I had an even worse experience setting up a DSL connection for one of the directors at my company. He bought two iBooks and wanted to use an AirPort base station to connect to the DSL modem. Because Bell Atlantic uses PPP over Ethernet instead of static IP addressing, there was no way to get their horribly broken proprietary client software to work with a Mac OS 9 iBook.
I went back and forth with them every day for over THREE MONTHS with Bell Atlantic never coming up with an acceptable solution. Finally I gave up and decided to go with another provider. Bell Atlantic told me that it was illegal for them to give me a list of other providers and that they had no idea who might be reselling service over their lines. My protests that this would make billing a little more difficult on their end didn't seem to get through to them.
Finally, I got in touch with another provider. I bounced around their voice mail system for another two weeks before they finally admitted that they were only going to accept customers from Covad because Bell Atlantic was impossible to deal with.
After a delay of several more weeks, I finally got the service running (with a static IP address) from an outfit run by two Russian guys working out of their apartment. Not exactly world class customer service, but at least the line has worked since then.
In all, it took close to six months between the time the order was placed and when the first packet went through the line. I spent hundreds of hours working on this and ranted at Bell Atlantic until I was blue in the face (their CEO never did reply to my emails).
Short answer: I would *NEVER AGAIN* recommend that anyone buy Internet access from Bell Atlantic. I don't know if other telcos are as bad, but this experience was terrible.
The Nitro was kind of the legendary holy grail of NeXT computing.
Here's a site with good info on the Nitro board:
http://www.channelu.com/NeXT/Black/Nitro/
Sam Goldberger of Spherical Solutions hacked up a 60Mhz board called the Pyro that he was selling from his site for a long time. I don't know where he is now, orb.com seems to have been taken over by the aggressively commercial orb.net.
NetBSD should run on your cube, and there is a 68k linux project out there, but I'm not sure if they are supporting the NeXT hardware or not.
I have used my 030 cube for ages, it made a great web server, AppleShare file server, and everything else I ever asked it to do.
Considering that the text file with the CC numbers was posted to usenet repeatedly around that time, by your logic male sopranos should be making a comeback in opera companies, right?
He was never found to have used any of those credit card numbers.
Reading through this thread, I'm really surprised that so many slashdot readers are not familiar with the details of Mitnick's case. If you read the trial documents, you will see that what he was charged with was *lying on the telephone*.
That's right, the Master Hacker's uber-leet k-rad tactic consisted primarily of calling up the help desk and saying, "Hi this is Bob. I forgot my password, could you remind me again?"
He then used the passwords to get onto internal servers and download source code. As far as anyone knows, he never redistributed the source or used it for personal gain.
Yes, he misrepresented himself and took source code that was not available to the public. He downloaded the source to Solaris and some cell phone switching software. If he had instead enrolled in any community college, Sun would have sold him a student source code license for $100. If he had waited until now, he could have downloaded the source for free!
And for this he gets five years in prison? He agreed to the plea bargain because he was still in jail two years after his sentence was up and it looked like the government was planning on keeping him in jail indefinitely until he admitted to something, anything.
I'm not in favor of malicious hacking, or destructive behavior in general, but I am favor of some sense of justice in a justice system. What Kevin did was unethical and wrong, but should he really have spent more time in jail than many people convicted of armed robbery, rape, or murder?
Interesting precedent. If Markoff and Shimomura were convicted of entrapment, would that mean that they would have to pay back the million bucks they got for Takedown?
I think they are planning this already. Ever since Gates decided to step down as CEO and become "chief software architect". It looks to me like they are anticipating being broken in three along the following lines:
They know it will be illegal to collaborate *after* the split, so my guess is that the busy beavers in Redmond are working like crazy collaborating in anticipation of the split. It would not surprise me if they have already written business plans (and marketing plans, coming from MS) to last them the next ten years.
Now that Microsoft is officially an illegal monopoly, and we now know that they intentionally put this back door in their software, it only reinforces that consumers actually are hurt by Microsoft's poor judgement.
If we can demonstrate harm done from this, I think that MS just might find themselves facing a huge class action lawsuit.
