The claims the parent article makes about the FDIC are entirely fictional.
The FDIC was created in 1933, because so many banks failed in the Great Depression. It had nothing whatsoever to do with the "old west", stagecoaches, or bank theft. It was a policy response to prevent the failure of most of the banking industry.
It's widely accepted as a good thing, but there are arguments that the FDIC actually increases the risk of bank failure, because it in effect allows the banks to take risks with their customers' money that they ordinarily would not, were it not for the "safety net". And customers don't do as much due diligence as they should, because they know they are protected by the Feds.
I don't really have any argument with the rest of the parent's comments, but couldn't let the FDIC thing slide.
Just because 51% of the population thinks that it's okay to tear an {insert member of current most-hated group} limb from limb on public television, that doesn't make it right.
If the majority of the country belongs to one Christian sect or another, is it ok to repress Buddhists or Wiccans or Satanists?
The form of the government is a very great deal less important than its scope. I'd rather have a monarchy with really strong protection of fundamental liberties than a democracy without.
(As an aside, that's part of the problem in Iraq; we're focused on "creating a democracy" instead of "creating a free nation.")
I thought that was actually one of the finest parts of the series; the mix of very old and very new technology. Not everyone is wealthy, and the poor make do as they always have; with old, outmoded technology. Horses are WAY better than nothing; if you have no industrial base to build tractors, horses are about it. If there ever ends up being any kind of human colonization of other planets, I very strongly suspect you'll see results very, very similar to Firefly's; horses and guns still in use, even when laser weapons and robots exist somewhere in the empire.
Hell, in real life, you have subsistence farmers scraping a living out of the dirt, using animal power and living in homes without electricity.... but they have t-shirts and Nikes.
It's only in the really developed places that everything in high-tech; look at the episode where they rob the hospital, for instance. Not a horse in sight.
I think that show could have been amazing, given enough time to find its legs. Whedon is good at changing his characters over time, so we probably wouldn't have had to suffer with "psycho super girl" for all that long. Would have been interesting to see where he went with it. I found that plotline a bit tedious (has been done too many times), but it probably wouldn't have lasted all that long.
My real complaint was probably the prostitute; I liked the character and the actress, but I wished they were a little less graphic about it.
I mentioned Firefly to my sister and she said "oh, that's the show with the.... space trollop, isn't it?" She wasn't impressed either.:-)
This runs the bzip locally; this is good if you're on the same network and your server carries a heavy load. If bandwidth rather than CPU is your limiting factor:
This runs the bzip remotely, which puts more of a load on the server being backed up, but sends a lot less data down the pipe.
Note that this is NOT a perfect backup strategy, unless you stop all your services first. It's a quick and dirty hack that will usually work. You should probably test every backup file you create this way; if a file changes while tar is reading it, it may give up and bail out before completing the backup. (I found this out the hard way, when I needed to restore but my file was incomplete; it stopped at/var/log and didn't get/var/www. This was bad.)
As far as your service thing goes... I mostly run debian, and I have found that many scripts won't run if/etc/profile hasn't been executed. So I often write quick little wrappers that source/etc/profile and ~/.bashrc and then start or stop my service. That fixes most scripts.
Probably would be smart to write a generic wrapper that does that and then runs the command I want, but I haven't done that yet.:-)
He's basically arguing that the whole Linux community should band together and make One True Window Manager, and cease development on all the alternatives.
It's like he thinks that someone can just... order this to happen. OK, I hereby decree that KDE is the One True Window Manager. If you're working on one of the others, stop work immediately and start learning to code KDE.
While I'm at it, I also decree that it will only rain at night, except when there's a game, and never on weekends.
There are lots and lots of good ideas for interfaces out there, and competition and the "marketplace" will eventually sort out which ones are superior. Over time, Linux may gradually move toward a single unified interface naturally. Maybe someday KDE will be so good that all the GNOME hackers will just stop working on their project, or vice-versa.
But it will NEVER happen based on decree or spontaneous cooperation by everyone involved in Linux. Different programmers, different opinions, different approaches. It's the old herding cats problem. Linux users like choice; in most cases, that is the ENTIRE REASON they are using it.
Taking the choice out of Linux might help it on the mainstream desktop for a short while, but long-term would destroy it, because all the developers would go elsewhere.
I could probably reduce this entire post to four words: Competition good. Monoculture bad.
Windows 3.1 was the first major PC windowed interface. (there were earlier versions and a couple of alternatives, but 3.1 was the first one to really hit the bigtime.)
NT 3.51 looked fairly similar but had some significant differences.
95 was different. NT 4.0 was revamped to look kind of like 95, but it was quite different once you did a little looking around. Fundamental user interaction was about the same, but for system administration they didn't resemble one another at all.
98 changed things again, and then 2000 was introduced with yet more changes. Wasn't it in 98 where they first tried that Web Desktop (single-click) interface?
XP redid it AGAIN, massively, to the point that I feel my use of the computer is significantly impaired until I get the interface elements all switched back to what I'm used to.
Every new iteration of Windows has changed interface; in the case of the Web Desktop, they tried to completely change the metaphor.
You can support your argument by simply decreeing that all these changes were "minor' ones, but personally I disagree with that.
When the first Tomb Raider came out, we had it in glorious accelerated 3DFX loveliness at the computer store where I worked. It was a popular demo.
One of my coworkers brought her daughter in to see it; she stayed and played it for hours. The next day she brought her friends over, and we had a whole gaggle of teenage girls totally engrossed in the game. I remember being pretty amazed at how hooked they were; one of them would play and they'd all watch, and every once in awhile, they'd change who was controlling Lara. They were *mesmerized*.
