is because of East Timor, which is a massacre of civilians at the behest of the Indonesian government. Go look it up - you'll see why I mention it specifically.
I got an email also indicating that mosquitos aren't that bad around Orlando. I had spent some time in a more rural area in Florida, and assumed that the horrendous mosquito problem was statewide. Apparently, I erred in that projection.
However, Disney World has one heck of a lot of standing water. I can't shake my curiosity about the lack of insects so easily. Disney has been there a long time -- perhaps they are part of the reason why there aren't many mosquitoes?
I don't do much programming, so my opinion doesn't count for much -- but if it were my code, I'd put a specific exclusion in the license, specifically disallowing the Chinese government any rights to use my code. I'd probably do some research and also exclude some other governments that are abusive of human rights, like Indonesia.
Not like it would stop any of them from actually using it, mind you, but I would feel better about it.
I wonder if I would include the US Government in that list of abusive countries? Frankly, the fact that the thought even crosses my mind is a bit sad.:(
About two years ago, we went to Disney World. Had a great time; it was lots of fun. At the time, a very small thing impressed me more than anything else: during the week I was there, I saw precisely one mosquito.
So why was this so impressive to me? I suddenly realized, "Hey, this is Florida, otherwise known as the subtropics. Why aren't you being eaten alive by bugs?"
I think Disney must be putting an astonishing amount of poison into the environment down there to keep the bugs down. The amounts they must be using would have to be enormous -- Florida is a gigantic swamp. It's all standing water. How far out do they have to go from the park to keep mosquitoes away from the guests, and what methods do they use to control them?
I suspect that environmentalists are not going to like the answers too much.
IMO, the right to privacy includes the right to privacy in aggregate. In other words, since I personally have an expectation of privacy. I don't see that I give up that right by being a member of a group.
Consider the smallest possible group, 2 people. Should I be able to track all the purchases a group of two people makes? What is the fundamental difference between tracking a group of 2 people and of 20? Where is the line where it becomes okay to publish purchasing records? 200? 2 million? I don't think that line exists.
We have the right to privacy in aggregate. We should be insistent on it. Programs like Purchase Circles at Amazon should absolutely be opt-in, not opt-out.
It's nasty of Real to be collecting this data without prior permission. Claiming that it is somehow okay because the data was 'in aggregate' is specious and false.
I'll never use their software again, personally. Even Microsoft has more respect for privacy than that.
Please pardon the spelling, I'm too lazy to look up "opthamalogy".:)
According to him, you will generally get very good results from it. As others have said here, it does occasionally (very infrequently) foul your vision up terribly, so it is important to do it with a very experienced doctor. Damage done by the laser is permanent -- if your sight is impaired, chances are you won't ever see well again.
One interesting point that I haven't seen anyone else mention is that after the surgery is done, your eye structure becomes very slightly unstable. What this means is that your vision will change on a day-to-day basis... some days are better than others. The effect isn't enough to bother you too much -- he said it was just enough to notice. Even a bad day will still be better than the best days before.
He explained to me that the instability happens because some of the eye's supporting structures are weakened by reshaping them: they are, in essence, cutting out small parts of your eye to cause the remaining muscles to form the eye into a perfect sphere. (Most vision impairments come from an imperfectly-round eye.) Immediately after the surgery, there is an 'oscillating' effect, as the muscles quiver and move around their new point of equilibrium. It does fade, but it never goes away completely. Some people are more affected by this than others.
Two caveats here: I am not a doctor, this is just what a doctor told me. And I had this conversation about two and a half years ago. The technology and approach may be different now. I haven't heard this elsewhere, so I do take this with a grain of salt -- it made enough sense to me, though, that I thought it was worth repeating.
I know that there is a project afoot to rewrite most of the basic utilities in perl. This would, in effect, allow an instant port to NT. Unfortunately, I don't remember the name of the project. Perhaps some kind soul out there could enlighten us?
I'm glad you've had good luck with it. In my experience, I have lost data from Linux machines and not from NT ones. And I have A LOT more NT machines than Linux ones. At last count, I was directly responsible for about a hundred NT boxes, 6 Unix boxes of different flavors, and 3 Linux boxes.
2 of the 3 Linux boxes have been at least partially corrupted due to a power failure. I had to erase and manually copy in a bunch of files that had been crosslinked. The Debian box has never been the same -- I can't even compile kernels on it anymore. I'm sure I could fix the problem if I took the time, but I haven't so far. The Redhat box seems all right, finally -- but I was chasing down and reinstalling missing files for quite some time. (and yes, I know that RPM could have done that for me, but I didn't know how at the time, and it was faster to just replace the files. I haven't found a similar facility in dpkg.)
Of the remaining 106 or so machines, precisely 0 have have had glitches due to power problems (and yes, they have had exactly the same number of power interruptions.)
I did lose one (1) NT server when two hard drives died simultaneously (same production run... was kind of eerie), but I don't hold that against NT -- that kind of hardware failure requires lots of expensive redundant equipment to prevent data loss. I would have lost the whole Linux partition too.
In that same timeframe, we lost data on 2 out of 3 Linux machines after (different) power failures.
If data integrity is all you measure, Linux looks just horrible -- absolutely shoddy -- from my sample. This is why I don't trust it with the crucial files from my business. Amazing uptimes or not, UPS or not, ext2 is fragile. Accidents happen. The combination of these two facts forces me to hold off deploying it for file-sharing applications.
Paranoid? You bet. I have to be. That's why they pay me.
I'm really looking forward to XFS or ext3. I'd love to have Linux on every server. It would save me *so much* time and hassle.
But first, I'll be beating on that new filesystem, while standing there toggling the power off and on a few dozen times, to see if I can kill off any files.:-)
That's not NTFS. That's Exchange's problem. If it's one massive file, that's more like a filesystem on top of a filesystem.
You can install Exchange on a FAT volume just as easy, and I doubt it will make the least difference in the overall stability of the database.
