I find it interesting how many people like having to compile everything and use a distro with such a archaic(not bad, just outdated compared to todays distros) install routine.
Archaic compared to what? Portage is infinitely better than RPM. And I'm not exaggerating... the Goodness of RPM is a negatively increasing value, while the Goodness of Portage is a positively increasing value. Portage is, while maybe not Heaven, certainly very nice. RPM is Hell.
In any case, you have a valid point, but about the wrong aspect of Gentoo. The initial installation is very primative, compared to... well... everything else. But the basic package manager is excellent. It has a superb and intelligent dependancy resolution mechanism, that supports complex version checking; it has a package specifications that are separate from the tarball distributions; it allows multiple versions of the same software to be installed at the same time; it really is quite sophisticated.
Read the article. The 1.4+ releases provide optional binary distributions, no compilation necessary. Personally, I don't think that this defeats Gentoo's purpose at all. What I like about it is that, in most cases, it lets you get Gentoo up and running fast. Then you can selectively "upgrade" -- compile ON your machine, optimized FOR your machine -- individual packages overnight. You'll probably do that over time anyway, with system upgrades.
Binary packages are even more necessary than Gentoo will admit, though. KDE can take a day to compile, Mozilla takes about the same, and I've heard horror stories about OpenOffice. On a laptop, compiling these packages can also present other problems; I've had some packages consume over 300MB of RAM during the compiling, which means that my laptop with its puny 256MB of RAM spends most of its time swapping. Mozilla during a build will happily eat 500MB of hard drive space. For massive applications like these, I'd rather take the tiny performance hit of having them compiled on another computer, and just download the binaries.
Mandrake (and other RPM-based distributions) are fine, as long as you never upgrade them. RPM-based systems can quickly become unmanageable, as I recently discovered with my laptop:-/
Have you truly never wasted a day screwing around with RPM? I spend half my time trying to resolve stupid dependancies that make no sense -- like starting with a PHP install for a headless web server and ending up needing to install X. That's not an actual example, BTW, but it is fairly typical.
RPM suffers from a (not insignificant) number of basic flaws, and most of these are not present in Gentoo's Portage (or Debian's dpkg, AFAIK).
I've been running Gentoo on a server for several months now (replaced Redhat), and it has been a pleasure to maintain. I'm not keen on some of the ways in which Gentoo lays out its file structure, but I can live with that. My laptop, which has been running Mandrake for about 18 months since I bought it, finally got to the point where I couldn't install anything anymore because of RPM. I was building and installing everything by hand, so I installed Gentoo over it. This was actually a lot easier than I thought it would be. Replacing Mandrake was painless; getting Gentoo configured, getting X running, and so on, was unneccessarily difficult. Gentoo has an extremely primative configuration system, which is to say, it doesn't have one. It uses the base package's configuration tools; for X, this is still xf86config. I mean, come on... when was the last time anybody on a modern operating system had to enter scan rates by hand? That is more XF86's fault than Gentoo's, but it does make Gentoo a pain to get running.
Basically, what I'm saying is that when a package manager is so bad that people have to avoid it to install software, as is the case with RPM, something else is needed.
Gentoo is relatively difficult to install, compared with other modern Linux distributions. However, once it is installed, it is fast and easy to maintain.
Please, don't bullshit. There are a lot of motivating factors for writing free software,
No bullshit. There is a conspicuous absence of a popular non-Linux Gnu distribution. This is the only proof I need to show that Linus, not RMS, was the motivating force behind Linux. Furthermore, I believe that even had Gnu never existed, Linux would still have come into being, and would have gained as much popularity as it now has. This is, obviously, pure conjecture on my part, but I think it is an opinion formed from a reasonable extrapolation of the history of Linux.
First, look up dogmatic and see if that is really what you mean.
dogÂma
n. pl. dogÂmas or dogÂmaÂta (-m-t)
A doctrine or a corpus of doctrines relating to matters such as morality and faith, set forth in an authoritative manner by a church.
An authoritative principle, belief, or statement of ideas or opinion, especially one considered to be absolutely true.
A principle or belief or a group of them
Yup. It means exactly what I think it means.
Linus tends to keep high politics (as opposed to interpersonal politics) out of what he does. RMS does just the opposite; from the GPL to his public statements, everything is shaded in politics, ethics, and dogma. This makes Linus easier to work with, because he isn't preaching at you. Again, you can agree with something, and still not want to have it shoved down your throat. I support PBS, but I treat their fundraisers just like any other telemarketter if I receive unsolicited calls.
