The iPaq people may be OSS-friendly... that Web site you mention is meant to encourage open source software that runs on handhelds. But all I really do with my handheld (a Palm) is move data to and from it. When will I be able to do that with an iPaq?
I use a PowerBook G3, triple booting Mac OS 9.1, Mac OS X beta, and LinuxPPC. Which one is my best choice for syncing my iPaq? --
It's not as simple as that. I know on my LinuxPPC system, the fonts look like hell. And that's even the TrueType fonts that I, um, ripped off from the Mac OS and serve using xfstt.
The fonts under Linux are fairly legible -- provided you find the one or two point sizes that actually look good, so the loops aren't closing up, etc. But under the Mac OS, ALL the point sizes look fine!
Maybe it has something to do with the fonts for the Mac being designed for a 72 dpi screen resolution, while X11 is designed at 75 dpi? But I thought TrueType was supposed to solve the resolution-dependence problem...
There are legions of corporations and individuals who have been disrespected by Apple--from the BeOS community to the Apple clone industry, all of whom comprise a form
The BeOS community is a formidable enemy? Come on. If Jean-Louis Gassee was so indispensible to the future of computing, Apple wouldn't have canned his ass to begin with.
Power Comuting is a formidable enemy? The people whose primary contributions to the Mac hardware platform were unsupported OEM CD-ROM drives, PS/2 keyboards, and the standard VGA port? (Okay, I'll give you the last one -- but Apple even ships those, now.)
Open up its hardware specs and software so that where now exists little more than a corporate cult, there might exist a vibrant autonomous industry of developers, hackers, and hardware vendors.
I hear this argument a lot. I don't get it. Why would opening up Apple's platform lead to more support? The world already has Linux. Linux is much more mature on Intel hardware than on the Mac/PowerPC. Why would having a mature, open source OS that ran on Mac hardware make people want to concentrate their efforts on the Mac instead of Intel? Even if Apple made the Mac ROMs public domain, why would that inspire people to use them instead of the Intel platform, which currently has a larger installed base than any other form of PC hardware? Especially if you're talking about the consumer market (Apple's primary targer)... it doesn't make any sense.
The only reason Power Computing met with any success at all in the Mac clone market was because they sensed there was demand for lower-cost hardware that ran this closed source OS. Take away the closed source OS, and the need for the "clone" hardware goes away too.
The only reason for an open source Mac OS is because you don't think Apple's hardware is being used well enough. Take away Apple's hardware, and the need for the OS disappears as well.
What Apple is doing now with OS X is continuing to do what it does best -- try to provide the consumer with an all-inclusive computing platform. You buy Apple's hardware, you get Apple's OS. It's a packaged product. Just because it isn't currently well-suited for tinkering, doesn't mean it should be. --
Stuff like this happens all the time. Go back through a few years of magazines and look for any ads, etc. which reference the pro wrestler Hulk Hogan. You may find that at the bottom of the ad there's a legend stating that the name "Hulk Hogan" is TM & Copyright Marvel Comics.
Why? Not because Marvel Comics invented the Hulk Hogan persona. No, it's because Marvel Comics has a character who's a big strong guy who's named the Hulk. Hulk Hogan was (arguably) a fictitious character. He was a big, strong guy. Ergo, in Marvel Comics's eyes, his use of that name was infringing on their trademark on big strong guys with this particular word of the English language in their name.
Now, let's take stock here: Hulk Hogan could certainly not bench press 18 tons. Hulk Hogan was certainly not green, and he didn't undergo any particular transformation when he got angry. It could be argued that his use of the word "hulk" was just plain English language, and nothing more.
Then again, you could argue that it WAS something more. And ultimately, somebody decided it was easier (and better for Hogan's career) if they let Marvel claim the trademark and Hogan just went on about his business of being a wrestler.
Nothing really more clandestine or sinister than that. Just like in this Blizzard case. It's just plain ol' trademark law, contract law, entertainment law, and what-have-you... and Slashdot should probably leave its interpretation to the experts. --
It is a well accepted economic principle that a large quantity of a good or service will cost more than a small quantity.