And I didn't say never loses data. I have never lost a byte myself. I know, however, that NTFS is designed to protect the integrity of the filesystem in a power failure or hard crash. You can indeed lose data, but you probably won't lose anything but what you were actually working on. If you're saving something when you crash, you will probably lose that data, but you can be just about certain that you won't lose your whole filesystem.
This is a good thing -- NT crashes way too often. It needs the extra protection.:)
I'm surprised, but I'm forced to agree with you here.
Look, I love Linux to death. I think it's absolutely wonderful. But it has flaws. It is not perfect. NT is better than Linux in at least a couple of areas: its filesystem and its access control lists.
I have been running Linux and NT both for years. I can tell you that Linux loses data from power failures, and I fear it because of that.
It appears that this is where the reality distortion field is setting in. Instead of admitting that Linux is inferior to NT in this area (which it is), a lot of folks here are claiming that I need more hardware.
Wait, isn't that Microsoft's argument? Buy another XXXX megs of RAM and all will be well. Throw another hundred megahertz at the problem. Buy a UPS.
I would expect most reasonable software engineers to say, "Yeah, you're right, we're working on that" or maybe "Cripes, I haven't lost any data, you must be unlucky" (which one person did.) Instead, they are blaming me for not throwing hardware at a software solution.
Whoah. This is reality distortion worthy of Steve Jobs. Somehow it is my fault that ext2 is fragile? I don't think so.
Now, for you Linux Doubters(tm) who are reading this and inserting the occasional pointed flame: keep in mind that MOST of what people tell you about Linux is true. It really is amazing in many ways.
However, there are some things things Linux is not:
1. The second coming of the Messiah. 2. An enterprise-class OS. (yet..check again in a year) 3. Mom-ready. (see #2: this is coming.) 4. Easy to learn. This will never happen: there may be shells on top of Linux that are easy, but Linux itself will be unrelentingly hostile forever. If it weren't, it wouldn't be Unix anymore. Note that power = hostility: you can almost bring about world peace with the Unix command line, if you study it long enough. But it will take some time.:-)
I haven't seen this reality distortion field directly myself before, and it worries me. This kind of unclear thinking will harm Linux more than anything else. It both prevents problems from being recognized and fixed, and also does a great deal of propaganda damage. If we are promising that it is all things to all people, we are lying. People remember that stuff. In five years, when it has swallowed much of the functionality of the other mainstream OSes, what are you going to say when people refuse to use it because of their bad experiences two or three years before?
It's worth thinking about. Pushing it too strongly this early is dangerous. It's not finished yet, all hype to the contrary.
Sure, we can help the fat, bloated kid up. But it would be essentially stupid: as soon as our collective back was turned, there'd be a knife in it.
Microsoft didn't get to where they have by being nice. They play serious hardball. Either you are a customer or you are prey. They cooperate only when they must and destroy most of their partners. Examples: Citrix and Sybase.
They are *bullies*, and treating them any other way is simply asking to lose teeth. Or worse.
The lack of a journaling filesystem for Linux is a real problem. Honestly, this lack is the only thing keeping me from deploying this thing as a server where I work. We're all pretty jacked about Linux, on the whole, but we only use it for DNS services at the moment.
Why? Simply because I have lost data from several Linux machines due to power failures.
Linux is highly reliable, but ext2 is fragile as hell. The last time I complained about this, someone said, "Well, gee, I've never lost a byte of data, I just used [some disk utility] and restored the 16th copy of the superblock and all was well. You must be a moron."
Apparently that was a valid criticism from his standpoint, but I have only a vague understanding of the layout of an ext2 partition, and I don't think I should be forced to learn it as an emergency data-recovery measure. I'm pretty geeky, but I'm not THAT geeky, and I don't think I'm ever likely to be, either. I'm interested in using the system to do fun stuff, not mucking about with the actual physical layout of the filesystem. (!)
NT, while it bluescreens and behaves mysteriously a lot more often than I would like, doesn't lose data when it crashes, except possibly the very last bit that was being saved. In my opinion, this strength makes up for many of its other failings. It may not be the most reliable OS in the world. It has horrible security. Administration takes way too much time. It's a tool that Microsoft uses to try to own everything in my life. But by god, it doesn't lose data from simple stuff like power failures.
Every OS has good points, and I think Microsoft really nailed it with two of NT's features: NTFS and their ACL system. I really like both. If I could have ACLs and a real journaling filesystem under Linux, without having to spend more than a day or two setting it up, I'd be very happy to deploy it all over my office.
I don't LIKE NT much... but I trust its filesystem.
Welcome aboard at last. I thought you had given up on the idea -- you still had the little ?'s in all your posts.:-)
I think you'll find that Linux is enormously empowering and tons of fun -- but don't forget to pay occasional attention to the wife and kids.:-)
Every Linuxer starts in the same place. What you are doing now is starting on building your knowledge web about how computers work. In the beginning, it is very hard -- it's all uncharted territory that doesn't relate to anything you've ever done before. Macs actually encourage this kind of ignorance, so your Mac experience hasn't been much help in understanding how computers really work on the inside.
Linux is different -- everything is right there in front of you. If you're willing to dig, you can control anything you want to control, right down to the iron. However, that can be an enormous amount of work. Seemingly simple things -- like dialup networking -- can be really remote and mysterious when you are starting.
It gets a lot easier. As your knowledge web fills out, you will begin to understand how things connect. Each new problem will (generally) be easier than the one before because you will have more of an idea of how to approach it. And new facts will stick more easily because they will relate to other things you already know. And, you will be learning how things REALLY work, instead of trying to understand through a layer of abstraction sitting on top. Abstraction is fine for Eloi, but Morlocks need to know how it really works.:-)
I've been working with Linux for about five years, off and on, and I am still learning things like mad. I'm actually a fairly good sysadmin with Linux at this point, but there's still a HUGE amount I don't know. And I'm reasonably intelligent and do the sysadmin stuff for a living. If I don't know the whole thing, being a professional and all, you can hardly be expected to.... so don't be too mean with yourself when you are still confused and have been learning for a couple of years. I'm still confused after 5.:-)
There is at least one problem with this approach: bugfixes.