At the time, I figured it was because it was a female protagonist. I remember being acutely uncomfortable with Lara Croft and her Breasts of Impalement being a role model for those girls; talk about your unrealistic images! Even Barbie is rather staid and normal-looking compared to Lara Croft.
Not sure exactly where I'm going with this. I guess I'm trying to point out that girls do indeed play video games if given material that interests them, and also that the typical male-oriented bimbo with the chainmail bikini may not be doing young women any favors. (or, probably, young men, for that matter.)
I'm not saying it's going to leave huge psychic scars, but conditioning is probably the single most powerful influence on human behavior, and I suspect that this particular kind of conditioning isn't very healthy.
(and yes, other media may be worse, but that doesn't make video games any better. It's still wrong for me to beat you with a stick, even if my neighbor would shoot you instead.)
Mark Hamill is a talented voice actor, however. He did the voices to a lot of early Lucas games, and you could NOT TELL it was him.
For instance, in Full Throttle, he voiced several characters, including the main bad guy (rather sepulchral, like the classic undertaker's voice). You would never, in a million years, think it was Hamill. He also did a number of voices in Indiana Jones and the Fate of Atlantis, and the only way to tell was by reading the credits.
Strange interview. Usually, the idea is to ask leading questions and let your subject expound. This one, um, isn't liek that. The author expounds at length about something he's interested in, and the developer of the game station says something along the line of "I agree", and then the 'interviewer' is off and running again. It's really an editorial in disguise.
Aces of the Deep, Sierra (386-486 range) ( I think this was after PoP by some time) M1 Tank Platoon, from Microprose (286) Red Storm Rising, also from Microprose (VERY old game, I think it ran fine on a 4.77Mhz XT)
Most PC games before about 1990 had horrid graphics and next to no sound. There aren't many games from that period that are still much fun.
There are some amazing classics from the Amiga in the 1985-1990 timeframe, though, and they are still quite playable on emulators.
As an aside, I remember when we Amiga owners believed that it would be impossible to ever emulate the machine (which was itself trying to emulate the PCs of the time and not doing a very good job). That wasn't as stupid as it sounded -- it takes about a gigahertz processor to emulate the 7.14Mhz Amiga well, and more horsepower doesn't hurt. That was some seriously amazing hardware.
The AI in Half-Life wasn't too bad. The teams of commandoes weren't that tough, but they communicated with each other and were at least a little coordinated. They'd actually throw grenades at you if they knew roughly where you were but couldn't see you. You do still end up blasting everything that moves, but the commando teams, at least, tried to make that a little harder on you.
I don't think Half-Life promised great AI on the box, either (though I could certainly be wrong, it's been years!). Underpromise and overdeliver; great recipe for success.
I'd watch HL2; the intro movies are sure interesting.
I suspect that the PC will remain the FPS king for one simple reason: the mouse. Until they come up with something equally elegant (thumbsticks aren't it!) on the consoles, that's where the real FPS fanatics are going to hang out.
If you read between the lines, they're saying that they are lying about the movies ("carefully crafted marketing image") and that the customers are catching on faster than they used to ("You could buy your gross for the weekend and overcome bad word of mouth, because it took time to filter out into the general audience.").
The old trick of shoveling out crap but still making money isn't working anymore. And instead of trying to fix the problems (make better movies and stop lying about the product), they're blaming the faster communication methods.
Eventually, of course, it's going to result in better movies; the companies will have to adapt to the new reality or die. Unlike with our friends at the RIAA, they won't be able to buy legislation to prop up their failing business model.
I don't think it's biased. On my standalone Linux server, my passwords are relatively protected. On my standalone OSX server, they aren't protected very well. Unless you are claiming that the NetInfo passwords on a standalone box are bogus, and the real ones are in/etc/shadow? That would be great, but it sure didn't look like that to me when I was last looking at this.
I don't generally use distributed passwords, even on medium-sized server farms; each machine is a standalone entity. From my perspective, OSX has a big hole in it that Linux does not.
And yes, ypcat does do those things. It has a very bad rep because of that (among other problems), and that's part of the reason they came out with NIS+, which works better.
I haven't studied NIS much myself, but I remember our Solaris guy telling us about this.... that NIS+ was a lot better and much more secure, but a lot more complex to get running properly. I got the impression that NIS 1 was sort of considered the bastard stepchild at Sun -- still around, but unwanted, and everyone was hoping it would disappear soon.
OSX emulating an old, sort-of-broken protocol is probably not a feature.
I'm sure Kerberos is great, but it seems like overkill for standalone boxes. A properly-locked-down Netinfo would seem to be a much better solution.
That's a good point. I'm just used to having control, and I don't like indeterminate things.... but hotplugging is a very good point.
In theory, the dependencies in startup scripts SHOULD be handled by your package manager. But in theory, Windows should never crash and we're all rich, rich, rich. *grin*
I'll check out daemonic, thanks. I tried everything I could think of (lots) to get it running back in November. I had no problem starting anything else, like dhcp or squid or apache. It was JUST postfix that broke. Incredibly frustrating.
There's a lot of info in a hash. Go look up "john the ripper" and "crack" for examples. The hash value doesn't directly tell you what the password is, but it's a powerful too to help brute-force the password, because it can be chomped at offline, on another computer completely.
