Completely amazing that no one has *adequately* answered you on this question.
That's true.
I can appreciate that one thing I hadn't realised, made clearer by the responses, was that postgresql doesn't yet have a stable Windows build, and I suppose that's a reasonable enough reason to use MySQL instead. I hadn't noticed the easy-installation issue at all, but then I run Debian. I installed postgres with apt-get, and didn't have much of a problem adjusting the configuration files afterwards.
Most of the rest of it seems to be about MySQL being driven by it's existing popularity. People use it because they already have it, or because the application they want to use only works with it, and so on.
So I suppose there's reason enough why people currently choose MySQL over others, especially Postgres which is free. But if those responses are anything to go by, then MySQL doesn't seem very encouraging in it's future prospects.
Re:Picking the right tool for the job
on
Why MySQL Grew So Fast
·
· Score: 4, Interesting
If application bigotry keeps you from choosing the right tool for a job, you will be a less valuable resource to those who employ you.
Pardon my ignorance, but what is it exactly that MySQL is good at that Postgresql isn't at least equal or better at? I'm all for using the right tool for the right job, but every time I've used at MySQL I've frequently encountered all sorts of crazy problems with no killer features to justify the change, that aren't already provided by both postgres and other databases. Maybe it's my application bigotry speaking, but I honestly can't see how MySQL is better at anything.
The only things I can think of are the familiarity aspect, for people who are already very knowledgable about MySQL not wanting to switch, and possibly some minor issues with moving database files around on the underlying file system.
If you haven't already, I'd go and read Asimov's stories about robots, or at least consider them in a different light from how you may have been doing so far.
He's openly stated on several occasions that the three laws of robotics obviously aren't that simple and that there are a lot of ambiguities. He's also commented that exploring these ambiguities, such as "what would a robot do in situation X?", has been a main plot device for many of his stories.
His stories intentionally aren't about robots running around killing people. They're about a society living amongst robots that have been programmed with the intent of being non-threatening, and how it might work. If it was all so predictible then there wouldn't be much to write about.
If you don't think fines are the solution, then what do you suggest? Gaol sentances? Might work, but who do you put in gaol?
The company executives. Unfortunately I don't suppose the law would directly support this.
The law is there to define what's not allowed. Fines are intended as a deterrant, they are not intended as compensation for society. Opting to simply pay the fines is not a valid excuse to break the law.
Company strategies don't appear out of thin air. They're designed and implemented, with intent, by people in the company. If it can be demonstrated that the company strategy involves ignoring the law in favour of fines, then it should be considered ethically okay for the company executives to be directly charged with corporate mis-management.
If the execs weren't involved, then by all means charge the next person down who did know, and if necessary disbar any lawyers who either agreed with or went along with breaking a law. But go after individuals who know that they're breaking the law.
If the accusations are correct that Microsoft's business plan is to break the law, then surely there are at least several people in-the-know who should immediately be able to be put on trial.
I agree completely. Most of the disagreements that I've had with people about management of "their" information that I have is a result of us having different perceptions of what's going on.
I get strange looks from some people when I mention in-passing to them that I rarely delete email. (From my perspective, disk space is cheap.) For the majority of techie people who I know, this is completely ordinary. But for other people who are used to deleting email soon after they've read it and not keeping it around, it seems weird. Sometimes they even express discomfort that I keep their email permanently, but from my point of view it's just keeping records of correspondence in the same way that many people don't throw away old letters.
I often get the same reaction when I look at the source code of someone's website. If someone's watching me they often consider it equivalent to hacking the site. I've had a few reactions like "how dare you look at my code?" and so on. From my perspective it's completely normal, since I just view HTML of a web page as open information that's not encoded any more than it is, and viewin the HTML directly is merely a different view of the same information. But from their point of view, it was never imagined that anyone would be able to see it.
I don't use IRC a lot, but if I did it would be completely natural for me to treat it as something that could be logged and saved. After all, from my perspective it's written communication and I like to be able to keep records of my written correspondence.
I think one of the main problems with this type of law is that it's based on an incorrect mental model of how something works. People don't expect something might happen because they've never used the service that way themselves, and so they assume that anyone who uses it differently must be up to something suspicious.
So should there be a law that you have to delete old emails unless you have permission from the people who sent them? Should there be a law that you have to throw out old letters that people have posted to you? Should there be a law against viewing the http transmissions of raw HTML text that are coming through your connection for your web browser to display? Should there be a law against keeping logs of IRC discussions? My personal beleif is that none of these laws would make sense.
Personally I live somewhere (New Zealand) that allows for correspondence to be recorded as long as at least one party knows about it. It nicely fits my mental model of information propagation, and I hope that law stays as it is.
I make a living off Plone support (and training/development), and see no difference from a "commercial" product.
Perhaps I've missed a point somewhere, but isn't this exactly what fahrvergnugen just claimed? You both seem to be in agreement that paid support for a product is a good thing, so I don't see the problem.
The grandparent comment was claiming that commercial support wasn't as important since standard open source developer support (unpaid for) might be just as good in different ways.
if you always get a negative reinforcement for an action, operant conditioning will cause the drivers to slow down. tickets and cops are not regular enough to train people to stop.