Consider the case of MFC42.DLL. There's a series of nasty Y2K date bugs in this DLL. If you are dynamically linked, your Y2K problems go away when Microsoft fixes them. If you are statically linked, you have to fix and recompile yourself -- and you may not even know that early versions of the DLL have these problems. Finding out by getting a panicked call from a user with a broken app is not a lot of fun.
Now, if you statically link against a perfect.DLL, you're set forever -- but Microsoft != perfection. And they often don't even admit to bugs, instead just quietly fixing them. This makes it tough to know when you should update your.DLLs and recompile for your users.
I don't think the.DLLs themselves are really the problem. I think the REAL problem is that Microsoft doesn't manage them very well.
That fellow who is trying hard to make personal flight vehicles (cars that fly) has done a lot of development of the Wankel engine, so all is not lost. Even if his idea fails (unknown at present), the improvements to the Wankel should be available to others to license... or possibly just use, if he didn't patent his improvements. (unlikely.)
Actually, the failure of DAT is probably more due to the record companies than Sony. They managed to get legislation requiring anti-piracy devices passed, along with (I think) a tax on the DAT media.
In essence, they killed off the technology to protect their monopoly a little longer. (It's going away anyway. MP3 and similar formats will destroy the record industry as it is today.)
At any rate, blaming DAT's failure on Sony is really a bit unfair. DAT was a great technology and would have boomed if left alone.
On NT, the drivers live on the print server. They are never downloaded to the client workstation. Print jobs are handed to the print spooler on the server 'as-is', using whatever internal format Windows uses for print jobs in progress. (this is documented but I am too lazy to look it up:) )
You are confused with 95, which downloads print drivers from an NT server IF the NT server is set up properly. 95 requires print drivers to be local. NT clients do not.
I don't know what 98 does. I've never tried to print with it on our network.
There's a lot of competition right now. Cable isn't the only way of getting high speed access, and panicking about monopolies this early is a bit foolish. Remember: the market is not 'cable modems'. The market is question is 'high speed bandwidth'.
Admittedly, cable companies are in a better position than are telephone companies, but they are rapidly pissing away their lead. While they have much better basic infrastructure for carrying large amounts of bandwidth, they are thinking cheap. A simple review of their membership agreement shows their real agenda: they want to sell you the service, but they don't want you to use it. Most activities that would use bandwidth in any sort of steady, consistent way are banned. Servers of any sort are not allowed on @home. You can't run a mail server OR a Quake server. If they catch you running one, they can shut you off, no questions asked, and no refunds.
Remember, their management is used to being 'a cable company'. They really have no clue what it means to be in the bandwidth market. Right now, their thinking appears to be, 'Okay, we'll sell you the bandwidth, but don't you dare use it." They are more in the position of the railroad barons of the 1800s, but few of them realize it.
Because of the noxious usage restrictions, I have avoided cablemodems, and have instead been signed up with Pacific Bell DSL since about February. Their AUP is very different: their whole attitude is different. The AUP is approximately: don't do illegal stuff.
The basic assumption is that you will USE the bandwidth you have. The rates are structured for that. I have 384K down/128K up guaranteed for $80/mo. (I pay for 5 IP addresses -- that costs $30/mo more.) That's a lot more than a cable modem, but note that I have no noxious restrictions on what I may run. I assume I could run an adult site if I wished, though of course doing that on a 128K upload restriction is silly.:)
I have had two major problem areas. My first was that, like clockwork, at 9PM most nights, my DSL would go offline. I ignored it for the first 2 or 3 weeks, because I figured they were doing maintenance -- because it was so regular, I assumed it was scheduled. However, when it went down and STAYED down for a day and I half, I called and squawked. They sent out a tech, who replaced the modem (I had an Alcatel A1000, rev A0: they replaced it with a B0) and all was well. I'm still not sure why it was always 9PM. Maybe someone was running a hair dryer close to the line or something.
The second problem was more recent. In the local area, the network was totally saturated. I have several coworkers also on PacBell DSL (at my recommendation primarily) and they were seeing the same thing. Our ping times had gone astronomical, and it was impossible to play action games online.
I finally got pissed and called and complained -- loudly -- and sent in traceroutes showing 300ms round trips starting 2 hops away. I didn't hear anything more from them -- but 2 weeks later the problems went away, and I'm back to about 60ms round trip times almost anyplace. I assume they upgraded an overloaded router. I wish they had called me, but at least they fixed it.
Overall, in my opinion, it's decent service. I don't expect 100% perfection at these rates, and I haven't gotten it either.:-)
What I really like is that, while they guarantee only 384K down, I actually get close to 1.5Mbps down. My upload rate is definitely hard capped at 128K, however. If the service came in a 384/384 speed I'd be fat and happy. 128K up is a bit skinny, but it works okay. I would rather have a capped, reliable rate than a fast, unreliable one.
PacBell understands what it means to be in the bandwidth business. If you know what you are doing, you can call their technical center and deal with the same people you'd work with if you had a T1. If you're a networking professional, I suggest you consider them. If they realize you know what you're talking about, they'll work with you on a professional level.
For all these reasons, there's not really any reason to panic yet about cable modems being run by just a few companies, or even just one. Again, the market is 'bandwidth', not 'cable modems'. The phone companies are behind but not THAT much. Their better understanding of the bandwidth business gives them a real service advantage, at least in many cases.
The time to worry is when you see a single company owning large fractions of more than one competing type of service. The fact that AT&T is selling DSL now actually worries me. I do not believe AT&T is likely to be very competitive with itself, and I know they own a cable company. THAT, I think, is a problem.
In comparison, @Home buying Excite is nearly irrelevant, IMO.
Cripes, Solaris has been in development for how many years now? With how many developers?
Linux is still basically a PC program that has been ported. PC hardware isn't designed to take this kind of load. The hardware itself can't keep up with it. Linux is not *designed* to be a massive server. It's meant to be a desktop for serious technoid users who like technical excellence in their home machines. It happens to make a good server because it's good software, not because it's really designed to be one.