As far as crediting contributors, I agree. The question is: at what point is a contribution considered significant enough to warrent high-level credit? Should we start calling it "Kerningham/Richie/Gnu/Linux"? K&R have made much more of a contribution to Linux than either Gnu or Linus. Maybe Gentoo should be called "Python/Gentoo", because so much of what defines Gentoo is Portage, which is a Python app.
This message was not run through a spell checker, including my own brain.
The GNU/Linux beef is one thing (language is inaccurate; koala bears aren't really bears), but calling Linus "the inventor of the Linux operating system" is about like crediting NT to the team that wrote KERNEL32.DLL.
No, calling Linus "the inventor of the Linux operating system" is like calling Bill Gates "the inventor of MS Windows." Linus probably contributed more code to his project than Bill did to his, but their most important contributions were similar in nature.
In both situations, it is less important who wrote the code than who was the central motivating factor in the creation of the phenomenon. Bill Gates did it through shrewd (amoral?) business practices; Linus did it by motivating people to contribute. However he did it, Linus was the binding glue that has propelled Linux to where it currently is.
As you said, most of the codebase is there. You imply that the kernel is almost a trivial aspect of the OS... so where is the Gnu OS that exists outside of Linux? Hurd is pretty old... why isn't anybody running it? Why is Linux more popular than BSD? Why is it more popular than any other well designed OS?
Linus motivates people; he's a cult of personality that encourages cooperation and construction. RMS's cult of personality is of a different, more evangelical sort, which hasn't, and doesn't, encourage people to join in. This is probably because RMS's motivations are morally dogmatic, where as Linus's are practical/hedonistic. You can be a communist and like Linux, but you don't have to.
So, I believe it is entirely reasonable (if not fair) to credit Linus with the creation of Linux. Even if he hadn't written a single line of code, and yet had otherwise done for Linux what he has, he'd still be able to claim that title. This all by the basic fact that without Linus, there would have been no Linux as we know it, no matter how much code Gnu churned out.
(Note: Chris does not advise you to follow in his footsteps.)
I don't believe that "warez" is an important enough issue to break the law over, I probably wouldn't morally approve of the activity if I thought about it enough, and I'm probably not clever enough (anymore) to crack software anyway.
However, one must wonder whether Chris' discouraging of people to follow in his footsteps is motivated by his inner feelings, or by the terms of his sentence / plea bargain / desire for early parole. The last, I can understand, for obvious reasons; the first two have always seemed just shy of legalized censorship.
To those who say this isn't news: it is. It may not be the first cordless optical mouse that comes with a recharging station, but it is the first from Logitech.
To those who say this is a useless product: beauty is in the eye of the beholder. Don't knock it until you try one of these mice. I don't think there are many people who would doubt the advantage of optical over mechanical mice. The wireless feature, however, is one of those "once you've had cordless, you won't go back" things. The recharging station is a nice feature, because without it, you're guaranteed to get annoyed for a couple of minutes at least once ever month or so when your mouse batteries run down. It also opens the possibility that Logitech has fixed a problem with their earlier versions of this mouse.
This article touches briefly on the sampling rate of this mouse, which was nice. When you go down and compare these optical, cordless mice at the store, you find that none of them have any hard specs about their sampling rates. I have one of the earlier Logitech cordless optical mice, and really liked it. However, it couldn't come close to the accuracy of a corded optical mouse. The optical sampling rate was low enough that the difference in accuracy was very noticable. This reduced sampling is done to improve the battery life. A low sampling rate is an issue that can make you hate your mouse.
With the inclusion of a recharging station, it is possible that Logitech modified the mouse to sample at a higher rate and use more energy, because battery life isn't as much of an issue any longer. The batteries are capable of driving the mouse over several days, even at high consumption rates, so as long as the user puts the mouse in the cradle overnight occasionally, the mouse can be less efficient and perform better. It is one of those myriad little user-friendliness things that can make or break a product's popularity.
Personally, I'm happy to see this mouse. Microsoft makes good ergonomic keyboards and crappy operating systems, but Logitech, IMHO, makes the best mice.
A basic understanding of inertia is needed here. When a plane loses an engine-- or all the engines, for that matter-- it doesn't just drop out of the sky.
I agree -- a basic understanding of inertia is needed.
There are many cases where inertia won't help you much. This plane, in all probability, will make a good glider. However, if you take an F15 and turn off the engines, it does effectively drop out of the sky, in the sense that the speed rapidly falls below the minimum speed required for its tiny wings to provide any lift.
There's another way inertia will not only not help, but will hurt. If this plane doesn't glide well, then the 20m fall isn't going to hurt as much as the 500 knot horizontal velocity will when it hits something.