There was a huge thread about this on one of the security mailing lists recently. There was a lot of debate about whether or not a honeypot was entrapment. The short answer is maybe. In some areas, and for some agencies (like the US military) honeypots are considered an illegal form of entrapment. If you are thinking at all about implementing one, check with you local authorities first.
You also need to take a good look at your security policy and determine what your security goals are. For most businesses, trying to catch the person who rooted your box is a secondary goal at best. The most important thing is to get the systems back on line and minimize the downtime. A honeypot only makes sense if you are trying to gather information that you will use to try to prosecute the attacker.
Getting decent evidence that will be admissable in court is extremely diffcult, so many people don't really try. For more information on gathering forensic evidence, check out this PDF from a recent SANS conference.
http://www.sans.org/TALKS/KRUSE.PDF
YMMV, but in my own opinion, the time and effort you put into a honeypot would be better spent securing your actual boxes.
Yuck, I hate to blow my own horn, but this is exactly the thing I was trying to get across in the Katz article from yesterday (http://slashdot.org/comments.pl?sid=00/03/17/1662 37&cid=66).
A micropay per play model is the best thing I have come up with yet to allow people to freely distribute music files and guarantee payment to the musicians.Just selling MP3 singles doesn't do much good if the files are going to be widely copied and distributed.
I will be the first one to say that relying on a continual stream of micropayments is not perfect, and there are things about it that I strongly dislike. If anyone has come up with a better way of resolving this problem, I would really like to hear about it.
I'm the person quoted here in Katz's article. One thing that he didn't say (probably because I didn't make it clear enough (or because babelfish still lacks a geek to reporter translater)), is that this was a thought experiment, not a business plan.
What I was trying to do was imagine a system where musicians who chose to do so could distribute music online without using a record company and without fear of losing royalties. I know that these are not the terms everyone works under, but these were the givens in the thought experiment.
I came up with the idea of using micropaymets every time the recording is played. This way, musicians would encourage the use of napster and trading files since playing the file itself would generate the revenue. Both the musician and the listener would benefit.
The more I thought about it, I liked some aspects and strongly disliked others. The idea of cumulative payments appeals to me. A song I really like is worth a lot to me and I am happy to see the performer get paid in direct proportion to my appreciation (this is another variation of the street performer idea, you throw more in the hat for a song you like).
I did not like the fact that it would require breaking the MP3 format and would open up a bunch of security issues. The security problems might not be that bad, as I imagined that the debit accounts would be very small; a $5 deposit would get you 5000 plays. At worst, if someone cracked the player we would only be back where I am.
Sorry about the repetition, but I wanted to make this part perfectly clear: I have no plans to wrap myself in an open source, copyleft, or any other kind of flag and I have absolutely no intention of taking this idea any further than an intellectual exercise.
I do consider the open source and copyleft ideals very important to me, so thank you 348 for being so quick to defend them!
The company and its products now take on the name LEGO, which is formed from the Danish words "LEg GOdt" ("play well"). Later, it is realised that in Latin the word means "I study", "I put together".
And here's the link:
http://www.lego.com/info/history/hist3259.asp
I guess this means we are both right, and I lambast you with a pedantic "neener neener nee-ner".
Have you tried using keykit? It is a MIDI tool with its own C-like programming language. Interesting stuff and available for Linux, Mac, and Microsoft OSes.
I had looked for tools to use in algorithmic composition for years before finding keykit. It works quite differently than some commercial MIDI packages, but for the experimenter/MIDI hacker it is a very useful tool
I think you're on to something. What about these services that offer small payments in exchange for surfing the net and letting some company record the clickstream?
I don't use these services because for me, the $20/mo or whatever is not worth the loss of privacy (plus they don't run on Linux ), but what about in a place where the average income is $20/mo? You could get the equivalent of a full time wage for sitting at home surfing the net, learning about programming or whatever to bootstrap your skills to the next level.
This has potential. It could really change the life of some aspiring young geek.
I actually thought about this quite a bit back in the early 90s when I was seriously thinking about opening an ISP in Mongolia. The nice people at Magic Net (http://www.magic.mn/) beat me to it though.
At the time, my plan was to initially sell access to the government, foreign corporations with offices in the country, and scientific research stations. I even had some prototype packet radio units that could be used on fossil digs.