That's a pretty obtuse statement. What about economies of scale? For instance, the printing industry -- forgetting print-on-demand for the moment, printing two million books has a lower cost per unit than printing one book.
So when you say something "will cost more," what are you talking about? Two McDonald's cheeseburgers will always cost more than one cheeseburger, true. But a two-cheeseburger meal costs less than buying one cheeseburger, a Coke, and some fries. It still "costs more" to get the two cheeseburgers. But it's a better deal.
If you then resell each cheeseburger to somebody at cost, then those people get a cheaper cheeseburger. Alternately, if you mark up the price, you make profit and then you can buy more cheeseburgers. That's about as "well accepted" an "economic principle" as you can get!
I think you hit on it more with this statement:
Rather, human ability and inginuity is required for the manufacture of IP networks.
...because that's just the trick. Your mileage may vary, but out here in California manpower is just the problem with network service. Manpower (and the cost thereof) is what's keeping DSL orders backed up for 6 weeks. Manpower (or the lack thereof, when telcos let go their support staff in an effort to offset costs) is what's keeping network service poor.
To me, it seems like it's practically to the point where there's no difference between ordering a $900/month 1.54Mbps Frame Relay circuit, and ordering a $40/month DSL circuit. It used to be that a Frame Relay reseller could tell their customers that they'd be getting better quality of service guarantees for that extra money; nowadays, there's almost no justification.
Part of the problem has been the whole DSL market share war, which has driven DSL prices artificially low, and in turn impacted the entire industry. And besides that, I'm sure other people can come up with a full laundry list of telco incompetences.
So is the cost of online access going to go up? Most definitely. And not because people are "using it too much" (with P2P etc.) -- it will go up because the telcos can barely cut costs any more than they have already, and they're still as hungry for profits as they've ever been.
But will the money go to the right places, that's the real question. What we should really be worrying about is the price of access continuing to go up while the quality of service continues to go down.
...so-called "hardcore programming" jobs aren't that easy to come by on a contract basis. On the other hand, Web jobs aren't too difficult to pick up. If server-side Java meets your qualifications for "programming," then why not try that? A lot of work in that field around here (Bay Area).
In my opinion, Asmimov's Foundation series would translate to the screen more easily. I'd love to see this on the big screen. Has anyone heard any rumors about this? At the very least, it would make a great mini-series.
They already did the Mule saga. It was called "The Usual Suspects."
Moebius is an artist who pioneered French comics in the 1970s and early 1980s, with his involvement with the magazine Metal Hurlant. If that sounds vaguely familiar, it's because an American version of the magazine concept was launched later, called Heavy Metal, and it reprinted a lot of translations of Moebius's comics.
Born Jean Giraud, he is most famous in Europe for his Western character, Blueberry, whose exploits Giraud signs with his given name. Blueberry still appears in various ongoing series of graphic novels.
In the US, however, Giraud is most widely-known by the pseudonym Moebius. He is the author of many popular comics stories in the science fiction genre, including the Arzach series, as well as The Airtight Garage, which was used as the inspiration for the video arcade in San Francisco's Sony Metreon entertainment center.
Moebius has also done production design work on some science fiction films, including Alien, Tron, and The Abyss. The latter was a rotten movie, but it had a visually interesting version of what alien spacecraft and technology might look like. Should work well for Rama.
If course, I noticed Jobs wasn't ready with a demo to prove this particular claim at his MacWorld keynote. Just a 1-minute video clip being encoded, raced side by side with a Pentium IV box the way he did that Photoshop demo, would have proved it for me.
Aw, complain all you want about not having eighteen mouse buttons on the Titanium PowerBook. Apple obviously feels that a single button is a nicer interface, and I for one am not going to say they're wrong.
What I thought was interesting about Mac OS X Public Beta, on the other hand, is that it AUTOMATICALLY supported the top two buttons of my Kensington 4-button mouse, out of the box. The second button worked the contextual menus, as expected.
And this is an ADB mouse, mind you -- old technology from Apple (all the new mice are USB), and NOT code that was built in to OpenStep!