I'm skeptical about the possibility of implementing this safely.
By one approach, you could implement it so that as the light goes red, traffic is started as usual from other directions. But then you run the risk of drivers getting confused by the unexpected change, and skidding through the intersection.
By the other approach, you could implement it so that the light simply turns red and the intersection is dormant for at least a few seconds. But then you run the risk that some selected drivers will learn that they can get away with running the red light without consequence. (Hey, if you could ticket them then you may as well have ticketed them for speeding in the first place, before the red light thing was proposed.) Though they might get away with it here, it could cause serious accidents at other intersections if a driver misunderstands what's going on.
Personally I think that adding an extra semantic meaning behind the reason to stop at a traffic light could create all sorts of new traffic problems.
Now instead of talented, inspired artists putting an album together that means something to them (Beatles Sgt Pepper, Pink Floyd Dark Side of the Moon), you get a Stripper singing meaningless lyrics to a computerized drumbeat and bassline, while drinking a Pepsi.
Maybe it's just the genre of music that I listen to, which tends to be trip hop, down beat, etc. (Ranging around things like Massive Attack, Zero 7, Bic Runga, Hooverphonic, Morcheeba, Delerium, Red Hot Chilli Peppers, and various forms of classical music.) Perhaps I have much worse taste than I thought I did, but I haven't experienced many strippers singing meaningless lyrics to a computerized drumbeat and bassline while drinking a Pepsi at all.
On the contrary, most of what I have amounts to complete albums with organisation that gels together nicely, and I frequntly listen to albums all the way through without getting sick of certain tracks. Granted that I do take my time before buying music, as I want to make sure that it's not crap.
I don't support the RIAA at all, but I'm sure that there are a lot of well crafted albums out there if you take the time to look for them. Saying that nearly all RIAA affiliated music is crap isn't correct -- from my experience it's only the top-level manufactured music that tends to be that way, and I don't buy that stuff anyway.
What desktop linux needs is ONE desktop to replace them all. That is; one set of widgets, one way of doing everything, and one interface for developing gui apps for linux. This kind of dictatorship works dandy at the core level of linux, and needs to be extended to include the GUI. [..] As long as there is choice, there will be no breakthrough. One more choice won't help either.
It's a cute idea, but like another response to your post I also don't see it as realistic and in some places I completely disagree with what you've said. Someone else might elaborate since I'm not an expert on kernel development, but I also think you've missed some important points about how things work.
Linux can have a dictatorship and "one way of doing everything" at the kernel level, because by definition the kernel is linux. If someone were to fork the kernel and do things differently, it wouldn't be linux any more. Similarly, if someone forks emacs it becomes something else. If someone forks X11, it becomes something else. And so on.
Nobody's seriously and successfully forked the linux kernel for one reason or another... or at least if they have, it's not called linux anymore. But there are several other kernels in existence that are available and work significantly differently. Even if nobody bothers to fork the linux kernel, some people may go and work on the BSD kernels, for instance, because they prefer the design.
In essence, as long as enough people disagree about the best way to do something, there will be a fork. It happens with nearly every application available as much as, if not more than, it happens with kernels.
Desktops are a huge area of disagreement. The design of them is mostly about usability, and we're still in infancy when it comes to understanding the best ways to do things. ACM has only been running HCI conferences since the early 1980's, and since then researchers have figured out that designing good desktops is very difficult. Putting rules on it might make it slightly easier to be compatible with or learn, but placing draconian enforcement on a policy that isn't known to be good is more likely just to leave us with another crappy desktop.
Windows is a crappy desktop from a usability perspective. Personally I prefer to avoid KDE and Gnome, both of which seem to want to mimic Windows in most ways, including most of it's bad features. For a linux desktop I prefer WindowMaker, which also isn't perfect, but is has several features that I just like. Having the option to switch and still have all of my X applications work is fantastic.
I was with you until you mentioned inter-car communication. Now you're allowing a whole new source of untrusted data into the decision-making process. What happens when a group of script-kiddies figure out how to DDOS a freeway with fake emergency brake messages?
Probably the same thing that would happen if someone hacked the railway signalling system to force a few trains to stop (or worse, crash). Some people may be killed or injured, lots of people get annoyed (many more people than a few sysadmins), the police chase them down, and they quite possibly get into a hell of a lot of trouble.
From the way you describe it, though, you also seem to be implying that this whole system may be somehow connected to the Internet. Bureaucratic decisions aside, there shouldn't be any need for the system to be connected to anything. The road indicates it's shape (edges, lanes, corners, turnoffs, etc), and the cars drive themselves on it.
Other ways that it might be broken could be to hack your car's driving system, or to simply stand near the road waving around some fake transmissions (depending on how the communications worked). My personal belief is that worrying about this type of thing happening is starting to get a bit too cautious. To me it doesn't seem like the sort of apparently harmless activity that most script kiddies would be interested in, aside from being a lot more cost and effort, and probably worse consequences.
Sure, the information provided by another car is untrusted. But so is just about everything about another car already that's usually taken for granted. When you're out driving on the road, for instance, you normally assume that the drivers around you are suitably trained to drive, that the bits of their vehicle aren't going to fall off, that their wheels have enough tread to properly grip the road, that the other drivers haven't been drinking, and so on.