I haven't looked into this recently, but as of two or three years ago, a normal alphanumeric password of eight characters was generally crackable in 12 hours or less on a fast machine. They have probably gone to better hashing algorithms since then, but PCs are a lot faster now, and clustering is becoming more and more common. There is just no reason to EVER reveal the password hash; it can't ever help, and can most certainly hurt. That is why they invented/etc/shadow, which broke a lot of stuff at the time. It was worth the pain.
As far as learning OSX goes.... my biggest struggles were in A) the nonexistent documentation, and B) trying to get the open-source stuff I'm used to, operational. Trying to compile most free software under OSX is really difficult (or at least it was when I was last really working at this, late last year). The Fink project helps a lot, but even that breaks sometimes, and in general it's not very well-integrated into the rest of the system. It's not their fault, they're making the best they can of a strange setup. Replacing Apple defaults with better alternatives is awkward. I have a lot of experience with Postfix, and I simply *could not* get the fink package to deliver mail how I wanted. I ended up falling back to a Debian install on another machine, which took maybe 15 minutes to set up.
The startup system under OSX sucks rocks. The whole/Library/SystemStartup thing is horribly broken. It's not predictable in terms of what order it will start things, and there's no shutdown sequence *at all*. You have to manually shut down your services before rebooting! And some things (like postfix) absolutely WILL NOT start from SystemStarter. (I spent a LONG time on this one; I still have to log into my Mac after a reboot and issue 'postfix start' as root. It WILL NOT run from SystemStarter.)
I can't be a lot more specific, as my memories are getting hazy. If you'd asked me last November or so I could have given you numerous and very specific examples. Unfortunately, I gave up about then and went back to Linux for my free software stuff.
OSX is certainly a long way ahead of where Linux was in the early years, and after it's had 10 years of development I expect it will be very slick indeed. But, at the moment, I think most people with Unix experience would have an easier time setting up a Linux server than an OSX one, at least assuming they need server software other than the defaults provided in OSX.
I am looking forward to Panther, despite the exorbitant upgrade charge. It's still fun to play with -- but to WORK with, I'll use Linux.
This is a major culture collision, unfortunately, and I just don't see how it's going to come out well.
I think, for the most part, the Mac users are RIGHT; case-sensitivity is incredibly, incredibly non-intuitive for a non-techie. MyDocument should be the same as mydocument, from a human perspective. Learning that capital-MD was different than lowercase-md was painful for me when I started using Unix, and I'm a techie!
I understand where these people come from; they're used to being able to use a computer without having to think about it much, and I can't really blame them for wanting to.
Unix, on the other hand, has 30 years of accumulated code, probably billions of lines, and any of it that touches filesystems believes in case sensitivity.
In general, code should adapt to people, instead of the other way around, so the RIGHT thing to do is probably to convert the Unix software to understand case-preserving-but-insensitive filesystems. But if that happens at all, it's gonna take a long time. I'm not holding my breath.
But even if you totally omit the cost of hardware (not unreasonable, as it is maintenance that really costs), the XServe isn't any easier to maintain than Linux is. I find OSX to be much harder to work with, from an administration perspective.
It's a better DESKTOP than Linux (not a ton better, but better) -- but as a SERVER it's not as good. The hybrid OS it's running will run most open source stuff (thanks, fink guys!) but getting that stuff working is often a royal PITA, *harder* than it is on Linux.
And the marriage of the Mac's non-cap-sensitive filesystem with the fundamental case expectations of Unix is FAR from a match made in heaven. It is just awkward and annoying as hell.
OSX is really pretty, and I own a dual G4 tower myself. But I seriously question the constant adulation it gets here on Slashdot; I have often wondered if the place is being astroturfed. Yes, it's pretty; yes, it's Unix. Apparently their laptops are pretty sweet. But as a desktop, I see it as being competitive but hardly compelling. I don't think it's going to be any cheaper to maintain than a well-run network of XP machines. And as a server, it's a LOT harder to deal with. I assume part of that is the learning curve, but it's not an instant fit to most open-source stuff the way Linux is.
On top of that, I don't particularly trust it. From a security perspective, I'm not at all sure about the quality of the design. Consider: the nidump utility dumps out *encrypted passwords* to ANY user on the box, even 'nobody'. In other words, OSX doesn't even have the equivalent of shadow passwords! That is just so overwhelmingly boneheaded that I wouldn't trust it with my critical data. Sure, I could remove nidump, but it's not SUID or anything -- at some interface level, the box will happily spit out its encrypted password hashes to any user, and it's not like Joe Hacker is going to have a hard time ftping nidump back in. Their password hashes may be better than most (no idea), but even if they are, that doesn't excuse handing them out to just anyone. No matter how hard they are to crack, the advance of CPU power makes them constantly easier. This is just *so* stupid that I question the fundamental design; are they even thinking about security at Apple?
The only spot where I can see Apple being really compelling is in the Unix laptop space. If you want Unix on the road, Apple is far and away the best choice, probably years ahead of anything you can buy in the Intel space. But for the "normal OS" laptop market, I suspect that a Centrino Thinkpad will mostly run rings around a Powerbook; similar build quality, equivalently nice screen (though not in widescreen format, AFAIK), better battery life, much faster, and cheaper.
I generally like Cringely, but this week's column is just purely ignorant. He's making absolutely wild assertions on no evidence whatsoever. IT people are under incredible cost pressure; if Apple was really cheaper and better, they'd be switching in droves.
And don't even get me started on "12 people and a bunch of contractors would run IBM's network better".... for a network supporting THREE HUNDRED THOUSAND EMPLOYEES plus god knows how many customers, directly or indirectly, dependent on that network.