I'd like to hear what the average rate of recovery for failed engines is, and how long the average restart time is. Not on this plane, in particular, but your average commercial airliner.
Care to give any specific examples as to how it does so with Lisp, Smalltalk or Forth?
I can't speak for Smalltalk or Forth, but Lisp (the original poster's example) -- which dates back to about 1975 -- was tied heavily to ITS on the PDP-10, which was "killed" by Unix. Unix was much more portable than ITS. While C didn't kill Lisp, it took most of the wind out of its sails, despite being less elegant -- Lisp is unarguably a second-class citizen in the modern computing world. C was also designed for portability.
So, I can see the correlation the original poster was making between Crush and Lisp.
I have a feeling that Crush will be a limitation to BRiX. By limiting the programming languages for BRiX, Brand is de facto alienating a huge group of people who probably won't like Crush for whatever reason. It is going to be difficult enough to gain sufficient momentum without limiting your contributor domain.
Does anybody have an idea of how they established the earnings? For example, did they count the money Mandrake received from their community dues? Did they count the money 'donated' to various projects, such as Gentoo? Did they only count the reported earnings of companies that are publically traded?
Armor piercing is great for some applications. Most anti-personell weapons, however, don't want armor piercing. An armor piercing round from even a moderate velocity weapon will go all the way through a human, doing relatively minimal damage. If you want to cause damage, what you want are soft bullets that expand when they hit soft stuff. If they expand, they do more damage, and cavitation effects are worse. This is why hollow points exist. This is why bullets are made of lead, not steel. You're better off if you're hit with an armor piercing round.
Nato 5.56mm rounds (M-16 rounds,.22 cal) are designed to tumble very early. They do a massive amount of damage for a small round, because they are designed to tumble very early upon hitting the body, split in two, and produce some massive cavitation. Despite this, the 5.56 doesn't kill as fast as the 7.62 used by the M1 Garand or the AK-47. This is on purpose. The US small arms tactic is to prefer wounding over killing. This isn't because of some noble humanitarian ideal; the military figured out that one wounded soldier takes three other soldiers out of combat just to take care of the 1 wounded person. Wounded people are much more expensive to take care of than dead people.
In any case, all other things being equal, you'd be better off getting hit by a Liquid Metal bullet than most of the other options. You'd have a better chance of surviving, if you got to a hospital.
There are as many different time proposals as there are people who believe that that we need something better. The only thing lacking is some agreement on how it should be done. If all of the metric/decimal time proponants could get together and agree on one standard, they'd be a sizeable group.
are a couple of good places to start. There are a significant number of opponents, as well.
In my experience, the difference in the standards are derived from one fundamental opinion: how much are you willing to deviate from how people use time to get universality?
Incidentally, one of the big problems with universal (using the term very loosely), decimal time (and the two do tend to go together) is that with universal time, it is very difficult for people to measure local time. What time do the stores open? What time is lunch? Using technology such as the Yes Watch solves this problem, and puts everything into perspective. The Yes Watch shows where you are in the day -- it is easy to see how far you are from sunrise or sunset; it is extremely convenient, and practical. If this technology were combined with a universal, decimal time format, universal time would become instantly practicaly. You look at the watch for your perspective on the day; you look at the decimal time to schedule your telephone conversation with someone in a different time zone. Very cool.
If you're interested in this, check out YesClock, a software implementation of the Yes Watch. It displays time using the same YesWatch paradigm, and has the option of displaying time in decimal format.
If you want a truely universal time measurement system, you won't get it, because of relativity. Theoretically, any time system we develop is going to fall to pieces fairly rapidly if we start travelling around at even moderately relativistic velocities. So you have to swallow that and pick the next best thing, say, measuring everything in seconds. Only, that sucks. Imagine trying to schedule a meeting for sometime next week. How many seconds is that from now? It doesn't scale very well, either. Plus, the fact of the matter is that people living on this planet live by a 24 hour day, and 'days' are a very useful concept. So you add the concept of a day. Now you're way off universality; as soon as we start living on Mars, the time system will have to be restructured. But you suck it up and continue. Years are really useful too -- most humans on the planet live by a cycle of seasons, and just counting 'days since the eruption of Mt. St. Helens' has scalability problems as well. So you add years -- heck, you're already dependant on the Earth by using days.
So you have years and days; when does a year start? When does a day start? Well, you could put the start of the day at the international date line, since the start point is arbitrary. But since the start point is completely arbitrary, why not start a day at a current standard start-point: UTC, AKA Greenwich Mean Time? It would make conversion easier.