The surprising thing is that many of these ISP that have popped up in places you least expect are thriving. They keep their equipment costs low and they don't have the super high customer acquisition costs that burn so much money for providers in the US. I don't know how profitable they are, but I think that they are probably doing better than most ISPs in developed countries.
Starting an ISP in a country with no telephone or network infrastructure is not for the easily intimidated. Clearly there are many places in the world where people are more concerned with survival than the color scheme in the YRO section of Slashdot. Still, starting an ISP in a developing country may provide a positive jump start in a way that would not happen otherwise.
I read the other day that Charles Schwab has installed 6 mainframes in the past year and that they are taking new ones as fast as the vendors can deliver them.
There are still some places where the dirt cheap PC style boxes have not infiltrated. As many people have pointed out, mainframes can handle much higher transaction volumes and have much better uptime figures than any other kind of system.
Also, remember that many of the outfits that use mainframes have huge amounts of legacy code in production that might not port so easily to a beowolf cluster, or anything else.
The costs are huge but you need to put them in perspective. I was talking with one of the local Sun techs in my area and he had recently been out to work on an Starfire that a client kept as a hot spare. This is a $2 million dollar machine and it's just the backup! The tech thought it was overkill but the client, a major brokerage house, does $4 million an hour in transactions. The spare machine would pay for itself ini preventing only 30 minutes of downtime.
For the organizations that really need them, the big iron is still worth every penny.
>2) Have software that warns you about every change to your system.
Is there anything equivalent to tripwire or tcpwrappers for Windows? For *nix based systems, these are indispensible tools to track the integrity of your systems.
At a securty conference I went to about a year ago, everyone was complaining that they were not available for win32 systems. Has that changed since then?
Isn't an affiliate program really just a commission sales model? The idea is that if the affiliates deliver customers who buy merchandise, they get paid.
Seems to me you wouldn't need to turn over too many rocks to come up with prior art on that one.
I think a standard desktop could be a real detriment to Linux. We are seeing Linux adopted in embedded systems, handhelds, supercomputers, and desktop machines. An interface that works for one of these applications may not work at all for another.
That's the whole reason behind the failing of WinCE and Windows for Pen-Based Computing before it. The Windows interface does not transfer well to a handheld and that frustrates users much more than learning an "unfamiliar" interface like the Palm OS.
Having some apps that work with KDE and others that work only with Gnome can be a pain. However, I think we should be careful not to tie Linux too closely to a single interface paradigm in case it causes problems for the whole range of Linux devices.
I have wondered about this as well. It seems to me that one of the real strengths of the open source model is that it frees developers from having to meet artificial ship dates in order to sell a bunch of upgrades to inflate the quarterly sales figures.
Yes, having a nifty one click installer would be realy nice. So would having a kernel that scales well above four processors, a free database with a full SQL implementation, a journaling file system, or any of a thousand other technical goals.
The real difference between Linux and a commercial OS is that Linux developers have the freedom to get the tech right. MS or any other commercial software house needs to ship releases quickly to keep the operation going, even if that means releasing half baked early beta quality software as a 1.0 release.
Easy installs and all the rest are worthy goals, but I don't think that technical goals should be sacrificed for the sake of aesthetic ones.
While I have certainly seen my share of ugly exchanges between Linux users, a lot of what comes off as anger is ideological zeal.
Some people who use and develop Linux do so because it is the right tool for the job, or because it is efficient, or because it doesn't cost anything, or other equally valid reasons. You cannot forget that there is also a large population of people who use and code for Linux because they feel it is the *right* thing to do.
Other platforms don't have this, and it is a huge difference. Can you imagine a Windows user saying that running Windows is morally superior to other types of computers? Even smug Mac users (and I speak as a longtime smug Mac user) feel a deep technical superiority and a sense of oppression from the MS juggernaut, but I don't think any of them thinks using a Mac makes one a better person.
Some of the people who have slogged away for years bringing Linux to its current level have this sense of occupying the moral high ground. RMS is the obvious example.