What this tells me is that Apple is going to lengths to preserve your possible preference, even though the preference of their hardware engineers is something else. As other people have pointed out, if you want more buttons, feel free to plug in a different mouse. I don't know what more you can really ask for.
A lot of DVD players don't support CD-R discs. Don't ask me why it's so; but I don't think it has anything to do with "piracy prevention," it's more by-product of the way the laser picks up data from the discs on DVD players.
My Pioneer DV-606D doesn't support CD-Rs either, for what it's worth.
I've actually been wanting a list of players that DO support CD-Rs for a while now. (Besides playing audio CDs, it's useful for VCDs.)
As near as I can tell, the deal with these Pinball 2000 machines is that they have NO moving parts. It's a last-ditch attempt to make an economically viable pinball machine. The ball slides around, interacting with magnets (?) while the targets are all "virtual," displayed on a video screen below.
Wait a minute... who are these tunnel-vision Mac users you claim have moved to LinuxPPC, anyway? Have you ever met one? If they won't use Windows, what in hell makes you think they'd use Linux?
And since when has the sole purpose of LinuxPPC development been to let Linux run on Macs, anyhow?
I honestly don't think you have the slightest clue what you're talking about. But keep on feeling superior anyhow.
He said it wouldn't be REALLY READY until September. I'd agree with that -- if i didn't think it was a gross underestimate. I can't see the masses of Mac users REALLY making the leap to OS X for another year at least.
Yeah, my Mac can anti-alias fonts at the OS level if you want it to. But personally, I think it ends up looking like blurry crap. Most of the time, I turn off the anti-aliasing.
Thing is, it's possible to do that on the Mac OS because the fonts don't look like hell when they're not anti-aliased. I don't really care about alpha channel blending etc... why is it that the default X courier font always looks like it got smashed in a garbage masher? Why do serif fonts always seem to look LESS legible under X (exactly the opposite of the reason serifs were put there to begin with)?
Everybody in the font-design community always bitches about how often fonts are ripped off, how you can't make any money selling fonts. So if you're resigned to that, font designers, then why not work on some good, legible, Open Source fonts that won't look like ass under X?
I've been looking into this recently myself. As far as I understand it, the only way to be "entitled to citizenship" is to either be born in the UK yourself, or have both parents be UK citizens and be born outside the UK because of some reason (military service etc.)
As you say, though, a grandparent who is a UK citizen can get you the "Right of Abode" -- provided you can document it, of course. If you are a Commonwealth citizen who has one parent who is a UK citizen by birth, you can also be entitled to the "Right of Abode."
This right is essentially the equivalent of a Green Card in the US, though I don't think it ever qualifies you for full-on Citizenship, no matter how long you live there, unlike the US.
We have stuff like FreeBSD binary support etc. for Linux/Intel. So how long, I wonder, before someone figures out how to bring OS X binary support to PPC Linux?
From this article, I can't tell that they're emulating anything more than the network layer on each platform. Where does it say anything about porting games to different platforms? Someone who speaks Japanese has to read the original source and explain what's going on.
Doesn't sound like much added security to me, for the effort. So you have your password-changing script go in with lynx and confirm all the changes -- how hard is that?
Even so, the way I have it currently set up is a problem. Someone could grab a copy of the local encrypted cookie, then use it to connect as the user from then on. The easiest way I can think of to solve this is to have the cookie be a combination of a timeout value and the encrypted pass, and store that value in the db as well ('till timeout) but even so, at least user passwords can't be read out of the database in my current setup.
Why do you need to store the encrypted password in your cookie to begin with? You could send the password once. If it matches the (encrypted) password stored in the database, the user is authenticated.
Once you determine they've got a valid password, you generate a cookie for them that includes a few different pieces of data. Say, you get one server-side "password" value (that the user, presumably, does not know); the user's IP address (from the server's environment); the user's login id; and a timeout value -- and you hash them all together with MD5.