How about the year start date? Do you make conversion easier, and start it with the definately Christian-centric BC/AD dividing line, or try for something less religiously tied, like the year of the eruption of some major volcano?
My point being, that when you really start discussing this issue, you start getting all of these questions that don't have clear answers, and you end up with purists, who support solutions which are -- inconvenient, at best, and a wide swath of people who compromise to some degree on all of the points. The end result is a huge number of proposals.
By the way, Swatch tried this with BMT. They even make metric time watches, that measure time in 1000ths of a day. Being arrogant, they put 000 at midnight, Buel Switzerland (the corporate home of Swatch), which was a mistake; they should have settled for either the IDL or GMT, both of which have reasonably strong arguments. They failed in not defining a metric date measurement to boot.
Personally, I measure all dates in Gregorian Year+Day (IE, today is *2002-186). It is a pretty good compromise.
Why is our Tux wearing a party hat?Tux is wearing his special outfit in honor of the first annual 'wIndependence Day' -- a special celebration which takes place this year on July 4th. wIndependence Day celebrates the burgeoning use of Linux on business, government, educational institution, and personal end-user computer desktops as an alternative to dependence on Microsoft's Windows operating system and associated products. Enter this year's essay contest for the best 500-1000 words on how you kicked the Windows habit before July 3rd at midnight. Awesome prizes are being provided from great companies including the leading desktop-oriented Linux distributions and productivity tools. Enter before the deadline to be eligible!
If it were illegal, the Govt. would have busted them.
Say what? They're an illegal monopoly, proven in court. Not only has the govornment not busted them, but it had them tried and convicted, and then backed off with a lame "remedy" that just ensures M.S. maintains its monopoly.
MS owns the current administration, bought and paid for in the last election.
Note that this doesn't prove that MS's tax evasion is illegal, it just proves that if it were illegal, the current administration wouldn't do anything about it.
Hopefully, us early adopters will drive that price down fairly quickly. I've been hearing numbers like under $1000 for the HT within a couple of years. If that's true, that's about the same as a decent bike.
14 mph is about a 4-minute mile, right? Isn't that pretty close to a flat-out sprint for the mortals among us? Would you really feel safe sprinting on a city sidewalk on a regular basis?
No. Would you ride a Segway at 14 MPH on a sidewalk? The device does have variable speed, you know.
People seem to be assuming that riders are going to be cruising at maximum speed on crowded sidewalks on these things. Impose sidewalk speed limits if it becomes a problem. Segways certainly appear to have better reaction times than other wheeled transport.
I'm disappointed at the "guilty until proven innocent" attitude of people here. Most of the statements against Segway in this discussion smack of fear mongering. Give Segway a chance to prove itself to be a hazard before you start accusing it of crimes.
If Dean Kamen had his way, most motorized traffic in cities would cease. The sidewalks would be the domain of pedestrians, and the streets would be filled with cyclists and segway riders.
My impression is that Kamen doesn't believe that most of the 4-wheel traffic in cities is neccessary. I believe what he's shooting for is a sort of utopia where cars are the domain of rural transportation, and most people have a personal transportation device and use mass transit for everything else. You aren't going to get rid of freight traffic, but isn't that already limited to the wee hours in places like New York?
Since it isn't likely that any city is going to eliminate most inner-city 4-wheel traffic, Kamen is trying to ensure that Segway will be a viable alternative to the car. The Segway is more appropriate on sidewalks with pedestrians than on the streets with hazards like cars and bicyclists.
Another difference is that when a skater has an accident, you usually end up with a 3-4 pound high-velocity missile that inflicts further collatoral damage. I suppose a segway could go on a rampage, but somehow I doubt it.
I had to restrain a lady who tried to bolt straight inside after I had finished. Maybe I shouldn't have...
Perhaps you remember the news item from last year about the elderly American tourists in Paris, in two of the tourists tried to avoid paying the toilet fee by the first holding the door open for the second. When the door closed on the second lady, the toilet went into it's wash cycle. The cleaning solution is a deep, rich blue.
I'm surprised this happens at private corporations as much as it does, but this is fairly typical of government organizations.
I work for a company that contracts with the government, and it has been my experience that software development in the government tends to suffer from some basic problems, especially in organizations which don't have a long history of software development:
Overdesign. Spending far too long designing systems.
No milestones. Setting due dates measured in years, rather than "partial solution" milestones measured in weeks.
Treating software development as a large, resource intensive project.
These problems stem, by and large, from the nature of government itself. First, the government has a habit of requiring detailed documentation and approval for the use of all resources, because of accountability. Therefore, things that could normally take a couple of weeks to implement can take a year. You have to write a proposal and submit that for approval. When that finally gets back to you, you have to design the system, and submit that for approval. When that finally gets back to you, you can begin development. If you require any changes in the system, or support from any other unit, you have to document that and submit that for approval. That will probably never get back to you.