The fact that it trounces NT so thoroughly isn't really an endorsement of Linux, it's simply stinging criticism of NT. Linux wasn't written to be a server: the fact that it is a better 'server' than NT (which IS meant to be a server) simply shows how low the MS standards have gotten.
The underlying technology is being worked on. Each year that goes by will see it get stronger and stronger. Someday it may run Ebay and IRC servers... but then again, it might not. And that is okay. Ebay is gravy.
There is a *reason* Solaris boxes cost $50K plus. And no, the reason isn't (just) that they are greedy. The hardware is designed for an entirely different scale of problem. Sun knows how to make servers, and how to approach the problem from the bottom up. Linux is an accidental server, pressed into use because it happens to be better than the only commercial alternative that most people can afford.
You might as well bash the VW Bug because it isn't as fast as a Ferrari and doesn't pull the load a Mack truck would.
Argh.
well, if 80% of the users are good...
on
Moderation Ideas
·
· Score: 2
why not just assume that the 80% good people will outweigh/undo the effects of the 20% bad?
Admittedly, 5 moderator points/day helps limit the damage that bad moderators can cause. However, it also limits the benefit that good moderators can bring. I don't think it fundamentally changes anything. It is sort of like a minimizing lens; the overall ratio is the same, but the effect is reduced.
I don't remember why the moderator point limit was implemented. It's possible it may have been done in reaction to abuse that was not being corrected. If that's the case, disregard this idea:
Why not try a week or two with UNLIMITED moderator points? That is, if you are selected as a moderator, you have no limit on how many points you use. Keep the number of moderators about the same, just give them unlimited, tyrannical power.:-) But track M2 points carefully -- if people get more than a few negative M2 points, sharply decrease their frequency of moderation.
I think what will happen is that yes, there will be some abuse, but that good moderators will correct that abuse, and that the meta-moderation will (after awhile) shut down the abusers.
If that's no good, your idea of 'a chance' to moderate any given comment is a possibility -- I'm thinking a 5% chance per positive karma point. That way, 20 karma means you can moderate any post. Most likely, anyone who has 20 karma is going to have a clue.:-)
I have meta-moderated a few times, and I don't think I have chosen 'unfair' once. In some cases I probably would have chosen differently, but I have not seen ANY examples of grievous boneheadedness. (I don't mark down moderators for making different choices than I would, as long as their choice seems at least faintly reasonable.)
It also seems like the slant on pro-open-source and pro-Linux tends to artificially inflate some posts. I have written posts pointing out limitations of open source (most recently in regard to open docs), and also have written ones in favor of it where I think it really helps. Even though (in my opinion), the open-docs-won't-happen argument was better thought out than most of what I've written, it got no positive moderation at all. Other posts I have put up have gotten a lot of positive moderation, even though they weren't nearly as good, IMO. Maybe I'm too close to the problem, but personally I thought the negative-slant article should have scored higher.
Strikes me as a good idea to remind moderators to be aware of their own biases. I try to correct for mine when I'm moderating. I'm not sure how well I succeed, mind you, but I do try.:-)
It strikes me that documentation will probably never be free in the same way that source code is. You'll probably be able to get basic docs (a la manpages) gratis, but no matter how much RMS says free docs ought to happen, I just don't think they will.
There's little benefit in releasing docs open source. If I document a program and release the documentation, patches to the docs will enhance them for OTHER readers but not for ME. I already know the program -- that's why I documented it.:-) That's not at all like code -- if I release code and get patches back, my program improves and I probably also learn a thing or three.
And I definitely lose $$$ by not selling the documentation. Docs are NOT fun to write. It's not like being an artist/programmer. Really good documentation hides the author as much as possible. It's an exercise in negative ego gratification. And to write really good docs, you have to go OVER and OVER and OVER them... very tedious work. To get talented people to do this sort of work, you most often have to pay them. It's not like coding at all -- there's not much of a rush in completing a chapter.
Altruism does exist, but I don't think it's nearly as powerful as some folks would like to think. Most individuals will behave in a way that they perceive benefits them the most. Some have longer sight than others, of course.... but even the longest view doesn't really reveal THAT MUCH benefit to really good, free documentation *to the writer of the docs*. If someone uses my program, they will tend to want me to maintain and improve it, and chances are pretty good they'll pay me to do it. If someone uses my docs, they're not locked into my way of writing in nearly the same way they are with code. Giving away free code is a bit like Netscape giving away free browsers... it resulted in a great deal of revenue. Giving away free docs is more like, say, a search engine... it can be profitable but doesn't have the loyalty. Switching code bases is painful; switching docs (or search engines) is instant and painless.
Competing with that is the simple fact that there's a lot of immediate benefit to be gained by selling those same docs instead.
For these reasons, I simply don't think open docs will ever achieve the same status that open source will. Before very much longer, open source will be the way most development is done. The advantages are enormous. That is not true with docs; the present capitalistic method will likely continue indefinitely.
Caveat: I didn't read 3001. I read the first chapter in a preview -- it sounded hopelessly dull.
But what angers me the most about the poster I'm responding to is his talk about `anti-inquisition forces'. I read 3001. The book was not only anti-Christian, anti-Jewish, anti-Islamic, anti-anything to do with religion (check out the amazon reviews!).
I wasn't saying Clarke is anti-inquisition, although I'm sure he would have to be. He obviously feared religion a good deal, and there are vast, vast reams of data to support his view. However, religion certainly isn't the only source of evil in the world: government has done its fair share.
I think most Western religions encourage one not to think and seek answers for oneself, but to blindly accept canned answers. That is dangerous. Nazi Germany did the same thing. It's dangerous no matter who uses it. Telling people not to think, to shut up and go along with the program, results in pain and violence almost 100% of the time. But the religions Clarke feared mostly have this as a fundamental tenet of the religion.. (Judaism is at least a little different than this, I understand, but I am not really that familiar with it.)
But then you go into how fraternities justify dirty deeds, and I don't get where you're coming from. What fraternities? What dirty deeds?