From then on, you rebuild the cookie on the server side every time the user reloads a page, and if it matches the cookie sent by the client's browser, the assumption is that they've been authenticated. They can't steal a cookie from someone else's machine, because the IP address is embedded in the cookie. They can't trick someone into using their machine to log in to a valid account and then steal that cookie for their own account, because the userid is embedded in there. And they wouldn't be able to just forge a cookie, even if they had somehow figured out what the format of the cookie is, so long as they didn't know what the hidden server-side value is.
I Am Not A Security Expert, but it sounds good to me -- provided we're not talking about security for a bank or something here.
Doesn't solve the problem of forgotten passwords, of course.
The thing that's funniest about C|Net reporting on this is that it's almost universal consensus amongst Web/IT workers in and around San Francisco that C|Net is one of the worst possible places you could ever work.
I've met a lot of people who have worked there, and I've never met one who came away happy about it.
Slashdot Mirror
The iPaq people may be OSS-friendly ... that Web site you mention is meant to encourage open source software that runs on handhelds. But all I really do with my handheld (a Palm) is move data to and from it. When will I be able to do that with an iPaq?
I use a PowerBook G3, triple booting Mac OS 9.1, Mac OS X beta, and LinuxPPC. Which one is my best choice for syncing my iPaq?
--
It's not as simple as that. I know on my LinuxPPC system, the fonts look like hell. And that's even the TrueType fonts that I, um, ripped off from the Mac OS and serve using xfstt.
The fonts under Linux are fairly legible -- provided you find the one or two point sizes that actually look good, so the loops aren't closing up, etc. But under the Mac OS, ALL the point sizes look fine!
Maybe it has something to do with the fonts for the Mac being designed for a 72 dpi screen resolution, while X11 is designed at 75 dpi? But I thought TrueType was supposed to solve the resolution-dependence problem...
Truly perplexing, this rotten X fonts thing.
--
Power Comuting is a formidable enemy? The people whose primary contributions to the Mac hardware platform were unsupported OEM CD-ROM drives, PS/2 keyboards, and the standard VGA port? (Okay, I'll give you the last one -- but Apple even ships those, now.)
I hear this argument a lot. I don't get it. Why would opening up Apple's platform lead to more support? The world already has Linux. Linux is much more mature on Intel hardware than on the Mac/PowerPC. Why would having a mature, open source OS that ran on Mac hardware make people want to concentrate their efforts on the Mac instead of Intel? Even if Apple made the Mac ROMs public domain, why would that inspire people to use them instead of the Intel platform, which currently has a larger installed base than any other form of PC hardware? Especially if you're talking about the consumer market (Apple's primary targer)The only reason Power Computing met with any success at all in the Mac clone market was because they sensed there was demand for lower-cost hardware that ran this closed source OS. Take away the closed source OS, and the need for the "clone" hardware goes away too.
The only reason for an open source Mac OS is because you don't think Apple's hardware is being used well enough. Take away Apple's hardware, and the need for the OS disappears as well.
What Apple is doing now with OS X is continuing to do what it does best -- try to provide the consumer with an all-inclusive computing platform. You buy Apple's hardware, you get Apple's OS. It's a packaged product. Just because it isn't currently well-suited for tinkering, doesn't mean it should be.
--
Come on, now. This OS won't go mainstream for another year, at least.
--
Actually, I thought Apple's Classic environment was supposed to take care of this problem.
OK, OK. For someone who's actually used the Classic environment, I realize this comment may come off as somewhat facetious.
--
Ah! Now it makes sense.
This is what Sun Microsystems is talking about, when they're saying they're going to put chips into refrigerators and run waffle irons on Java.
--
Stuff like this happens all the time. Go back through a few years of magazines and look for any ads, etc. which reference the pro wrestler Hulk Hogan. You may find that at the bottom of the ad there's a legend stating that the name "Hulk Hogan" is TM & Copyright Marvel Comics.
... and Slashdot should probably leave its interpretation to the experts.
Why? Not because Marvel Comics invented the Hulk Hogan persona. No, it's because Marvel Comics has a character who's a big strong guy who's named the Hulk. Hulk Hogan was (arguably) a fictitious character. He was a big, strong guy. Ergo, in Marvel Comics's eyes, his use of that name was infringing on their trademark on big strong guys with this particular word of the English language in their name.