While this happens to some degree in private industry, IME it is much worse in government.
A side effect of this is that, delays which are damaging in private industry, are deadly in government. You get administration changes, congressional budget changes, and so on, which can affect a project without the project even being a consideration. Normally, in industry, somebody's actually made a conscious decision that the project is unneccessary.
In private industry, you have organizations built around marketting. In the government, the organizations are built around beaurocracy. I think that software development in these organizations tends to reflect this.
I'm looking forward to WebDAV support in Konqueror, and (I haven't checked to see if/when this will be included) SVG support. And even more KIO plugins! camera://!! ssh://! dav://!! The whole world as an URL!
Once again, the wizard worked his magic, and there arose an extraordinary, modern IT operating system that evolved into the utterly rock-solid Windows 2000.
Rock solid from who's perspective? The only people I've heard using "rock solid" and "Windows 2000" in the same sentance that also doesn't include "not" are Windows 95 users who've upgraded. My wife, who uses Linux at home and W2K at work, still complains about W2K crashing.
I'm not sure the author was completely unbiased, although he did tend toward the center of the road.
Slashdot Mirror
Archaic compared to what? Portage is infinitely better than RPM. And I'm not exaggerating... the Goodness of RPM is a negatively increasing value, while the Goodness of Portage is a positively increasing value. Portage is, while maybe not Heaven, certainly very nice. RPM is Hell.
In any case, you have a valid point, but about the wrong aspect of Gentoo. The initial installation is very primative, compared to... well... everything else. But the basic package manager is excellent. It has a superb and intelligent dependancy resolution mechanism, that supports complex version checking; it has a package specifications that are separate from the tarball distributions; it allows multiple versions of the same software to be installed at the same time; it really is quite sophisticated.
Binary packages are even more necessary than Gentoo will admit, though. KDE can take a day to compile, Mozilla takes about the same, and I've heard horror stories about OpenOffice. On a laptop, compiling these packages can also present other problems; I've had some packages consume over 300MB of RAM during the compiling, which means that my laptop with its puny 256MB of RAM spends most of its time swapping. Mozilla during a build will happily eat 500MB of hard drive space. For massive applications like these, I'd rather take the tiny performance hit of having them compiled on another computer, and just download the binaries.
Have you truly never wasted a day screwing around with RPM? I spend half my time trying to resolve stupid dependancies that make no sense -- like starting with a PHP install for a headless web server and ending up needing to install X. That's not an actual example, BTW, but it is fairly typical.
RPM suffers from a (not insignificant) number of basic flaws, and most of these are not present in Gentoo's Portage (or Debian's dpkg, AFAIK).
I've been running Gentoo on a server for several months now (replaced Redhat), and it has been a pleasure to maintain. I'm not keen on some of the ways in which Gentoo lays out its file structure, but I can live with that. My laptop, which has been running Mandrake for about 18 months since I bought it, finally got to the point where I couldn't install anything anymore because of RPM. I was building and installing everything by hand, so I installed Gentoo over it. This was actually a lot easier than I thought it would be. Replacing Mandrake was painless; getting Gentoo configured, getting X running, and so on, was unneccessarily difficult. Gentoo has an extremely primative configuration system, which is to say, it doesn't have one. It uses the base package's configuration tools; for X, this is still xf86config. I mean, come on... when was the last time anybody on a modern operating system had to enter scan rates by hand? That is more XF86's fault than Gentoo's, but it does make Gentoo a pain to get running.
Basically, what I'm saying is that when a package manager is so bad that people have to avoid it to install software, as is the case with RPM, something else is needed.
Gentoo is relatively difficult to install, compared with other modern Linux distributions. However, once it is installed, it is fast and easy to maintain.
No bullshit. There is a conspicuous absence of a popular non-Linux Gnu distribution. This is the only proof I need to show that Linus, not RMS, was the motivating force behind Linux. Furthermore, I believe that even had Gnu never existed, Linux would still have come into being, and would have gained as much popularity as it now has. This is, obviously, pure conjecture on my part, but I think it is an opinion formed from a reasonable extrapolation of the history of Linux.
Yup. It means exactly what I think it means.
Linus tends to keep high politics (as opposed to interpersonal politics) out of what he does. RMS does just the opposite; from the GPL to his public statements, everything is shaded in politics, ethics, and dogma. This makes Linus easier to work with, because he isn't preaching at you. Again, you can agree with something, and still not want to have it shoved down your throat. I support PBS, but I treat their fundraisers just like any other telemarketter if I receive unsolicited calls.