And then you go really wacked out and say that Katz' motivation of protecting us from the inquisition is a bad one.
Say what? Where the hell did you get that? Katz was reviewing a book, remember? I guess you swallowed the assertion that he's a Mason hook, line, and sinker. And where 'enemies of Jews' came from is entirely beyond me.
I agree with you that anti-inquisition probably doesn't belong here, and judging from the overall quality of this thread, I believe this will be my last post on it.
Slashdot Mirror
is because of East Timor, which is a massacre of civilians at the behest of the Indonesian government. Go look it up - you'll see why I mention it specifically.
Hey, that wasn't a troll, that was an honest opinion. That was an unfair moderation.
Heh, shouldn't that be a mosquito line? :)
I got an email also indicating that mosquitos aren't that bad around Orlando. I had spent some time in a more rural area in Florida, and assumed that the horrendous mosquito problem was statewide. Apparently, I erred in that projection.
However, Disney World has one heck of a lot of standing water. I can't shake my curiosity about the lack of insects so easily. Disney has been there a long time -- perhaps they are part of the reason why there aren't many mosquitoes?
Just a thought.
I don't do much programming, so my opinion doesn't count for much -- but if it were my code, I'd put a specific exclusion in the license, specifically disallowing the Chinese government any rights to use my code. I'd probably do some research and also exclude some other governments that are abusive of human rights, like Indonesia.
:(
Not like it would stop any of them from actually using it, mind you, but I would feel better about it.
I wonder if I would include the US Government in that list of abusive countries? Frankly, the fact that the thought even crosses my mind is a bit sad.
About two years ago, we went to Disney World. Had a great time; it was lots of fun. At the time, a very small thing impressed me more than anything else: during the week I was there, I saw precisely one mosquito.
So why was this so impressive to me? I suddenly realized, "Hey, this is Florida, otherwise known as the subtropics. Why aren't you being eaten alive by bugs?"
I think Disney must be putting an astonishing amount of poison into the environment down there to keep the bugs down. The amounts they must be using would have to be enormous -- Florida is a gigantic swamp. It's all standing water. How far out do they have to go from the park to keep mosquitoes away from the guests, and what methods do they use to control them?
I suspect that environmentalists are not going to like the answers too much.
Wow... great comments. I hope you are moderated up. Your link re:work addiction was really interesting.
:)
Thanks for posting
It's why I don't shop at Amazon anymore.
IMO, the right to privacy includes the right to privacy in aggregate. In other words, since I personally have an expectation of privacy. I don't see that I give up that right by being a member of a group.
Consider the smallest possible group, 2 people. Should I be able to track all the purchases a group of two people makes? What is the fundamental difference between tracking a group of 2 people and of 20? Where is the line where it becomes okay to publish purchasing records? 200? 2 million? I don't think that line exists.
We have the right to privacy in aggregate. We should be insistent on it. Programs like Purchase Circles at Amazon should absolutely be opt-in, not opt-out.
It's nasty of Real to be collecting this data without prior permission. Claiming that it is somehow okay because the data was 'in aggregate' is specious and false.
I'll never use their software again, personally. Even Microsoft has more respect for privacy than that.
Please pardon the spelling, I'm too lazy to look up "opthamalogy". :)
According to him, you will generally get very good results from it. As others have said here, it does occasionally (very infrequently) foul your vision up terribly, so it is important to do it with a very experienced doctor. Damage done by the laser is permanent -- if your sight is impaired, chances are you won't ever see well again.
One interesting point that I haven't seen anyone else mention is that after the surgery is done, your eye structure becomes very slightly unstable. What this means is that your vision will change on a day-to-day basis... some days are better than others. The effect isn't enough to bother you too much -- he said it was just enough to notice. Even a bad day will still be better than the best days before.
He explained to me that the instability happens because some of the eye's supporting structures are weakened by reshaping them: they are, in essence, cutting out small parts of your eye to cause the remaining muscles to form the eye into a perfect sphere. (Most vision impairments come from an imperfectly-round eye.) Immediately after the surgery, there is an 'oscillating' effect, as the muscles quiver and move around their new point of equilibrium. It does fade, but it never goes away completely. Some people are more affected by this than others.
Two caveats here: I am not a doctor, this is just what a doctor told me. And I had this conversation about two and a half years ago. The technology and approach may be different now. I haven't heard this elsewhere, so I do take this with a grain of salt -- it made enough sense to me, though, that I thought it was worth repeating.
Good luck!
I know that there is a project afoot to rewrite most of the basic utilities in perl. This would, in effect, allow an instant port to NT. Unfortunately, I don't remember the name of the project. Perhaps some kind soul out there could enlighten us?
I'm glad you've had good luck with it. In my experience, I have lost data from Linux machines and not from NT ones. And I have A LOT more NT machines than Linux ones. At last count, I was directly responsible for about a hundred NT boxes, 6 Unix boxes of different flavors, and 3 Linux boxes.
:-)
2 of the 3 Linux boxes have been at least partially corrupted due to a power failure. I had to erase and manually copy in a bunch of files that had been crosslinked. The Debian box has never been the same -- I can't even compile kernels on it anymore. I'm sure I could fix the problem if I took the time, but I haven't so far. The Redhat box seems all right, finally -- but I was chasing down and reinstalling missing files for quite some time. (and yes, I know that RPM could have done that for me, but I didn't know how at the time, and it was faster to just replace the files. I haven't found a similar facility in dpkg.)
Of the remaining 106 or so machines, precisely 0 have have had glitches due to power problems (and yes, they have had exactly the same number of power interruptions.)
I did lose one (1) NT server when two hard drives died simultaneously (same production run... was kind of eerie), but I don't hold that against NT -- that kind of hardware failure requires lots of expensive redundant equipment to prevent data loss. I would have lost the whole Linux partition too.
In that same timeframe, we lost data on 2 out of 3 Linux machines after (different) power failures.
If data integrity is all you measure, Linux looks just horrible -- absolutely shoddy -- from my sample. This is why I don't trust it with the crucial files from my business. Amazing uptimes or not, UPS or not, ext2 is fragile. Accidents happen. The combination of these two facts forces me to hold off deploying it for file-sharing applications.