Now, let's take stock here: Hulk Hogan could certainly not bench press 18 tons. Hulk Hogan was certainly not green, and he didn't undergo any particular transformation when he got angry. It could be argued that his use of the word "hulk" was just plain English language, and nothing more.
Then again, you could argue that it WAS something more. And ultimately, somebody decided it was easier (and better for Hogan's career) if they let Marvel claim the trademark and Hogan just went on about his business of being a wrestler.
Nothing really more clandestine or sinister than that. Just like in this Blizzard case. It's just plain ol' trademark law, contract law, entertainment law, and what-have-you
--
So when you say something "will cost more," what are you talking about? Two McDonald's cheeseburgers will always cost more than one cheeseburger, true. But a two-cheeseburger meal costs less than buying one cheeseburger, a Coke, and some fries. It still "costs more" to get the two cheeseburgers. But it's a better deal.
If you then resell each cheeseburger to somebody at cost, then those people get a cheaper cheeseburger. Alternately, if you mark up the price, you make profit and then you can buy more cheeseburgers. That's about as "well accepted" an "economic principle" as you can get!
I think you hit on it more with this statement:
To me, it seems like it's practically to the point where there's no difference between ordering a $900/month 1.54Mbps Frame Relay circuit, and ordering a $40/month DSL circuit. It used to be that a Frame Relay reseller could tell their customers that they'd be getting better quality of service guarantees for that extra money; nowadays, there's almost no justification.
Part of the problem has been the whole DSL market share war, which has driven DSL prices artificially low, and in turn impacted the entire industry. And besides that, I'm sure other people can come up with a full laundry list of telco incompetences.
So is the cost of online access going to go up? Most definitely. And not because people are "using it too much" (with P2P etc.) -- it will go up because the telcos can barely cut costs any more than they have already, and they're still as hungry for profits as they've ever been.
But will the money go to the right places, that's the real question. What we should really be worrying about is the price of access continuing to go up while the quality of service continues to go down.
--
...so-called "hardcore programming" jobs aren't that easy to come by on a contract basis. On the other hand, Web jobs aren't too difficult to pick up. If server-side Java meets your qualifications for "programming," then why not try that? A lot of work in that field around here (Bay Area).
They already did the Mule saga. It was called "The Usual Suspects."
Born Jean Giraud, he is most famous in Europe for his Western character, Blueberry, whose exploits Giraud signs with his given name. Blueberry still appears in various ongoing series of graphic novels.
In the US, however, Giraud is most widely-known by the pseudonym Moebius. He is the author of many popular comics stories in the science fiction genre, including the Arzach series, as well as The Airtight Garage, which was used as the inspiration for the video arcade in San Francisco's Sony Metreon entertainment center.
Moebius has also done production design work on some science fiction films, including Alien, Tron, and The Abyss. The latter was a rotten movie, but it had a visually interesting version of what alien spacecraft and technology might look like. Should work well for Rama.
See? You can learn something new every day.
Wonder why he didn't?
Hey ... you smell something?
Aw, complain all you want about not having eighteen mouse buttons on the Titanium PowerBook. Apple obviously feels that a single button is a nicer interface, and I for one am not going to say they're wrong.
What I thought was interesting about Mac OS X Public Beta, on the other hand, is that it AUTOMATICALLY supported the top two buttons of my Kensington 4-button mouse, out of the box. The second button worked the contextual menus, as expected.
And this is an ADB mouse, mind you -- old technology from Apple (all the new mice are USB), and NOT code that was built in to OpenStep!
What this tells me is that Apple is going to lengths to preserve your possible preference, even though the preference of their hardware engineers is something else. As other people have pointed out, if you want more buttons, feel free to plug in a different mouse. I don't know what more you can really ask for.
A lot of DVD players don't support CD-R discs. Don't ask me why it's so; but I don't think it has anything to do with "piracy prevention," it's more by-product of the way the laser picks up data from the discs on DVD players.
My Pioneer DV-606D doesn't support CD-Rs either, for what it's worth.