As far as crediting contributors, I agree. The question is: at what point is a contribution considered significant enough to warrent high-level credit? Should we start calling it "Kerningham/Richie/Gnu/Linux"? K&R have made much more of a contribution to Linux than either Gnu or Linus. Maybe Gentoo should be called "Python/Gentoo", because so much of what defines Gentoo is Portage, which is a Python app.
This message was not run through a spell checker, including my own brain.
No, calling Linus "the inventor of the Linux operating system" is like calling Bill Gates "the inventor of MS Windows." Linus probably contributed more code to his project than Bill did to his, but their most important contributions were similar in nature.
In both situations, it is less important who wrote the code than who was the central motivating factor in the creation of the phenomenon. Bill Gates did it through shrewd (amoral?) business practices; Linus did it by motivating people to contribute. However he did it, Linus was the binding glue that has propelled Linux to where it currently is.
As you said, most of the codebase is there. You imply that the kernel is almost a trivial aspect of the OS... so where is the Gnu OS that exists outside of Linux? Hurd is pretty old... why isn't anybody running it? Why is Linux more popular than BSD? Why is it more popular than any other well designed OS?
Linus motivates people; he's a cult of personality that encourages cooperation and construction. RMS's cult of personality is of a different, more evangelical sort, which hasn't, and doesn't, encourage people to join in. This is probably because RMS's motivations are morally dogmatic, where as Linus's are practical/hedonistic. You can be a communist and like Linux, but you don't have to.
So, I believe it is entirely reasonable (if not fair) to credit Linus with the creation of Linux. Even if he hadn't written a single line of code, and yet had otherwise done for Linux what he has, he'd still be able to claim that title. This all by the basic fact that without Linus, there would have been no Linux as we know it, no matter how much code Gnu churned out.
I don't believe that "warez" is an important enough issue to break the law over, I probably wouldn't morally approve of the activity if I thought about it enough, and I'm probably not clever enough (anymore) to crack software anyway.
However, one must wonder whether Chris' discouraging of people to follow in his footsteps is motivated by his inner feelings, or by the terms of his sentence / plea bargain / desire for early parole. The last, I can understand, for obvious reasons; the first two have always seemed just shy of legalized censorship.
To those who say this isn't news: it is. It may not be the first cordless optical mouse that comes with a recharging station, but it is the first from Logitech.
To those who say this is a useless product: beauty is in the eye of the beholder. Don't knock it until you try one of these mice. I don't think there are many people who would doubt the advantage of optical over mechanical mice. The wireless feature, however, is one of those "once you've had cordless, you won't go back" things. The recharging station is a nice feature, because without it, you're guaranteed to get annoyed for a couple of minutes at least once ever month or so when your mouse batteries run down. It also opens the possibility that Logitech has fixed a problem with their earlier versions of this mouse.
This article touches briefly on the sampling rate of this mouse, which was nice. When you go down and compare these optical, cordless mice at the store, you find that none of them have any hard specs about their sampling rates. I have one of the earlier Logitech cordless optical mice, and really liked it. However, it couldn't come close to the accuracy of a corded optical mouse. The optical sampling rate was low enough that the difference in accuracy was very noticable. This reduced sampling is done to improve the battery life. A low sampling rate is an issue that can make you hate your mouse.
With the inclusion of a recharging station, it is possible that Logitech modified the mouse to sample at a higher rate and use more energy, because battery life isn't as much of an issue any longer. The batteries are capable of driving the mouse over several days, even at high consumption rates, so as long as the user puts the mouse in the cradle overnight occasionally, the mouse can be less efficient and perform better. It is one of those myriad little user-friendliness things that can make or break a product's popularity.
Personally, I'm happy to see this mouse. Microsoft makes good ergonomic keyboards and crappy operating systems, but Logitech, IMHO, makes the best mice.
Wow. You have talking carp?
I agree -- a basic understanding of inertia is needed.
There are many cases where inertia won't help you much. This plane, in all probability, will make a good glider. However, if you take an F15 and turn off the engines, it does effectively drop out of the sky, in the sense that the speed rapidly falls below the minimum speed required for its tiny wings to provide any lift.
There's another way inertia will not only not help, but will hurt. If this plane doesn't glide well, then the 20m fall isn't going to hurt as much as the 500 knot horizontal velocity will when it hits something.
I'd like to hear what the average rate of recovery for failed engines is, and how long the average restart time is. Not on this plane, in particular, but your average commercial airliner.