Just think of the install process as your first puzzle.:-)
You can absolutely play Infocom games on your Palm. I was doing that three or four years ago. I think I had to pay for the interpreter, but after this long there may be free ones out. Look up Infocom on any Palm site and you should run across something.
I was pleased to finally get around to playing Leather Goddesses of Phobos. The other people on the train, however, must have thought I was pretty bizarre -- staring at this tiny handheld device, scribbling away furiously, smiling frequently and occasionally outright laughing. It is probably fortunate that the Classics CD didn't come with the scratch-and-sniff of the original. *grin*
The original Graffiti Palms are not ideal for doing text adventures, as the input is fairly slow, but it beats nothing. I imagine the newer input methods would work better.
I also used that interpreter, whatever it was called, to work most of the way through a freebie called Christminster, by Gareth Rees. I strongly suggest you look into it. Personally I think it stands with the very best of the Infocom games. Not everyone agrees with that, but the fact that it could even be argued tells you that it's good.
And btw, I LOVED Trinity back in the 80s, but had trouble getting back into it this go-round. It's HARD. It appears I got more impatient, more stupid, or both.:-)
I may be feeding a troll here, but I'm not quite sure whether this guy is serious or not.
You say "[nobody] is preparing to challenge SCO and get this resolved", but that's utter crap. Many, many requests have been made to SCO to reveal the infringing source code.
SCO doesn't actually want to reveal it, because they know that it will be replaced with clean code almost instantly. The fact that they're unwilling to talk about which code infringes shows that they are dealing in bad faith. Their purpose is to damage Linux and (attempt to) extort protection money from the gullible.
Until SCO tells the community what code infringes, there is nothing that the community can do. You seem to be trying to blame the people on this side of the fence, who would like nothing better than to resolve the problem. It is SCO that is preventing any real solution. Their purpose is *not* protecting their IP. If it were, the problem would already have been resolved.
And as far as "having implications at some time in the future" -- also probably incorrect. I have downloaded a copy of the Linux kernel from SCO *well after* they filed their IBM suit, so they have knowingly given me their code under the GPL. I can now transfer that code to anyone else I choose, also under the GPL. So I am safe, and anyone I work for is safe. Hell, anyone I even vaguely KNOW is safe; I can email the file to anyone that asks.
You also call it "a seemingly large problem", but it really isn't. Even if the GPL doesn't hold up in court (unlikely), it'll probably take about two weeks for all infringing code to be removed as soon as it's revealed.
But, hey, if you feel safer, you go right ahead and call SCO and offer to pay them their protection money. The head of Linux IP Licensing, Mr. Barnum, will call you right back.
(warning: I'm just tossing this out from memory without doing any double-checking on it first, so read with caution and pay attention to replies.)
I believe that's basically what they're already doing.
If I understood what I read correctly, the "X86" CPUs on the market aren't really X86 CPUs anymore. Instead, they are essentially a super-fast hardware emulator of an instruction set. The real instruction set of these chips doesn't resemble X86 *at all*; the chip decodes on the fly from the X86 macro-ops down to the chip's native micro-ops, which are smaller and simpler and easier to track when running in parallel across several execution units.
That's part of why most software emulation is so slow -- you are in essence comparing generalized software solutions to incredibly well-engineered hardware solutions.
If we had a different instruction set, would we really benefit? For the vast majority of us, even the Slashdot crowd, no. The compiler guys would probably like it a lot, but very few programmers work in anything lower than C. The actual "machine language" is mostly unimportant. And it's not even REALLY the machine language of the chip anymore!
Even assembly coders, these days, are writing in a form of interpreted language. The "bare metal" guys aren't REALLY at the bare metal anymore; even they are working at a level of abstraction.
After vast effort and the expenditure of moderate sums of cash (SETI is really done on the cheap), scientists will be elated to finally discover an alien signal, and chagrined to find that it translates to the rough equivalent of a rude hand gesture and a raspberry.:-)
(or possibly "If all you can detect is this signal, you're too stupid to bother with. Stay home and evolve for a few million years.")
To some degree, it's the old instant gratification versus delayed gratification. You can have one dollar now, or you can have two dollars later. If you take the dollar now, you can put it to work elsewhere. If your extended project will net you two dollars sooner than investing the one dollar would, then from an abstract perspective, it's a worthwhile investment.
However, you must first be able to survive long enough to collect your two dollars, so if it's a matter of "publish or perish", you'd better publish and then clean up the mess afterward.
Additionally, a company has many competing needs for money. If you're not in a publish or perish situation, and if you can show that using the money on your project will result in more profitability than using it on other projects, then in theory, you will get funded.
Real politics, of course, often don't work this way, but the theory sounds good.
Ultimately, process and engineering are only important, to a company, from a profit perspective. A high-quality process will generally result in a high-quality product, but not all companies care about quality. Process purely for the sake of process is wasted money.
Process to meet business objectives may or may not be wasteful, depending on other investment opportunities.
Slashdot Mirror
The claims the parent article makes about the FDIC are entirely fictional.
The FDIC was created in 1933, because so many banks failed in the Great Depression. It had nothing whatsoever to do with the "old west", stagecoaches, or bank theft. It was a policy response to prevent the failure of most of the banking industry.
It's widely accepted as a good thing, but there are arguments that the FDIC actually increases the risk of bank failure, because it in effect allows the banks to take risks with their customers' money that they ordinarily would not, were it not for the "safety net". And customers don't do as much due diligence as they should, because they know they are protected by the Feds.