Paranoid? You bet. I have to be. That's why they pay me.
I'm really looking forward to XFS or ext3. I'd love to have Linux on every server. It would save me *so much* time and hassle.
But first, I'll be beating on that new filesystem, while standing there toggling the power off and on a few dozen times, to see if I can kill off any files.
That's not NTFS. That's Exchange's problem. If it's one massive file, that's more like a filesystem on top of a filesystem.
:)
You can install Exchange on a FAT volume just as easy, and I doubt it will make the least difference in the overall stability of the database.
And I didn't say never loses data. I have never lost a byte myself. I know, however, that NTFS is designed to protect the integrity of the filesystem in a power failure or hard crash. You can indeed lose data, but you probably won't lose anything but what you were actually working on. If you're saving something when you crash, you will probably lose that data, but you can be just about certain that you won't lose your whole filesystem.
This is a good thing -- NT crashes way too often. It needs the extra protection.
Look, I love Linux to death. I think it's absolutely wonderful. But it has flaws. It is not perfect. NT is better than Linux in at least a couple of areas: its filesystem and its access control lists.
I have been running Linux and NT both for years. I can tell you that Linux loses data from power failures, and I fear it because of that.
It appears that this is where the reality distortion field is setting in. Instead of admitting that Linux is inferior to NT in this area (which it is), a lot of folks here are claiming that I need more hardware.
Wait, isn't that Microsoft's argument? Buy another XXXX megs of RAM and all will be well. Throw another hundred megahertz at the problem. Buy a UPS.
I would expect most reasonable software engineers to say, "Yeah, you're right, we're working on that" or maybe "Cripes, I haven't lost any data, you must be unlucky" (which one person did.) Instead, they are blaming me for not throwing hardware at a software solution.
Whoah. This is reality distortion worthy of Steve Jobs. Somehow it is my fault that ext2 is fragile? I don't think so.
Now, for you Linux Doubters(tm) who are reading this and inserting the occasional pointed flame: keep in mind that MOST of what people tell you about Linux is true. It really is amazing in many ways.
However, there are some things things Linux is not:
I haven't seen this reality distortion field directly myself before, and it worries me. This kind of unclear thinking will harm Linux more than anything else. It both prevents problems from being recognized and fixed, and also does a great deal of propaganda damage. If we are promising that it is all things to all people, we are lying. People remember that stuff. In five years, when it has swallowed much of the functionality of the other mainstream OSes, what are you going to say when people refuse to use it because of their bad experiences two or three years before?
It's worth thinking about. Pushing it too strongly this early is dangerous. It's not finished yet, all hype to the contrary.
Sure, we can help the fat, bloated kid up. But it would be essentially stupid: as soon as our collective back was turned, there'd be a knife in it.
Microsoft didn't get to where they have by being nice. They play serious hardball. Either you are a customer or you are prey. They cooperate only when they must and destroy most of their partners. Examples: Citrix and Sybase.
They are *bullies*, and treating them any other way is simply asking to lose teeth. Or worse.
The lack of a journaling filesystem for Linux is a real problem. Honestly, this lack is the only thing keeping me from deploying this thing as a server where I work. We're all pretty jacked about Linux, on the whole, but we only use it for DNS services at the moment.
Why? Simply because I have lost data from several Linux machines due to power failures.
Linux is highly reliable, but ext2 is fragile as hell. The last time I complained about this, someone said, "Well, gee, I've never lost a byte of data, I just used [some disk utility] and restored the 16th copy of the superblock and all was well. You must be a moron."
Apparently that was a valid criticism from his standpoint, but I have only a vague understanding of the layout of an ext2 partition, and I don't think I should be forced to learn it as an emergency data-recovery measure. I'm pretty geeky, but I'm not THAT geeky, and I don't think I'm ever likely to be, either. I'm interested in using the system to do fun stuff, not mucking about with the actual physical layout of the filesystem. (!)
NT, while it bluescreens and behaves mysteriously a lot more often than I would like, doesn't lose data when it crashes, except possibly the very last bit that was being saved. In my opinion, this strength makes up for many of its other failings. It may not be the most reliable OS in the world. It has horrible security. Administration takes way too much time. It's a tool that Microsoft uses to try to own everything in my life. But by god, it doesn't lose data from simple stuff like power failures.
Every OS has good points, and I think Microsoft really nailed it with two of NT's features: NTFS and their ACL system. I really like both. If I could have ACLs and a real journaling filesystem under Linux, without having to spend more than a day or two setting it up, I'd be very happy to deploy it all over my office.
I don't LIKE NT much... but I trust its filesystem.
Welcome aboard at last. I thought you had given up on the idea -- you still had the little ?'s in all your posts. :-)
:-)
:-)
:-)
I think you'll find that Linux is enormously empowering and tons of fun -- but don't forget to pay occasional attention to the wife and kids.
Every Linuxer starts in the same place. What you are doing now is starting on building your knowledge web about how computers work. In the beginning, it is very hard -- it's all uncharted territory that doesn't relate to anything you've ever done before. Macs actually encourage this kind of ignorance, so your Mac experience hasn't been much help in understanding how computers really work on the inside.
Linux is different -- everything is right there in front of you. If you're willing to dig, you can control anything you want to control, right down to the iron. However, that can be an enormous amount of work. Seemingly simple things -- like dialup networking -- can be really remote and mysterious when you are starting.
It gets a lot easier. As your knowledge web fills out, you will begin to understand how things connect. Each new problem will (generally) be easier than the one before because you will have more of an idea of how to approach it. And new facts will stick more easily because they will relate to other things you already know. And, you will be learning how things REALLY work, instead of trying to understand through a layer of abstraction sitting on top. Abstraction is fine for Eloi, but Morlocks need to know how it really works.