Mozilla segfaults on my machine, and OpenOffice doesn't. Maybe your hardware is buggy.
I can't speak for Smalltalk or Forth, but Lisp (the original poster's example) -- which dates back to about 1975 -- was tied heavily to ITS on the PDP-10, which was "killed" by Unix. Unix was much more portable than ITS. While C didn't kill Lisp, it took most of the wind out of its sails, despite being less elegant -- Lisp is unarguably a second-class citizen in the modern computing world. C was also designed for portability.
So, I can see the correlation the original poster was making between Crush and Lisp.
I have a feeling that Crush will be a limitation to BRiX. By limiting the programming languages for BRiX, Brand is de facto alienating a huge group of people who probably won't like Crush for whatever reason. It is going to be difficult enough to gain sufficient momentum without limiting your contributor domain.
Does anybody have an idea of how they established the earnings? For example, did they count the money Mandrake received from their community dues? Did they count the money 'donated' to various projects, such as Gentoo? Did they only count the reported earnings of companies that are publically traded?
Armor piercing is great for some applications. Most anti-personell weapons, however, don't want armor piercing. An armor piercing round from even a moderate velocity weapon will go all the way through a human, doing relatively minimal damage. If you want to cause damage, what you want are soft bullets that expand when they hit soft stuff. If they expand, they do more damage, and cavitation effects are worse. This is why hollow points exist. This is why bullets are made of lead, not steel. You're better off if you're hit with an armor piercing round.
Nato 5.56mm rounds (M-16 rounds, .22 cal) are designed to tumble very early. They do a massive amount of damage for a small round, because they are designed to tumble very early upon hitting the body, split in two, and produce some massive cavitation. Despite this, the 5.56 doesn't kill as fast as the 7.62 used by the M1 Garand or the AK-47. This is on purpose. The US small arms tactic is to prefer wounding over killing. This isn't because of some noble humanitarian ideal; the military figured out that one wounded soldier takes three other soldiers out of combat just to take care of the 1 wounded person. Wounded people are much more expensive to take care of than dead people.
In any case, all other things being equal, you'd be better off getting hit by a Liquid Metal bullet than most of the other options. You'd have a better chance of surviving, if you got to a hospital.
- DecimalTime.org
- Universal-Time.org (one of the oldest that I know of)
are a couple of good places to start. There are a significant number of opponents, as well.In my experience, the difference in the standards are derived from one fundamental opinion: how much are you willing to deviate from how people use time to get universality?
Incidentally, one of the big problems with universal (using the term very loosely), decimal time (and the two do tend to go together) is that with universal time, it is very difficult for people to measure local time. What time do the stores open? What time is lunch? Using technology such as the Yes Watch solves this problem, and puts everything into perspective. The Yes Watch shows where you are in the day -- it is easy to see how far you are from sunrise or sunset; it is extremely convenient, and practical. If this technology were combined with a universal, decimal time format, universal time would become instantly practicaly. You look at the watch for your perspective on the day; you look at the decimal time to schedule your telephone conversation with someone in a different time zone. Very cool.
If you're interested in this, check out YesClock, a software implementation of the Yes Watch. It displays time using the same YesWatch paradigm, and has the option of displaying time in decimal format.
If you want a truely universal time measurement system, you won't get it, because of relativity. Theoretically, any time system we develop is going to fall to pieces fairly rapidly if we start travelling around at even moderately relativistic velocities. So you have to swallow that and pick the next best thing, say, measuring everything in seconds. Only, that sucks. Imagine trying to schedule a meeting for sometime next week. How many seconds is that from now? It doesn't scale very well, either. Plus, the fact of the matter is that people living on this planet live by a 24 hour day, and 'days' are a very useful concept. So you add the concept of a day. Now you're way off universality; as soon as we start living on Mars, the time system will have to be restructured. But you suck it up and continue. Years are really useful too -- most humans on the planet live by a cycle of seasons, and just counting 'days since the eruption of Mt. St. Helens' has scalability problems as well. So you add years -- heck, you're already dependant on the Earth by using days.
So you have years and days; when does a year start? When does a day start? Well, you could put the start of the day at the international date line, since the start point is arbitrary. But since the start point is completely arbitrary, why not start a day at a current standard start-point: UTC, AKA Greenwich Mean Time? It would make conversion easier.
How about the year start date? Do you make conversion easier, and start it with the definately Christian-centric BC/AD dividing line, or try for something less religiously tied, like the year of the eruption of some major volcano?
My point being, that when you really start discussing this issue, you start getting all of these questions that don't have clear answers, and you end up with purists, who support solutions which are -- inconvenient, at best, and a wide swath of people who compromise to some degree on all of the points. The end result is a huge number of proposals.