I don't really have any argument with the rest of the parent's comments, but couldn't let the FDIC thing slide.
Beware the tyranny of the majority.
Just because 51% of the population thinks that it's okay to tear an {insert member of current most-hated group} limb from limb on public television, that doesn't make it right.
If the majority of the country belongs to one Christian sect or another, is it ok to repress Buddhists or Wiccans or Satanists?
The form of the government is a very great deal less important than its scope. I'd rather have a monarchy with really strong protection of fundamental liberties than a democracy without.
(As an aside, that's part of the problem in Iraq; we're focused on "creating a democracy" instead of "creating a free nation.")
Democracy, in and of itself, is just mob rule.
"Markets can remain irrational longer than you can remain solvent." -- John Maynard Keynes
:-) )
(for the most part I don't like his economics, but that's a great quote
I thought that was actually one of the finest parts of the series; the mix of very old and very new technology. Not everyone is wealthy, and the poor make do as they always have; with old, outmoded technology. Horses are WAY better than nothing; if you have no industrial base to build tractors, horses are about it. If there ever ends up being any kind of human colonization of other planets, I very strongly suspect you'll see results very, very similar to Firefly's; horses and guns still in use, even when laser weapons and robots exist somewhere in the empire.
.... space trollop, isn't it?" She wasn't impressed either. :-)
Hell, in real life, you have subsistence farmers scraping a living out of the dirt, using animal power and living in homes without electricity.... but they have t-shirts and Nikes.
It's only in the really developed places that everything in high-tech; look at the episode where they rob the hospital, for instance. Not a horse in sight.
I think that show could have been amazing, given enough time to find its legs. Whedon is good at changing his characters over time, so we probably wouldn't have had to suffer with "psycho super girl" for all that long. Would have been interesting to see where he went with it. I found that plotline a bit tedious (has been done too many times), but it probably wouldn't have lasted all that long.
My real complaint was probably the prostitute; I liked the character and the actress, but I wished they were a little less graphic about it.
I mentioned Firefly to my sister and she said "oh, that's the show with the
One non-obvious application of this is remote backup from one machine to another.
/etc/nobackup file that looks like this:
/; tar -X /etc/nobackup cf - *" | bzip2 -9 >remotehost.tar.bz2
/; tar -X /etc/nobackup -cjf - *" >remotehost.tar.bz2
/var/log and didn't get /var/www. This was bad.)
/etc/profile hasn't been executed. So I often write quick little wrappers that source /etc/profile and ~/.bashrc and then start or stop my service. That fixes most scripts.
:-)
You can create an
proc/*
tmp/*
dev/*
var/log/*
and then you can:
ssh -l root remotehost "cd
This runs the bzip locally; this is good if you're on the same network and your server carries a heavy load. If bandwidth rather than CPU is your limiting factor:
ssh -l root remotehost "cd
This runs the bzip remotely, which puts more of a load on the server being backed up, but sends a lot less data down the pipe.
Note that this is NOT a perfect backup strategy, unless you stop all your services first. It's a quick and dirty hack that will usually work. You should probably test every backup file you create this way; if a file changes while tar is reading it, it may give up and bail out before completing the backup. (I found this out the hard way, when I needed to restore but my file was incomplete; it stopped at
As far as your service thing goes... I mostly run debian, and I have found that many scripts won't run if
Probably would be smart to write a generic wrapper that does that and then runs the command I want, but I haven't done that yet.
He's basically arguing that the whole Linux community should band together and make One True Window Manager, and cease development on all the alternatives.
It's like he thinks that someone can just... order this to happen. OK, I hereby decree that KDE is the One True Window Manager. If you're working on one of the others, stop work immediately and start learning to code KDE.
While I'm at it, I also decree that it will only rain at night, except when there's a game, and never on weekends.
There are lots and lots of good ideas for interfaces out there, and competition and the "marketplace" will eventually sort out which ones are superior. Over time, Linux may gradually move toward a single unified interface naturally. Maybe someday KDE will be so good that all the GNOME hackers will just stop working on their project, or vice-versa.
But it will NEVER happen based on decree or spontaneous cooperation by everyone involved in Linux. Different programmers, different opinions, different approaches. It's the old herding cats problem. Linux users like choice; in most cases, that is the ENTIRE REASON they are using it.
Taking the choice out of Linux might help it on the mainstream desktop for a short while, but long-term would destroy it, because all the developers would go elsewhere.
I could probably reduce this entire post to four words: Competition good. Monoculture bad.
Windows 3.1 was the first major PC windowed interface. (there were earlier versions and a couple of alternatives, but 3.1 was the first one to really hit the bigtime.)
NT 3.51 looked fairly similar but had some significant differences.
95 was different. NT 4.0 was revamped to look kind of like 95, but it was quite different once you did a little looking around. Fundamental user interaction was about the same, but for system administration they didn't resemble one another at all.
98 changed things again, and then 2000 was introduced with yet more changes. Wasn't it in 98 where they first tried that Web Desktop (single-click) interface?
XP redid it AGAIN, massively, to the point that I feel my use of the computer is significantly impaired until I get the interface elements all switched back to what I'm used to.
Every new iteration of Windows has changed interface; in the case of the Web Desktop, they tried to completely change the metaphor.
You can support your argument by simply decreeing that all these changes were "minor' ones, but personally I disagree with that.