I've been working with Linux for about five years, off and on, and I am still learning things like mad. I'm actually a fairly good sysadmin with Linux at this point, but there's still a HUGE amount I don't know. And I'm reasonably intelligent and do the sysadmin stuff for a living. If I don't know the whole thing, being a professional and all, you can hardly be expected to.... so don't be too mean with yourself when you are still confused and have been learning for a couple of years. I'm still confused after 5.
There is at least one problem with this approach: bugfixes.
.DLL, you're set forever -- but Microsoft != perfection. And they often don't even admit to bugs, instead just quietly fixing them. This makes it tough to know when you should update your .DLLs and recompile for your users.
.DLLs themselves are really the problem. I think the REAL problem is that Microsoft doesn't manage them very well.
Consider the case of MFC42.DLL. There's a series of nasty Y2K date bugs in this DLL. If you are dynamically linked, your Y2K problems go away when Microsoft fixes them. If you are statically linked, you have to fix and recompile yourself -- and you may not even know that early versions of the DLL have these problems. Finding out by getting a panicked call from a user with a broken app is not a lot of fun.
Now, if you statically link against a perfect
I don't think the
That fellow who is trying hard to make personal flight vehicles (cars that fly) has done a lot of development of the Wankel engine, so all is not lost. Even if his idea fails (unknown at present), the improvements to the Wankel should be available to others to license... or possibly just use, if he didn't patent his improvements. (unlikely.)
Actually, the failure of DAT is probably more due to the record companies than Sony. They managed to get legislation requiring anti-piracy devices passed, along with (I think) a tax on the DAT media.
In essence, they killed off the technology to protect their monopoly a little longer. (It's going away anyway. MP3 and similar formats will destroy the record industry as it is today.)
At any rate, blaming DAT's failure on Sony is really a bit unfair. DAT was a great technology and would have boomed if left alone.
On NT, the drivers live on the print server. They are never downloaded to the client workstation. Print jobs are handed to the print spooler on the server 'as-is', using whatever internal format Windows uses for print jobs in progress. (this is documented but I am too lazy to look it up :) )
You are confused with 95, which downloads print drivers from an NT server IF the NT server is set up properly. 95 requires print drivers to be local. NT clients do not.
I don't know what 98 does. I've never tried to print with it on our network.
There's a lot of competition right now. Cable isn't the only way of getting high speed access, and panicking about monopolies this early is a bit foolish. Remember: the market is not 'cable modems'. The market is question is 'high speed bandwidth'.
:)
:-)
Admittedly, cable companies are in a better position than are telephone companies, but they are rapidly pissing away their lead. While they have much better basic infrastructure for carrying large amounts of bandwidth, they are thinking cheap. A simple review of their membership agreement shows their real agenda: they want to sell you the service, but they don't want you to use it. Most activities that would use bandwidth in any sort of steady, consistent way are banned. Servers of any sort are not allowed on @home. You can't run a mail server OR a Quake server. If they catch you running one, they can shut you off, no questions asked, and no refunds.
Remember, their management is used to being 'a cable company'. They really have no clue what it means to be in the bandwidth market. Right now, their thinking appears to be, 'Okay, we'll sell you the bandwidth, but don't you dare use it." They are more in the position of the railroad barons of the 1800s, but few of them realize it.
Because of the noxious usage restrictions, I have avoided cablemodems, and have instead been signed up with Pacific Bell DSL since about February. Their AUP is very different: their whole attitude is different. The AUP is approximately: don't do illegal stuff.
The basic assumption is that you will USE the bandwidth you have. The rates are structured for that. I have 384K down/128K up guaranteed for $80/mo. (I pay for 5 IP addresses -- that costs $30/mo more.) That's a lot more than a cable modem, but note that I have no noxious restrictions on what I may run. I assume I could run an adult site if I wished, though of course doing that on a 128K upload restriction is silly.
I have had two major problem areas. My first was that, like clockwork, at 9PM most nights, my DSL would go offline. I ignored it for the first 2 or 3 weeks, because I figured they were doing maintenance -- because it was so regular, I assumed it was scheduled. However, when it went down and STAYED down for a day and I half, I called and squawked. They sent out a tech, who replaced the modem (I had an Alcatel A1000, rev A0: they replaced it with a B0) and all was well. I'm still not sure why it was always 9PM. Maybe someone was running a hair dryer close to the line or something.
The second problem was more recent. In the local area, the network was totally saturated. I have several coworkers also on PacBell DSL (at my recommendation primarily) and they were seeing the same thing. Our ping times had gone astronomical, and it was impossible to play action games online.
I finally got pissed and called and complained -- loudly -- and sent in traceroutes showing 300ms round trips starting 2 hops away. I didn't hear anything more from them -- but 2 weeks later the problems went away, and I'm back to about 60ms round trip times almost anyplace. I assume they upgraded an overloaded router. I wish they had called me, but at least they fixed it.
Overall, in my opinion, it's decent service. I don't expect 100% perfection at these rates, and I haven't gotten it either.
What I really like is that, while they guarantee only 384K down, I actually get close to 1.5Mbps down. My upload rate is definitely hard capped at 128K, however. If the service came in a 384/384 speed I'd be fat and happy. 128K up is a bit skinny, but it works okay. I would rather have a capped, reliable rate than a fast, unreliable one.
PacBell understands what it means to be in the bandwidth business. If you know what you are doing, you can call their technical center and deal with the same people you'd work with if you had a T1. If you're a networking professional, I suggest you consider them. If they realize you know what you're talking about, they'll work with you on a professional level.
For all these reasons, there's not really any reason to panic yet about cable modems being run by just a few companies, or even just one. Again, the market is 'bandwidth', not 'cable modems'. The phone companies are behind but not THAT much. Their better understanding of the bandwidth business gives them a real service advantage, at least in many cases.
The time to worry is when you see a single company owning large fractions of more than one competing type of service. The fact that AT&T is selling DSL now actually worries me. I do not believe AT&T is likely to be very competitive with itself, and I know they own a cable company. THAT, I think, is a problem.
In comparison, @Home buying Excite is nearly irrelevant, IMO.
Cripes, Solaris has been in development for how many years now? With how many developers?