By the way, Swatch tried this with BMT. They even make metric time watches, that measure time in 1000ths of a day. Being arrogant, they put 000 at midnight, Buel Switzerland (the corporate home of Swatch), which was a mistake; they should have settled for either the IDL or GMT, both of which have reasonably strong arguments. They failed in not defining a metric date measurement to boot.
Personally, I measure all dates in Gregorian Year+Day (IE, today is *2002-186). It is a pretty good compromise.
Say what? They're an illegal monopoly, proven in court. Not only has the govornment not busted them, but it had them tried and convicted, and then backed off with a lame "remedy" that just ensures M.S. maintains its monopoly.
MS owns the current administration, bought and paid for in the last election.
Note that this doesn't prove that MS's tax evasion is illegal, it just proves that if it were illegal, the current administration wouldn't do anything about it.
Hopefully, us early adopters will drive that price down fairly quickly. I've been hearing numbers like under $1000 for the HT within a couple of years. If that's true, that's about the same as a decent bike.
No. Would you ride a Segway at 14 MPH on a sidewalk? The device does have variable speed, you know.
People seem to be assuming that riders are going to be cruising at maximum speed on crowded sidewalks on these things. Impose sidewalk speed limits if it becomes a problem. Segways certainly appear to have better reaction times than other wheeled transport.
I'm disappointed at the "guilty until proven innocent" attitude of people here. Most of the statements against Segway in this discussion smack of fear mongering. Give Segway a chance to prove itself to be a hazard before you start accusing it of crimes.
If Dean Kamen had his way, most motorized traffic in cities would cease. The sidewalks would be the domain of pedestrians, and the streets would be filled with cyclists and segway riders.
My impression is that Kamen doesn't believe that most of the 4-wheel traffic in cities is neccessary. I believe what he's shooting for is a sort of utopia where cars are the domain of rural transportation, and most people have a personal transportation device and use mass transit for everything else. You aren't going to get rid of freight traffic, but isn't that already limited to the wee hours in places like New York?
Since it isn't likely that any city is going to eliminate most inner-city 4-wheel traffic, Kamen is trying to ensure that Segway will be a viable alternative to the car. The Segway is more appropriate on sidewalks with pedestrians than on the streets with hazards like cars and bicyclists.
Another difference is that when a skater has an accident, you usually end up with a 3-4 pound high-velocity missile that inflicts further collatoral damage. I suppose a segway could go on a rampage, but somehow I doubt it.
Or:
...everyone begins to look like data.
Perhaps you remember the news item from last year about the elderly American tourists in Paris, in two of the tourists tried to avoid paying the toilet fee by the first holding the door open for the second. When the door closed on the second lady, the toilet went into it's wash cycle. The cleaning solution is a deep, rich blue.
That's instant karma.
I work for a company that contracts with the government, and it has been my experience that software development in the government tends to suffer from some basic problems, especially in organizations which don't have a long history of software development:
- Overdesign. Spending far too long designing systems.
- No milestones. Setting due dates measured in years, rather than "partial solution" milestones measured in weeks.
- Treating software development as a large, resource intensive project.
These problems stem, by and large, from the nature of government itself. First, the government has a habit of requiring detailed documentation and approval for the use of all resources, because of accountability. Therefore, things that could normally take a couple of weeks to implement can take a year. You have to write a proposal and submit that for approval. When that finally gets back to you, you have to design the system, and submit that for approval. When that finally gets back to you, you can begin development. If you require any changes in the system, or support from any other unit, you have to document that and submit that for approval. That will probably never get back to you.While this happens to some degree in private industry, IME it is much worse in government.
A side effect of this is that, delays which are damaging in private industry, are deadly in government. You get administration changes, congressional budget changes, and so on, which can affect a project without the project even being a consideration. Normally, in industry, somebody's actually made a conscious decision that the project is unneccessary.
In private industry, you have organizations built around marketting. In the government, the organizations are built around beaurocracy. I think that software development in these organizations tends to reflect this.
I'm looking forward to WebDAV support in Konqueror, and (I haven't checked to see if/when this will be included) SVG support. And even more KIO plugins! camera://!! ssh://! dav://!! The whole world as an URL!
Mwahahahahahaaa!
Rock solid from who's perspective? The only people I've heard using "rock solid" and "Windows 2000" in the same sentance that also doesn't include "not" are Windows 95 users who've upgraded. My wife, who uses Linux at home and W2K at work, still complains about W2K crashing.
I'm not sure the author was completely unbiased, although he did tend toward the center of the road.