When the first Tomb Raider came out, we had it in glorious accelerated 3DFX loveliness at the computer store where I worked. It was a popular demo.
One of my coworkers brought her daughter in to see it; she stayed and played it for hours. The next day she brought her friends over, and we had a whole gaggle of teenage girls totally engrossed in the game. I remember being pretty amazed at how hooked they were; one of them would play and they'd all watch, and every once in awhile, they'd change who was controlling Lara. They were *mesmerized*.
At the time, I figured it was because it was a female protagonist. I remember being acutely uncomfortable with Lara Croft and her Breasts of Impalement being a role model for those girls; talk about your unrealistic images! Even Barbie is rather staid and normal-looking compared to Lara Croft.
Not sure exactly where I'm going with this. I guess I'm trying to point out that girls do indeed play video games if given material that interests them, and also that the typical male-oriented bimbo with the chainmail bikini may not be doing young women any favors. (or, probably, young men, for that matter.)
I'm not saying it's going to leave huge psychic scars, but conditioning is probably the single most powerful influence on human behavior, and I suspect that this particular kind of conditioning isn't very healthy.
(and yes, other media may be worse, but that doesn't make video games any better. It's still wrong for me to beat you with a stick, even if my neighbor would shoot you instead.)
For instance, in Full Throttle, he voiced several characters, including the main bad guy (rather sepulchral, like the classic undertaker's voice). You would never, in a million years, think it was Hamill. He also did a number of voices in Indiana Jones and the Fate of Atlantis, and the only way to tell was by reading the credits.
I figured I'd double-check on this, and found a site that lists a lot of voice work that he's done: http://voices.fuzzy.com/actor.idc?actor_id=472
They're showing 203 credits for him at the moment, and they didn't list the Indy game I mentioned, so they're definitely not (yet) exhaustive.
He has more depth than people like to believe.
Strange interview. Usually, the idea is to ask leading questions and let your subject expound. This one, um, isn't liek that. The author expounds at length about something he's interested in, and the developer of the game station says something along the line of "I agree", and then the 'interviewer' is off and running again. It's really an editorial in disguise.
Still interesting. I may buy one of these.
Aces of the Deep, Sierra (386-486 range) ( I think this was after PoP by some time)
M1 Tank Platoon, from Microprose (286)
Red Storm Rising, also from Microprose (VERY old game, I think it ran fine on a 4.77Mhz XT)
Most PC games before about 1990 had horrid graphics and next to no sound. There aren't many games from that period that are still much fun.
There are some amazing classics from the Amiga in the 1985-1990 timeframe, though, and they are still quite playable on emulators.
As an aside, I remember when we Amiga owners believed that it would be impossible to ever emulate the machine (which was itself trying to emulate the PCs of the time and not doing a very good job). That wasn't as stupid as it sounded -- it takes about a gigahertz processor to emulate the 7.14Mhz Amiga well, and more horsepower doesn't hurt. That was some seriously amazing hardware.
The AI in Half-Life wasn't too bad. The teams of commandoes weren't that tough, but they communicated with each other and were at least a little coordinated. They'd actually throw grenades at you if they knew roughly where you were but couldn't see you. You do still end up blasting everything that moves, but the commando teams, at least, tried to make that a little harder on you.
I don't think Half-Life promised great AI on the box, either (though I could certainly be wrong, it's been years!). Underpromise and overdeliver; great recipe for success.
I'd watch HL2; the intro movies are sure interesting.
I suspect that the PC will remain the FPS king for one simple reason: the mouse. Until they come up with something equally elegant (thumbsticks aren't it!) on the consoles, that's where the real FPS fanatics are going to hang out.
If you read between the lines, they're saying that they are lying about the movies ("carefully crafted marketing image") and that the customers are catching on faster than they used to ("You could buy your gross for the weekend and overcome bad word of mouth, because it took time to filter out into the general audience.").
The old trick of shoveling out crap but still making money isn't working anymore. And instead of trying to fix the problems (make better movies and stop lying about the product), they're blaming the faster communication methods.
Eventually, of course, it's going to result in better movies; the companies will have to adapt to the new reality or die. Unlike with our friends at the RIAA, they won't be able to buy legislation to prop up their failing business model.
I don't think it's biased. On my standalone Linux server, my passwords are relatively protected. On my standalone OSX server, they aren't protected very well. Unless you are claiming that the NetInfo passwords on a standalone box are bogus, and the real ones are in /etc/shadow? That would be great, but it sure didn't look like that to me when I was last looking at this.
.... that NIS+ was a lot better and much more secure, but a lot more complex to get running properly. I got the impression that NIS 1 was sort of considered the bastard stepchild at Sun -- still around, but unwanted, and everyone was hoping it would disappear soon.
I don't generally use distributed passwords, even on medium-sized server farms; each machine is a standalone entity. From my perspective, OSX has a big hole in it that Linux does not.
And yes, ypcat does do those things. It has a very bad rep because of that (among other problems), and that's part of the reason they came out with NIS+, which works better.
I haven't studied NIS much myself, but I remember our Solaris guy telling us about this
OSX emulating an old, sort-of-broken protocol is probably not a feature.
I'm sure Kerberos is great, but it seems like overkill for standalone boxes. A properly-locked-down Netinfo would seem to be a much better solution.
That's a good point. I'm just used to having control, and I don't like indeterminate things.... but hotplugging is a very good point.