The need to make sure that people and vehicles who drive on the road are using it safely is the reason that most governments operate systems that both require drivers to be licensed, and (in this case) require cars to be maintained to a certain standard and operate in particular ways. This is for the safety of everyone else as much as the safety of any particular vehicle's occupants. There shouldn't be much reason why such requirements can't be extended to the implementation and interface of an auto driving system that would work in this way. Probably people will hack them from time to time to provide false information to other parts of the system, and probably those people will get into an appropriate amount of trouble because of it.
In any case, I'm just brainstorming at the moment as I was when I wrote the previous comments. My personal belief is that untrusted information would not be a very serious problem under the circumstances.
What if, unexpectedly, a car's engine stalls? If computers could brake immediately, there is still the issue of decelerating...
Well I'm not a physics expert, but the thought that immediately comes to mind is that if an engine stalls, a vehicle still has momentum and won't stop suddenly unless the brakes are actually applied. Consequently, everything behind could still decelerate at the same rate.
But maybe you really meant what might happen if one vehicle simply has to stop suddenly by slamming on the brakes. You'd certainly hope that if one car did that, then all of the following cars would immediately do the same to avoid collision... almost certainly, all following cars could quite easily be stopping before their occupants realised there was a problem. If this were all done well, with proper inter-car communication, there doesn't seem to be much reason why it might not work.
Depending on the types of cars, the computer in each car would still need to allow for some gap in the following distance, based on what it knows about how much time it would take to stop, versus how much time the car in front would take to stop. It'd also need to take into account a required safe deceleration rate that wouldn't kill the people inside the car if it had to stop suddenly. (Irrespective of debates about whether governments should force people to wear seat belts, I still think it's stupid not to wear one.)
A lot of the distance that's presently needed for safe driving on roads is because of driver reaction time. Lots of drivers seem to have the impression that because they're driving a modern car, it's safer for them to drive faster and (to an extent) more recklessly. Unfortunately, as long as a human is controlling the car, you can't beat the human reaction time. It simply takes x fractions of a second for people to react, irrespective of how fast they think they're reacting, and irrespecting of how suddenly their vehicle can stop after they've hit the brakes. A computer controlled car would remove most of this reaction time, making it easier to reduce the safety gap between cars.
This is completely true. Another major advantage of autonomous cars is that they could simply follow more closely, letting more cars occupy the road at once, travelling at higher speeds, getting to where they're going more quickly and reducing much of the energy waste and pollution that's often associated with low speed stop-start driving.
With a road that's designed for it, as well as cars that are designed to communuicate with the road and the other cars on it, traffic congestion could be reduced hugely. As long as they have reliable data, computers are capable of reacting several orders of magnitude faster than humans are.
It might be substantially more difficult to implement a system like this for city driving, mostly due to uncontrollable parameters like pedestrians. But it doesn't seem that unreasonable to implement it on high speed roads. The main barriers are upgrading the roads to support the cars, and making the cars capable of driving on the roads. Perhaps, to do this, you might allocate a lane or two at a time for a while, as the infrastructure changes.
I can't shake the feeling that these people are getting paid a lot solely because they are associated directly with the characters, not because voice talent is hard to come by.
Part of me is inclined to agree, but then the other part of me realises that they've been doing this for more than 15 years. It's quite possibly a lot more of a tedious chore than it is interesting.
They've identified a figure for which they'll be satisfied to put up with the job in future times, no matter how boring or frustrating it gets. The alternative is that Fox may decide it's too much --- the show will end, and the voice actors can go and spend their time doing something that they find much more interesting. It's quite possible that they may be more than happy to go and do another show for much less, simply because it's different.
I suppose that just because the show is popular doesn't mean that the voice actors should be required to do it forever at a wage set by the studio. The raise is what they're claiming will be needed in order to keep them satisfied to continue doing it... perhaps they won't get it and they will no longer work on the show, but I'm sure they won't mind.
They will just issue themselves diplomatic passports so they can simply bypass those procedures.
Well I tend to agree with you, although it should be acknokledged that the extra rights available with a diplomatic passport still only exist because the hosting country decides to grant them. Revoking the rights would probably mean breaking international agreements, though, which I guess is why it's never going to happen. (If you actually care about things enough to break major international agreements, chances are that things are already much more seriously messed up than what this particular activity would resemble.)
I tended to think the same for a while, but having considered it, I have to admit that most international US travellers who I've met here (I'm in New Zealand) have been much more enlightened about the rest of the world than the people who probably back this type of legislation in the US.
I don't know if it's a smart thing to punish them just because their government does something stupid, and I also don't want to contribute to discouraging US citizens from actually getting out and seeing the rest of the world.
On the other hand, I'd have absolutely no problem with requiring tedious entry procedures for US government officials. The thought of Bush and his entire travel comeraderie being required to have their photo and fingerprints taken before entering other countries just makes me laugh. It won't happen, of course -- the US just has far too much international influence.
I would love to see a concerted effort by everyone to avoid doing any business with the Americans until they come down off their high horse and start treat others with some respect.