I've actually been wanting a list of players that DO support CD-Rs for a while now. (Besides playing audio CDs, it's useful for VCDs.)
As near as I can tell, the deal with these Pinball 2000 machines is that they have NO moving parts. It's a last-ditch attempt to make an economically viable pinball machine. The ball slides around, interacting with magnets (?) while the targets are all "virtual," displayed on a video screen below.
Wait a minute ... who are these tunnel-vision Mac users you claim have moved to LinuxPPC, anyway? Have you ever met one? If they won't use Windows, what in hell makes you think they'd use Linux?
And since when has the sole purpose of LinuxPPC development been to let Linux run on Macs, anyhow?
I honestly don't think you have the slightest clue what you're talking about. But keep on feeling superior anyhow.
He said it wouldn't be REALLY READY until September. I'd agree with that -- if i didn't think it was a gross underestimate. I can't see the masses of Mac users REALLY making the leap to OS X for another year at least.
Is anti-aliasing really the answer?
Yeah, my Mac can anti-alias fonts at the OS level if you want it to. But personally, I think it ends up looking like blurry crap. Most of the time, I turn off the anti-aliasing.
Thing is, it's possible to do that on the Mac OS because the fonts don't look like hell when they're not anti-aliased. I don't really care about alpha channel blending etc... why is it that the default X courier font always looks like it got smashed in a garbage masher? Why do serif fonts always seem to look LESS legible under X (exactly the opposite of the reason serifs were put there to begin with)?
Everybody in the font-design community always bitches about how often fonts are ripped off, how you can't make any money selling fonts. So if you're resigned to that, font designers, then why not work on some good, legible, Open Source fonts that won't look like ass under X?
I've been looking into this recently myself. As far as I understand it, the only way to be "entitled to citizenship" is to either be born in the UK yourself, or have both parents be UK citizens and be born outside the UK because of some reason (military service etc.)
As you say, though, a grandparent who is a UK citizen can get you the "Right of Abode" -- provided you can document it, of course. If you are a Commonwealth citizen who has one parent who is a UK citizen by birth, you can also be entitled to the "Right of Abode."
This right is essentially the equivalent of a Green Card in the US, though I don't think it ever qualifies you for full-on Citizenship, no matter how long you live there, unlike the US.
Good! I was sick of hearing from all these "K-Elite Juarez D00DZ" anyway.
We have stuff like FreeBSD binary support etc. for Linux/Intel. So how long, I wonder, before someone figures out how to bring OS X binary support to PPC Linux?
From this article, I can't tell that they're emulating anything more than the network layer on each platform. Where does it say anything about porting games to different platforms? Someone who speaks Japanese has to read the original source and explain what's going on.
Doesn't sound like much added security to me, for the effort. So you have your password-changing script go in with lynx and confirm all the changes -- how hard is that?
Why do you need to store the encrypted password in your cookie to begin with? You could send the password once. If it matches the (encrypted) password stored in the database, the user is authenticated.
Once you determine they've got a valid password, you generate a cookie for them that includes a few different pieces of data. Say, you get one server-side "password" value (that the user, presumably, does not know); the user's IP address (from the server's environment); the user's login id; and a timeout value -- and you hash them all together with MD5.
From then on, you rebuild the cookie on the server side every time the user reloads a page, and if it matches the cookie sent by the client's browser, the assumption is that they've been authenticated. They can't steal a cookie from someone else's machine, because the IP address is embedded in the cookie. They can't trick someone into using their machine to log in to a valid account and then steal that cookie for their own account, because the userid is embedded in there. And they wouldn't be able to just forge a cookie, even if they had somehow figured out what the format of the cookie is, so long as they didn't know what the hidden server-side value is.
I Am Not A Security Expert, but it sounds good to me -- provided we're not talking about security for a bank or something here.
Doesn't solve the problem of forgotten passwords, of course.
The thing that's funniest about C|Net reporting on this is that it's almost universal consensus amongst Web/IT workers in and around San Francisco that C|Net is one of the worst possible places you could ever work.
I've met a lot of people who have worked there, and I've never met one who came away happy about it.