In theory, the dependencies in startup scripts SHOULD be handled by your package manager. But in theory, Windows should never crash and we're all rich, rich, rich. *grin*
I'll check out daemonic, thanks. I tried everything I could think of (lots) to get it running back in November. I had no problem starting anything else, like dhcp or squid or apache. It was JUST postfix that broke. Incredibly frustrating.
There's a lot of info in a hash. Go look up "john the ripper" and "crack" for examples. The hash value doesn't directly tell you what the password is, but it's a powerful too to help brute-force the password, because it can be chomped at offline, on another computer completely.
/etc/shadow, which broke a lot of stuff at the time. It was worth the pain.
/Library/SystemStartup thing is horribly broken. It's not predictable in terms of what order it will start things, and there's no shutdown sequence *at all*. You have to manually shut down your services before rebooting! And some things (like postfix) absolutely WILL NOT start from SystemStarter. (I spent a LONG time on this one; I still have to log into my Mac after a reboot and issue 'postfix start' as root. It WILL NOT run from SystemStarter.)
I haven't looked into this recently, but as of two or three years ago, a normal alphanumeric password of eight characters was generally crackable in 12 hours or less on a fast machine. They have probably gone to better hashing algorithms since then, but PCs are a lot faster now, and clustering is becoming more and more common. There is just no reason to EVER reveal the password hash; it can't ever help, and can most certainly hurt. That is why they invented
As far as learning OSX goes.... my biggest struggles were in A) the nonexistent documentation, and B) trying to get the open-source stuff I'm used to, operational. Trying to compile most free software under OSX is really difficult (or at least it was when I was last really working at this, late last year). The Fink project helps a lot, but even that breaks sometimes, and in general it's not very well-integrated into the rest of the system. It's not their fault, they're making the best they can of a strange setup. Replacing Apple defaults with better alternatives is awkward. I have a lot of experience with Postfix, and I simply *could not* get the fink package to deliver mail how I wanted. I ended up falling back to a Debian install on another machine, which took maybe 15 minutes to set up.
The startup system under OSX sucks rocks. The whole
I can't be a lot more specific, as my memories are getting hazy. If you'd asked me last November or so I could have given you numerous and very specific examples. Unfortunately, I gave up about then and went back to Linux for my free software stuff.
OSX is certainly a long way ahead of where Linux was in the early years, and after it's had 10 years of development I expect it will be very slick indeed. But, at the moment, I think most people with Unix experience would have an easier time setting up a Linux server than an OSX one, at least assuming they need server software other than the defaults provided in OSX.
I am looking forward to Panther, despite the exorbitant upgrade charge. It's still fun to play with -- but to WORK with, I'll use Linux.
This is a major culture collision, unfortunately, and I just don't see how it's going to come out well.
I think, for the most part, the Mac users are RIGHT; case-sensitivity is incredibly, incredibly non-intuitive for a non-techie. MyDocument should be the same as mydocument, from a human perspective. Learning that capital-MD was different than lowercase-md was painful for me when I started using Unix, and I'm a techie!
I understand where these people come from; they're used to being able to use a computer without having to think about it much, and I can't really blame them for wanting to.
Unix, on the other hand, has 30 years of accumulated code, probably billions of lines, and any of it that touches filesystems believes in case sensitivity.
In general, code should adapt to people, instead of the other way around, so the RIGHT thing to do is probably to convert the Unix software to understand case-preserving-but-insensitive filesystems. But if that happens at all, it's gonna take a long time. I'm not holding my breath.
But even if you totally omit the cost of hardware (not unreasonable, as it is maintenance that really costs), the XServe isn't any easier to maintain than Linux is. I find OSX to be much harder to work with, from an administration perspective.
It's a better DESKTOP than Linux (not a ton better, but better) -- but as a SERVER it's not as good. The hybrid OS it's running will run most open source stuff (thanks, fink guys!) but getting that stuff working is often a royal PITA, *harder* than it is on Linux.
And the marriage of the Mac's non-cap-sensitive filesystem with the fundamental case expectations of Unix is FAR from a match made in heaven. It is just awkward and annoying as hell.
OSX is really pretty, and I own a dual G4 tower myself. But I seriously question the constant adulation it gets here on Slashdot; I have often wondered if the place is being astroturfed. Yes, it's pretty; yes, it's Unix. Apparently their laptops are pretty sweet. But as a desktop, I see it as being competitive but hardly compelling. I don't think it's going to be any cheaper to maintain than a well-run network of XP machines. And as a server, it's a LOT harder to deal with. I assume part of that is the learning curve, but it's not an instant fit to most open-source stuff the way Linux is.
On top of that, I don't particularly trust it. From a security perspective, I'm not at all sure about the quality of the design. Consider: the nidump utility dumps out *encrypted passwords* to ANY user on the box, even 'nobody'. In other words, OSX doesn't even have the equivalent of shadow passwords! That is just so overwhelmingly boneheaded that I wouldn't trust it with my critical data. Sure, I could remove nidump, but it's not SUID or anything -- at some interface level, the box will happily spit out its encrypted password hashes to any user, and it's not like Joe Hacker is going to have a hard time ftping nidump back in. Their password hashes may be better than most (no idea), but even if they are, that doesn't excuse handing them out to just anyone. No matter how hard they are to crack, the advance of CPU power makes them constantly easier. This is just *so* stupid that I question the fundamental design; are they even thinking about security at Apple?
The only spot where I can see Apple being really compelling is in the Unix laptop space. If you want Unix on the road, Apple is far and away the best choice, probably years ahead of anything you can buy in the Intel space. But for the "normal OS" laptop market, I suspect that a Centrino Thinkpad will mostly run rings around a Powerbook; similar build quality, equivalently nice screen (though not in widescreen format, AFAIK), better battery life, much faster, and cheaper.