Personally I would at least like to see all people who are closely connected with the current US federal administration (diplomats and presidents included) be forced to go through the process of having their photo and fingerprints taken whenever they try to travel to any other country.
I'm tempted to wish it for all US citizens, but I don't particularly want to sink to that level. To their credit, most Americans who actually travel internationally tend to be a bit more open minded (in my experience of meeting them so far) than the people who tend to support this.
All the farmers come there to sell their produce, eggs, honey, etc. they all pretty much have the same things (tomatoes, beets, lettuce, eggs, etc) and so all are basically in competetion with each other. HOWEVER, listen in when the farmers are talking to each other, and you will find that they very readily help each other out with tips, advice, and even labor.
It's an interesting analogy, but I'm not convinced that it describes things perfectly. Farmers who sell produce are in a different type of market from software. The former is very general with lots of small players. The latter is quite specialist, with several very big players who have a lot of influence as well as lots of smaller players who often have to follow.
Although farmers may often sell the same things in the same locations at the same time, they're not automatically in direct competition with each other. The simple reason is that on their own, one farmer can't supply enough goods to meet all of the demand. One farmer can almost never set the market price, and as long as s/he is selling at the market price a farmer is likely to sell all of the goods... with the occasional exception in different circumstances.
With the occasional exception, sales for one farmer aren't lost because another farmer has more to sell. And because there are so many small players in the market who are selling the same goods, any one farmer doesn't have much influence over the price, anyway. All the goods will probably be sold at the price set by the market. (Exceptions may be in small markets where there are only a couple of farmers and nobody else anywhere selling the same thing.)
It'd be funny to browse the logs if it did, but I'd hope that most people wouldn't do that as habit. It would pollute the statistics for website developers that describe how many people are using Mozilla/whatever, possibly making it appear less popular than it actually is -- at least to the less-educated.
If a simple count shows 85% IE and 12% Mozilla, it's much more likely that more emphasis will be placed on Mozilla support than if it showed 85% IE and 15% unidentifiable... in which case it's probably more likely for a typical admin to simply support IE and request that anyone who cares switches to it.
In either case, a good developer would hopefully at least code to standards wherever possible, but not all developers are perfect and depending on the application it's not always possible to do something using available published standards.
This whole argument that's tied in with many people's need to divide things into discrete groups, even if such groups don't exist, strikes me as more than a bit childish. We already know what it is, who cares how it's categorised? Let's just get on with learning more about it.
Richard Feynman had some great things to say about this on a BBC Horizon documentary some time ago. You can attach names to things as much as you like, but having a name for it doesn't compare with actually understanding it.
My parents have been using Netscape for browsing and email since they started using the Internet... I think it was a carryover from one of the ISP CD's some time ago, but they upgraded to Netscape 6 when it came out.
Recently, after my dad bought a new computer with XP pre-installed, he was even able to set up a dial-up networking connection. But since Netscape wasn't available he didn't know how to get to the web. (He couldn't actually find IE, even though it was certainly installed.) Anyway, I've now installed Mozilla for him, and he's really happy with it.
Interesting to note is the distinct lack of a 10th planet:)
You can talk about whether Sedna should be called a planet or an asteroid for ever, but really it's just trivial. You could also ask why several other objects haven't been called planets, or why pluto has. The best answer you're ever likely to get is that changing things would be too much controversy to be bothered with, it would make lots of teaching material out of date, and it would start a slippery slope to make the names more ambiguous than they already are.
What's really interesting about Sedna is that nobody expected it to be there at all, and nobody yet has any well accepted theory about how it got there. It's further away than the expected range of the Kuiper belt, it's too close for the Oort cloud, and until now nearly everyone expected that to be mostly empty space... certainly nowhere that an object of Sedna's mass would be found. It also has a very eccentric orbit. Most astronomers out there are much more interested in this type of thing than trivial naming issues. It doesn't stop the media from trying to create a story out of nothing, however.
Well I'd like colonists to at least be careful
on
Melting Europa
·
· Score: 1
Well I'm an amateur astronomer and although I don't think that type of progress can be stopped, I do tend to sympathise.
Realistically if you think about the history of the western culture that, today, dominates space exploration, it's very likely to result in a lot of plundering and abuse for personal (or possibly corporate) gain. This will be thanks to the relative lawlessness on another planet before any serious long term colony effort comes along and will be required to pick up the pieces.
Historically this is just what happens. Even if you discount the American colonials wiping out the Indians, Pizarro destroying civilisations in South America for his personal fortune, as well as the Brits beating up nearly every tribal culture they came accross in their conquest of the world, there's still the big impact of environmental damage that's caused by colonials making dramatic changes such as burning down all the forests so they can overfarm the land.
I realise that a lot of this will be inevitable, and obviously Mars is a bit different. But despite it's immediate irrelevance to easy human colonisation, Mars does have it's own environment that's has all sorts of interesting things about it. I would, for once, like to see a new place being valued for its existing qualities when it's first properly visited, instead of them all being turned upside down for short term profits of a select few, only to have future generations kicking themselves at what's later discovered to have been lost.
I think stuff like MySql is great for small operations, but they are hardly enterprise worthy.