Linux is still basically a PC program that has been ported. PC hardware isn't designed to take this kind of load. The hardware itself can't keep up with it. Linux is not *designed* to be a massive server. It's meant to be a desktop for serious technoid users who like technical excellence in their home machines. It happens to make a good server because it's good software, not because it's really designed to be one.
The fact that it trounces NT so thoroughly isn't really an endorsement of Linux, it's simply stinging criticism of NT. Linux wasn't written to be a server: the fact that it is a better 'server' than NT (which IS meant to be a server) simply shows how low the MS standards have gotten.
The underlying technology is being worked on. Each year that goes by will see it get stronger and stronger. Someday it may run Ebay and IRC servers... but then again, it might not. And that is okay. Ebay is gravy.
There is a *reason* Solaris boxes cost $50K plus. And no, the reason isn't (just) that they are greedy. The hardware is designed for an entirely different scale of problem. Sun knows how to make servers, and how to approach the problem from the bottom up. Linux is an accidental server, pressed into use because it happens to be better than the only commercial alternative that most people can afford.
You might as well bash the VW Bug because it isn't as fast as a Ferrari and doesn't pull the load a Mack truck would.
Argh.
Admittedly, 5 moderator points/day helps limit the damage that bad moderators can cause. However, it also limits the benefit that good moderators can bring. I don't think it fundamentally changes anything. It is sort of like a minimizing lens; the overall ratio is the same, but the effect is reduced.
I don't remember why the moderator point limit was implemented. It's possible it may have been done in reaction to abuse that was not being corrected. If that's the case, disregard this idea:
I think what will happen is that yes, there will be some abuse, but that good moderators will correct that abuse, and that the meta-moderation will (after awhile) shut down the abusers.
If that's no good, your idea of 'a chance' to moderate any given comment is a possibility -- I'm thinking a 5% chance per positive karma point. That way, 20 karma means you can moderate any post. Most likely, anyone who has 20 karma is going to have a clue.
I have meta-moderated a few times, and I don't think I have chosen 'unfair' once. In some cases I probably would have chosen differently, but I have not seen ANY examples of grievous boneheadedness. (I don't mark down moderators for making different choices than I would, as long as their choice seems at least faintly reasonable.)
It also seems like the slant on pro-open-source and pro-Linux tends to artificially inflate some posts. I have written posts pointing out limitations of open source (most recently in regard to open docs), and also have written ones in favor of it where I think it really helps. Even though (in my opinion), the open-docs-won't-happen argument was better thought out than most of what I've written, it got no positive moderation at all. Other posts I have put up have gotten a lot of positive moderation, even though they weren't nearly as good, IMO. Maybe I'm too close to the problem, but personally I thought the negative-slant article should have scored higher.
Strikes me as a good idea to remind moderators to be aware of their own biases. I try to correct for mine when I'm moderating. I'm not sure how well I succeed, mind you, but I do try.
It strikes me that documentation will probably never be free in the same way that source code is. You'll probably be able to get basic docs (a la manpages) gratis, but no matter how much RMS says free docs ought to happen, I just don't think they will.
:-) That's not at all like code -- if I release code and get patches back, my program improves and I probably also learn a thing or three.
:-)
There's little benefit in releasing docs open source. If I document a program and release the documentation, patches to the docs will enhance them for OTHER readers but not for ME. I already know the program -- that's why I documented it.
And I definitely lose $$$ by not selling the documentation. Docs are NOT fun to write. It's not like being an artist/programmer. Really good documentation hides the author as much as possible. It's an exercise in negative ego gratification. And to write really good docs, you have to go OVER and OVER and OVER them... very tedious work. To get talented people to do this sort of work, you most often have to pay them. It's not like coding at all -- there's not much of a rush in completing a chapter.
Altruism does exist, but I don't think it's nearly as powerful as some folks would like to think. Most individuals will behave in a way that they perceive benefits them the most. Some have longer sight than others, of course.... but even the longest view doesn't really reveal THAT MUCH benefit to really good, free documentation *to the writer of the docs*. If someone uses my program, they will tend to want me to maintain and improve it, and chances are pretty good they'll pay me to do it. If someone uses my docs, they're not locked into my way of writing in nearly the same way they are with code. Giving away free code is a bit like Netscape giving away free browsers... it resulted in a great deal of revenue. Giving away free docs is more like, say, a search engine... it can be profitable but doesn't have the loyalty. Switching code bases is painful; switching docs (or search engines) is instant and painless.
Competing with that is the simple fact that there's a lot of immediate benefit to be gained by selling those same docs instead.
For these reasons, I simply don't think open docs will ever achieve the same status that open source will. Before very much longer, open source will be the way most development is done. The advantages are enormous. That is not true with docs; the present capitalistic method will likely continue indefinitely.
Sorry, RMS. People are selfish.
Bah.
Real Luddites use toggle switches.
:-)
I wasn't saying Clarke is anti-inquisition, although I'm sure he would have to be. He obviously feared religion a good deal, and there are vast, vast reams of data to support his view. However, religion certainly isn't the only source of evil in the world: government has done its fair share.
I think most Western religions encourage one not to think and seek answers for oneself, but to blindly accept canned answers. That is dangerous. Nazi Germany did the same thing. It's dangerous no matter who uses it. Telling people not to think, to shut up and go along with the program, results in pain and violence almost 100% of the time. But the religions Clarke feared mostly have this as a fundamental tenet of the religion.. (Judaism is at least a little different than this, I understand, but I am not really that familiar with it.)
But then you go into how fraternities justify dirty deeds, and I don't get where you're coming from. What fraternities? What dirty deeds?
And then you go really wacked out and say that Katz' motivation of protecting us from the inquisition is a bad one.
Say what? Where the hell did you get that? Katz was reviewing a book, remember? I guess you swallowed the assertion that he's a Mason hook, line, and sinker. And where 'enemies of Jews' came from is entirely beyond me.
I agree with you that anti-inquisition probably doesn't belong here, and judging from the overall quality of this thread, I believe this will be my last post on it.