I generally like Cringely, but this week's column is just purely ignorant. He's making absolutely wild assertions on no evidence whatsoever. IT people are under incredible cost pressure; if Apple was really cheaper and better, they'd be switching in droves.
And don't even get me started on "12 people and a bunch of contractors would run IBM's network better".... for a network supporting THREE HUNDRED THOUSAND EMPLOYEES plus god knows how many customers, directly or indirectly, dependent on that network.
Words fail me.
Where's "-1, Evil"?
Just think of the install process as your first puzzle. :-)
:-)
You can absolutely play Infocom games on your Palm. I was doing that three or four years ago. I think I had to pay for the interpreter, but after this long there may be free ones out. Look up Infocom on any Palm site and you should run across something.
I was pleased to finally get around to playing Leather Goddesses of Phobos. The other people on the train, however, must have thought I was pretty bizarre -- staring at this tiny handheld device, scribbling away furiously, smiling frequently and occasionally outright laughing. It is probably fortunate that the Classics CD didn't come with the scratch-and-sniff of the original. *grin*
The original Graffiti Palms are not ideal for doing text adventures, as the input is fairly slow, but it beats nothing. I imagine the newer input methods would work better.
I also used that interpreter, whatever it was called, to work most of the way through a freebie called Christminster, by Gareth Rees. I strongly suggest you look into it. Personally I think it stands with the very best of the Infocom games. Not everyone agrees with that, but the fact that it could even be argued tells you that it's good.
And btw, I LOVED Trinity back in the 80s, but had trouble getting back into it this go-round. It's HARD. It appears I got more impatient, more stupid, or both.
You say "[nobody] is preparing to challenge SCO and get this resolved", but that's utter crap. Many, many requests have been made to SCO to reveal the infringing source code.
SCO doesn't actually want to reveal it, because they know that it will be replaced with clean code almost instantly. The fact that they're unwilling to talk about which code infringes shows that they are dealing in bad faith. Their purpose is to damage Linux and (attempt to) extort protection money from the gullible.
Until SCO tells the community what code infringes, there is nothing that the community can do. You seem to be trying to blame the people on this side of the fence, who would like nothing better than to resolve the problem. It is SCO that is preventing any real solution. Their purpose is *not* protecting their IP. If it were, the problem would already have been resolved.
And as far as "having implications at some time in the future" -- also probably incorrect. I have downloaded a copy of the Linux kernel from SCO *well after* they filed their IBM suit, so they have knowingly given me their code under the GPL. I can now transfer that code to anyone else I choose, also under the GPL. So I am safe, and anyone I work for is safe. Hell, anyone I even vaguely KNOW is safe; I can email the file to anyone that asks.
You also call it "a seemingly large problem", but it really isn't. Even if the GPL doesn't hold up in court (unlikely), it'll probably take about two weeks for all infringing code to be removed as soon as it's revealed.
But, hey, if you feel safer, you go right ahead and call SCO and offer to pay them their protection money. The head of Linux IP Licensing, Mr. Barnum, will call you right back.
(warning: I'm just tossing this out from memory without doing any double-checking on it first, so read with caution and pay attention to replies.)
I believe that's basically what they're already doing.
If I understood what I read correctly, the "X86" CPUs on the market aren't really X86 CPUs anymore. Instead, they are essentially a super-fast hardware emulator of an instruction set. The real instruction set of these chips doesn't resemble X86 *at all*; the chip decodes on the fly from the X86 macro-ops down to the chip's native micro-ops, which are smaller and simpler and easier to track when running in parallel across several execution units.
That's part of why most software emulation is so slow -- you are in essence comparing generalized software solutions to incredibly well-engineered hardware solutions.
If we had a different instruction set, would we really benefit? For the vast majority of us, even the Slashdot crowd, no. The compiler guys would probably like it a lot, but very few programmers work in anything lower than C. The actual "machine language" is mostly unimportant. And it's not even REALLY the machine language of the chip anymore!
Even assembly coders, these days, are writing in a form of interpreted language. The "bare metal" guys aren't REALLY at the bare metal anymore; even they are working at a level of abstraction.
After vast effort and the expenditure of moderate sums of cash (SETI is really done on the cheap), scientists will be elated to finally discover an alien signal, and chagrined to find that it translates to the rough equivalent of a rude hand gesture and a raspberry. :-)
(or possibly "If all you can detect is this signal, you're too stupid to bother with. Stay home and evolve for a few million years.")
To some degree, it's the old instant gratification versus delayed gratification. You can have one dollar now, or you can have two dollars later. If you take the dollar now, you can put it to work elsewhere. If your extended project will net you two dollars sooner than investing the one dollar would, then from an abstract perspective, it's a worthwhile investment.
However, you must first be able to survive long enough to collect your two dollars, so if it's a matter of "publish or perish", you'd better publish and then clean up the mess afterward.
Additionally, a company has many competing needs for money. If you're not in a publish or perish situation, and if you can show that using the money on your project will result in more profitability than using it on other projects, then in theory, you will get funded.
Real politics, of course, often don't work this way, but the theory sounds good.
Ultimately, process and engineering are only important, to a company, from a profit perspective. A high-quality process will generally result in a high-quality product, but not all companies care about quality. Process purely for the sake of process is wasted money.
Process to meet business objectives may or may not be wasteful, depending on other investment opportunities.