I agree with you to an extent, although moreso for Postgres than MySQL, the latter of which is insulting and not worthy of being labelled a real database, imho.
I also think that this is exactly why open source is such a threat to the big products like Oracle and SQL Server. The big databases certainly do have a lot of features. Certainly they're capable of much more than open source products. But if you think about it really, how many of the users actualy use all of those features?
The places where OSS products can cut into the market are with all of those customers who have a big, expensive commercial database that they really don't need. If you only use your databases for inserting, updating, selecting, stored procedures and having some integrity built in, and of course if you have an admin with some idea of what they're doing, then something like postgresql may be perfectly reasonable for your needs. It's not a top level database but for what it does, it does well and just as reliably as anything else on the market. It may not be the best choice if you have extreme load conditions or whatnot, but a lot of commercial vendor's customers don't.
This is what the commmercial vendors really have to watch out for. Although most OSS databases don't offer the breadth of features, they are starting to be viable substitutes for the majority of database tasks. It's certainly possible that the commercial vendors might suddenly find a lot of their customers disappearing from underneath them.
Slashdot Mirror
That's true.
I can appreciate that one thing I hadn't realised, made clearer by the responses, was that postgresql doesn't yet have a stable Windows build, and I suppose that's a reasonable enough reason to use MySQL instead. I hadn't noticed the easy-installation issue at all, but then I run Debian. I installed postgres with apt-get, and didn't have much of a problem adjusting the configuration files afterwards.
Most of the rest of it seems to be about MySQL being driven by it's existing popularity. People use it because they already have it, or because the application they want to use only works with it, and so on.
So I suppose there's reason enough why people currently choose MySQL over others, especially Postgres which is free. But if those responses are anything to go by, then MySQL doesn't seem very encouraging in it's future prospects.
Pardon my ignorance, but what is it exactly that MySQL is good at that Postgresql isn't at least equal or better at? I'm all for using the right tool for the right job, but every time I've used at MySQL I've frequently encountered all sorts of crazy problems with no killer features to justify the change, that aren't already provided by both postgres and other databases. Maybe it's my application bigotry speaking, but I honestly can't see how MySQL is better at anything.
The only things I can think of are the familiarity aspect, for people who are already very knowledgable about MySQL not wanting to switch, and possibly some minor issues with moving database files around on the underlying file system.
If you haven't already, I'd go and read Asimov's stories about robots, or at least consider them in a different light from how you may have been doing so far.
He's openly stated on several occasions that the three laws of robotics obviously aren't that simple and that there are a lot of ambiguities. He's also commented that exploring these ambiguities, such as "what would a robot do in situation X?", has been a main plot device for many of his stories.
His stories intentionally aren't about robots running around killing people. They're about a society living amongst robots that have been programmed with the intent of being non-threatening, and how it might work. If it was all so predictible then there wouldn't be much to write about.
The company executives. Unfortunately I don't suppose the law would directly support this.
The law is there to define what's not allowed. Fines are intended as a deterrant, they are not intended as compensation for society. Opting to simply pay the fines is not a valid excuse to break the law.
Company strategies don't appear out of thin air. They're designed and implemented, with intent, by people in the company. If it can be demonstrated that the company strategy involves ignoring the law in favour of fines, then it should be considered ethically okay for the company executives to be directly charged with corporate mis-management.
If the execs weren't involved, then by all means charge the next person down who did know, and if necessary disbar any lawyers who either agreed with or went along with breaking a law. But go after individuals who know that they're breaking the law.
If the accusations are correct that Microsoft's business plan is to break the law, then surely there are at least several people in-the-know who should immediately be able to be put on trial.
I agree completely. Most of the disagreements that I've had with people about management of "their" information that I have is a result of us having different perceptions of what's going on.
I get strange looks from some people when I mention in-passing to them that I rarely delete email. (From my perspective, disk space is cheap.) For the majority of techie people who I know, this is completely ordinary. But for other people who are used to deleting email soon after they've read it and not keeping it around, it seems weird. Sometimes they even express discomfort that I keep their email permanently, but from my point of view it's just keeping records of correspondence in the same way that many people don't throw away old letters.
I often get the same reaction when I look at the source code of someone's website. If someone's watching me they often consider it equivalent to hacking the site. I've had a few reactions like "how dare you look at my code?" and so on. From my perspective it's completely normal, since I just view HTML of a web page as open information that's not encoded any more than it is, and viewin the HTML directly is merely a different view of the same information. But from their point of view, it was never imagined that anyone would be able to see it.
I don't use IRC a lot, but if I did it would be completely natural for me to treat it as something that could be logged and saved. After all, from my perspective it's written communication and I like to be able to keep records of my written correspondence.
I think one of the main problems with this type of law is that it's based on an incorrect mental model of how something works. People don't expect something might happen because they've never used the service that way themselves, and so they assume that anyone who uses it differently must be up to something suspicious.
So should there be a law that you have to delete old emails unless you have permission from the people who sent them? Should there be a law that you have to throw out old letters that people have posted to you? Should there be a law against viewing the http transmissions of raw HTML text that are coming through your connection for your web browser to display? Should there be a law against keeping logs of IRC discussions? My personal beleif is that none of these laws would make sense.
