as Microsoft's business model. Namely, to write the software once, sell it over and over, and sell the extras separately.
I recently discovered that Acquisition (a popular Mac OS X gnutella client) is using GPLed Limewire code. It's not anywhere on the main Acquisition website, acquisitionx.com. The website implies the whole thing is written by David Watanabe. It's shareware with nag screens. Most users will never know there's Limewire code used. The most obvious place it's mentioned is the fine print of the "About" box.
There is some source available, at AcquisitionX.org. (There are no links from the other site. Found it through some googling.) This is the "core" of Acquisition, a modified version of LimeWire's core code. But the actual UI code is not released. The developer claims this is "full and complete compliance with the LimeWire GPL", but it's not. The key characteristic of the GPL is that you have to release applications that use GPLed libraries under the GPL themselves. He's following the terms of the LGPL instead. If they'd meant to release it under that license, they would have! He's profiting from other people's work without following their license or giving them proper credit.
(Sorry for making people click through. I deliberately have no links to either Acquisition site because I don't want to increase his PageRank.)
And your system won't work anymore, either. The order is important, and if you just add a "&" to the end of the commands, you lose control of that. Conventional init satisfies dependencies by prefixing the scripts with numbers indicating when they are to be launched - S10network, S12syslog, etc - and doing so sequentially. But this is overly conservative; the whole system doesn't start until a single service does, even if nothing depends on it. It also means you have to maintain a total order instead of a DAG, which sucks.
NetBSD 1.5 replaced init, too. See The Design and Implementation of the NetBSD rc.d system for details.
There was also an IBM article called Boot Linux faster on replacing initscripts with make. (This is less weird than it seems. Correctly resolving parallel dependencies is what make was designed to do.)
So Apple's not the only one to see a problem here and replace it. I don't know why the major Linux distributions haven't yet; the init system really sucks. Hopefully they'll take a look at launchd, which I believe is open source.
sigsafe's advantage is that it works with a lot more syscalls than just the file polling ones, and (hopefully) your code is simpler. The porting's not as bad as you'd think, either. There's no hope of it running on a new system without writing new code, but there's not a lot of code to write. I've got it working on a bunch of systems.