Personally I live somewhere (New Zealand) that allows for correspondence to be recorded as long as at least one party knows about it. It nicely fits my mental model of information propagation, and I hope that law stays as it is.
Perhaps I've missed a point somewhere, but isn't this exactly what fahrvergnugen just claimed? You both seem to be in agreement that paid support for a product is a good thing, so I don't see the problem.
The grandparent comment was claiming that commercial support wasn't as important since standard open source developer support (unpaid for) might be just as good in different ways.
Okay, maybe I have a wide taste -- they both sing, right? :)
I'm skeptical about the possibility of implementing this safely.
By one approach, you could implement it so that as the light goes red, traffic is started as usual from other directions. But then you run the risk of drivers getting confused by the unexpected change, and skidding through the intersection.
By the other approach, you could implement it so that the light simply turns red and the intersection is dormant for at least a few seconds. But then you run the risk that some selected drivers will learn that they can get away with running the red light without consequence. (Hey, if you could ticket them then you may as well have ticketed them for speeding in the first place, before the red light thing was proposed.) Though they might get away with it here, it could cause serious accidents at other intersections if a driver misunderstands what's going on.
Personally I think that adding an extra semantic meaning behind the reason to stop at a traffic light could create all sorts of new traffic problems.
Maybe it's just the genre of music that I listen to, which tends to be trip hop, down beat, etc. (Ranging around things like Massive Attack, Zero 7, Bic Runga, Hooverphonic, Morcheeba, Delerium, Red Hot Chilli Peppers, and various forms of classical music.) Perhaps I have much worse taste than I thought I did, but I haven't experienced many strippers singing meaningless lyrics to a computerized drumbeat and bassline while drinking a Pepsi at all.
On the contrary, most of what I have amounts to complete albums with organisation that gels together nicely, and I frequntly listen to albums all the way through without getting sick of certain tracks. Granted that I do take my time before buying music, as I want to make sure that it's not crap.
I don't support the RIAA at all, but I'm sure that there are a lot of well crafted albums out there if you take the time to look for them. Saying that nearly all RIAA affiliated music is crap isn't correct -- from my experience it's only the top-level manufactured music that tends to be that way, and I don't buy that stuff anyway.
It's a cute idea, but like another response to your post I also don't see it as realistic and in some places I completely disagree with what you've said. Someone else might elaborate since I'm not an expert on kernel development, but I also think you've missed some important points about how things work.
Linux can have a dictatorship and "one way of doing everything" at the kernel level, because by definition the kernel is linux. If someone were to fork the kernel and do things differently, it wouldn't be linux any more. Similarly, if someone forks emacs it becomes something else. If someone forks X11, it becomes something else. And so on.
Nobody's seriously and successfully forked the linux kernel for one reason or another ... or at least if they have, it's not called linux anymore. But there are several other kernels in existence that are available and work significantly differently. Even if nobody bothers to fork the linux kernel, some people may go and work on the BSD kernels, for instance, because they prefer the design.
In essence, as long as enough people disagree about the best way to do something, there will be a fork. It happens with nearly every application available as much as, if not more than, it happens with kernels.
Desktops are a huge area of disagreement. The design of them is mostly about usability, and we're still in infancy when it comes to understanding the best ways to do things. ACM has only been running HCI conferences since the early 1980's, and since then researchers have figured out that designing good desktops is very difficult. Putting rules on it might make it slightly easier to be compatible with or learn, but placing draconian enforcement on a policy that isn't known to be good is more likely just to leave us with another crappy desktop.
Windows is a crappy desktop from a usability perspective. Personally I prefer to avoid KDE and Gnome, both of which seem to want to mimic Windows in most ways, including most of it's bad features. For a linux desktop I prefer WindowMaker, which also isn't perfect, but is has several features that I just like. Having the option to switch and still have all of my X applications work is fantastic.
Probably the same thing that would happen if someone hacked the railway signalling system to force a few trains to stop (or worse, crash). Some people may be killed or injured, lots of people get annoyed (many more people than a few sysadmins), the police chase them down, and they quite possibly get into a hell of a lot of trouble.
From the way you describe it, though, you also seem to be implying that this whole system may be somehow connected to the Internet. Bureaucratic decisions aside, there shouldn't be any need for the system to be connected to anything. The road indicates it's shape (edges, lanes, corners, turnoffs, etc), and the cars drive themselves on it.
Other ways that it might be broken could be to hack your car's driving system, or to simply stand near the road waving around some fake transmissions (depending on how the communications worked). My personal belief is that worrying about this type of thing happening is starting to get a bit too cautious. To me it doesn't seem like the sort of apparently harmless activity that most script kiddies would be interested in, aside from being a lot more cost and effort, and probably worse consequences.
Sure, the information provided by another car is untrusted. But so is just about everything about another car already that's usually taken for granted. When you're out driving on the road, for instance, you normally assume that the drivers around you are suitably trained to drive, that the bits of their vehicle aren't going to fall off, that their wheels have enough tread to properly grip the road, that the other drivers haven't been drinking, and so on.
The need to make sure that people and vehicles who drive on the road are using it safely is the reason that most governments operate systems that both require drivers to be licensed, and (in this case) require cars to be maintained to a certain standard and operate in particular ways. This is for the safety of everyone else as much as the safety of any particular vehicle's occupants. There shouldn't be much reason why such requirements can't be extended to the implementation and interface of an auto driving system that would work in this way. Probably people will hack them from time to time to provide false information to other parts of the system, and probably those people will get into an appropriate amount of trouble because of it.
In any case, I'm just brainstorming at the moment as I was when I wrote the previous comments. My personal belief is that untrusted information would not be a very serious problem under the circumstances.
Well I'm not a physics expert, but the thought that immediately comes to mind is that if an engine stalls, a vehicle still has momentum and won't stop suddenly unless the brakes are actually applied. Consequently, everything behind could still decelerate at the same rate.
But maybe you really meant what might happen if one vehicle simply has to stop suddenly by slamming on the brakes. You'd certainly hope that if one car did that, then all of the following cars would immediately do the same to avoid collision... almost certainly, all following cars could quite easily be stopping before their occupants realised there was a problem. If this were all done well, with proper inter-car communication, there doesn't seem to be much reason why it might not work.
Depending on the types of cars, the computer in each car would still need to allow for some gap in the following distance, based on what it knows about how much time it would take to stop, versus how much time the car in front would take to stop. It'd also need to take into account a required safe deceleration rate that wouldn't kill the people inside the car if it had to stop suddenly. (Irrespective of debates about whether governments should force people to wear seat belts, I still think it's stupid not to wear one.)
A lot of the distance that's presently needed for safe driving on roads is because of driver reaction time. Lots of drivers seem to have the impression that because they're driving a modern car, it's safer for them to drive faster and (to an extent) more recklessly. Unfortunately, as long as a human is controlling the car, you can't beat the human reaction time. It simply takes x fractions of a second for people to react, irrespective of how fast they think they're reacting, and irrespecting of how suddenly their vehicle can stop after they've hit the brakes. A computer controlled car would remove most of this reaction time, making it easier to reduce the safety gap between cars.
This is completely true. Another major advantage of autonomous cars is that they could simply follow more closely, letting more cars occupy the road at once, travelling at higher speeds, getting to where they're going more quickly and reducing much of the energy waste and pollution that's often associated with low speed stop-start driving.
With a road that's designed for it, as well as cars that are designed to communuicate with the road and the other cars on it, traffic congestion could be reduced hugely. As long as they have reliable data, computers are capable of reacting several orders of magnitude faster than humans are.
It might be substantially more difficult to implement a system like this for city driving, mostly due to uncontrollable parameters like pedestrians. But it doesn't seem that unreasonable to implement it on high speed roads. The main barriers are upgrading the roads to support the cars, and making the cars capable of driving on the roads. Perhaps, to do this, you might allocate a lane or two at a time for a while, as the infrastructure changes.
Good point -- that would be very amusing.
Part of me is inclined to agree, but then the other part of me realises that they've been doing this for more than 15 years. It's quite possibly a lot more of a tedious chore than it is interesting.
They've identified a figure for which they'll be satisfied to put up with the job in future times, no matter how boring or frustrating it gets. The alternative is that Fox may decide it's too much --- the show will end, and the voice actors can go and spend their time doing something that they find much more interesting. It's quite possible that they may be more than happy to go and do another show for much less, simply because it's different.
I suppose that just because the show is popular doesn't mean that the voice actors should be required to do it forever at a wage set by the studio. The raise is what they're claiming will be needed in order to keep them satisfied to continue doing it... perhaps they won't get it and they will no longer work on the show, but I'm sure they won't mind.
Well I tend to agree with you, although it should be acknokledged that the extra rights available with a diplomatic passport still only exist because the hosting country decides to grant them. Revoking the rights would probably mean breaking international agreements, though, which I guess is why it's never going to happen. (If you actually care about things enough to break major international agreements, chances are that things are already much more seriously messed up than what this particular activity would resemble.)
I tended to think the same for a while, but having considered it, I have to admit that most international US travellers who I've met here (I'm in New Zealand) have been much more enlightened about the rest of the world than the people who probably back this type of legislation in the US.
I don't know if it's a smart thing to punish them just because their government does something stupid, and I also don't want to contribute to discouraging US citizens from actually getting out and seeing the rest of the world.
On the other hand, I'd have absolutely no problem with requiring tedious entry procedures for US government officials. The thought of Bush and his entire travel comeraderie being required to have their photo and fingerprints taken before entering other countries just makes me laugh. It won't happen, of course -- the US just has far too much international influence.
Personally I would at least like to see all people who are closely connected with the current US federal administration (diplomats and presidents included) be forced to go through the process of having their photo and fingerprints taken whenever they try to travel to any other country.
I'm tempted to wish it for all US citizens, but I don't particularly want to sink to that level. To their credit, most Americans who actually travel internationally tend to be a bit more open minded (in my experience of meeting them so far) than the people who tend to support this.
It's an interesting analogy, but I'm not convinced that it describes things perfectly. Farmers who sell produce are in a different type of market from software. The former is very general with lots of small players. The latter is quite specialist, with several very big players who have a lot of influence as well as lots of smaller players who often have to follow.
Although farmers may often sell the same things in the same locations at the same time, they're not automatically in direct competition with each other. The simple reason is that on their own, one farmer can't supply enough goods to meet all of the demand. One farmer can almost never set the market price, and as long as s/he is selling at the market price a farmer is likely to sell all of the goods... with the occasional exception in different circumstances.
With the occasional exception, sales for one farmer aren't lost because another farmer has more to sell. And because there are so many small players in the market who are selling the same goods, any one farmer doesn't have much influence over the price, anyway. All the goods will probably be sold at the price set by the market. (Exceptions may be in small markets where there are only a couple of farmers and nobody else anywhere selling the same thing.)
It'd be funny to browse the logs if it did, but I'd hope that most people wouldn't do that as habit. It would pollute the statistics for website developers that describe how many people are using Mozilla/whatever, possibly making it appear less popular than it actually is -- at least to the less-educated.
If a simple count shows 85% IE and 12% Mozilla, it's much more likely that more emphasis will be placed on Mozilla support than if it showed 85% IE and 15% unidentifiable... in which case it's probably more likely for a typical admin to simply support IE and request that anyone who cares switches to it.
In either case, a good developer would hopefully at least code to standards wherever possible, but not all developers are perfect and depending on the application it's not always possible to do something using available published standards.
This whole argument that's tied in with many people's need to divide things into discrete groups, even if such groups don't exist, strikes me as more than a bit childish. We already know what it is, who cares how it's categorised? Let's just get on with learning more about it.
Richard Feynman had some great things to say about this on a BBC Horizon documentary some time ago. You can attach names to things as much as you like, but having a name for it doesn't compare with actually understanding it.
My parents have been using Netscape for browsing and email since they started using the Internet... I think it was a carryover from one of the ISP CD's some time ago, but they upgraded to Netscape 6 when it came out.
Recently, after my dad bought a new computer with XP pre-installed, he was even able to set up a dial-up networking connection. But since Netscape wasn't available he didn't know how to get to the web. (He couldn't actually find IE, even though it was certainly installed.) Anyway, I've now installed Mozilla for him, and he's really happy with it.
You can talk about whether Sedna should be called a planet or an asteroid for ever, but really it's just trivial. You could also ask why several other objects haven't been called planets, or why pluto has. The best answer you're ever likely to get is that changing things would be too much controversy to be bothered with, it would make lots of teaching material out of date, and it would start a slippery slope to make the names more ambiguous than they already are.
What's really interesting about Sedna is that nobody expected it to be there at all, and nobody yet has any well accepted theory about how it got there. It's further away than the expected range of the Kuiper belt, it's too close for the Oort cloud, and until now nearly everyone expected that to be mostly empty space... certainly nowhere that an object of Sedna's mass would be found. It also has a very eccentric orbit. Most astronomers out there are much more interested in this type of thing than trivial naming issues. It doesn't stop the media from trying to create a story out of nothing, however.
Well I'm an amateur astronomer and although I don't think that type of progress can be stopped, I do tend to sympathise.
Realistically if you think about the history of the western culture that, today, dominates space exploration, it's very likely to result in a lot of plundering and abuse for personal (or possibly corporate) gain. This will be thanks to the relative lawlessness on another planet before any serious long term colony effort comes along and will be required to pick up the pieces.
Historically this is just what happens. Even if you discount the American colonials wiping out the Indians, Pizarro destroying civilisations in South America for his personal fortune, as well as the Brits beating up nearly every tribal culture they came accross in their conquest of the world, there's still the big impact of environmental damage that's caused by colonials making dramatic changes such as burning down all the forests so they can overfarm the land.
I realise that a lot of this will be inevitable, and obviously Mars is a bit different. But despite it's immediate irrelevance to easy human colonisation, Mars does have it's own environment that's has all sorts of interesting things about it. I would, for once, like to see a new place being valued for its existing qualities when it's first properly visited, instead of them all being turned upside down for short term profits of a select few, only to have future generations kicking themselves at what's later discovered to have been lost.
I agree with you to an extent, although moreso for Postgres than MySQL, the latter of which is insulting and not worthy of being labelled a real database, imho.
I also think that this is exactly why open source is such a threat to the big products like Oracle and SQL Server. The big databases certainly do have a lot of features. Certainly they're capable of much more than open source products. But if you think about it really, how many of the users actualy use all of those features?
The places where OSS products can cut into the market are with all of those customers who have a big, expensive commercial database that they really don't need. If you only use your databases for inserting, updating, selecting, stored procedures and having some integrity built in, and of course if you have an admin with some idea of what they're doing, then something like postgresql may be perfectly reasonable for your needs. It's not a top level database but for what it does, it does well and just as reliably as anything else on the market. It may not be the best choice if you have extreme load conditions or whatnot, but a lot of commercial vendor's customers don't.
This is what the commmercial vendors really have to watch out for. Although most OSS databases don't offer the breadth of features, they are starting to be viable substitutes for the majority of database tasks. It's certainly possible that the commercial vendors might suddenly find a lot of their customers disappearing from underneath them.