Linux means a lot to these people, much more than Windows does to an MCSE. It is unavoidable that they are going to have strong feelings and that, sometimes, this will look like spleen venting. Maybe there is a parallel with religion. Some religious people might respond to a precieved insult to their beliefs very badly. It doesn't excuse rude behaviour but the motivation is much deeper than a simple flame.
The part I was most impressed by was the fact that BootX is not required. I think BootX is a great tool (and that Ben Herrenschmidt is One Cool Guy for writing it), but I like the idea of not needing a Mac OS partition.
It looks like IBM is finally getting their POP board spec finished. With any luck we should be seeing dirt cheap non-Apple branded hardware before too long.
LinuxPPC has been working with IBM on software for the POP systems, so it is not too surprising to see their distro targeted towards these machines. However, YDL has some kind of a contract with IBM, so I would not be surprised to see them supporting the new hardware pretty soon as well.
Well done, LinuxPPC! If I weren't downloading an ISO of mkLinux right now, I'd be grabbing the new release already.
I think Katz has it all backwards. Yes, many geeks were picked on in school and some may not have Emily Post approved social skills, but this is not what makes up geeks. Reading the first few thousand words of the excerpt he seems to be arguing that social isolation is what causes geekdom. In fact the opposite is true. It is the focus on technology that causes the distance from "normal" society.
When I was 12 years old, the only thing I cared about was hacking my TRS-80. I wrote thousands and thousands of lines of code for that beast because it was what I loved to do. It took precedence over playing sports or going to dances at school or other things that some other kids were into. But the fact that I was not interested in their Lord of the Flies style social Darwinism did not make me code better.
I think Katz fails to appreciate that we are differenct because we are different. The societal pressure didn't make us learn assembly language. This is the same reason he wants to lump artists and other nonconformists together with geeks. He sees the external similarities between groups of social outcasts, but misses the essence of geekdom entirely.
I don't want to seem to harsh on him, though. At least Katz is putting forward the effort to try and understand who geeks are and what makes us tick. He's not there yet, but at least he's trying, and that's a lot more than I can say about most non-geeks I encounter.
Think of the orbit as a bicycle wheel. The earth is at the hub and the orbits are at different points along the spokes.
As the wheel spins, the parts that are farther away from the hub havce to travel farther to make a complete revolution, but since the wheel turns as a unit, the outer points have to turn faster than the inner points.
The difference between a satellite in orbit and a bicycle wheel is that the satellite is being pulled toward the Earth. To calculate a geosynchronous orbit, you find the point where the speed needed to keep from crashing back to earth is the same as the speed needed to match the Earth's rotation. It turns out that the point where these two match is at an altitude of about 30,000km.
At 675km, the gravitational attraction is much stronger and a satellite needs to orbit at a higher speed to keep aloft. If you slowed the orbital speed to match the rotation of the Earth you would need to constantly burn fuel to counteract the force of gravity.
That's pretty much how I would explain it to my four year old. I hope it makes sense.
Slashdot Mirror
I had an even worse experience setting up a DSL connection for one of the directors at my company. He bought two iBooks and wanted to use an AirPort base station to connect to the DSL modem. Because Bell Atlantic uses PPP over Ethernet instead of static IP addressing, there was no way to get their horribly broken proprietary client software to work with a Mac OS 9 iBook.
I went back and forth with them every day for over THREE MONTHS with Bell Atlantic never coming up with an acceptable solution. Finally I gave up and decided to go with another provider. Bell Atlantic told me that it was illegal for them to give me a list of other providers and that they had no idea who might be reselling service over their lines. My protests that this would make billing a little more difficult on their end didn't seem to get through to them.
Finally, I got in touch with another provider. I bounced around their voice mail system for another two weeks before they finally admitted that they were only going to accept customers from Covad because Bell Atlantic was impossible to deal with.
After a delay of several more weeks, I finally got the service running (with a static IP address) from an outfit run by two Russian guys working out of their apartment. Not exactly world class customer service, but at least the line has worked since then.
In all, it took close to six months between the time the order was placed and when the first packet went through the line. I spent hundreds of hours working on this and ranted at Bell Atlantic until I was blue in the face (their CEO never did reply to my emails).
Short answer: I would *NEVER AGAIN* recommend that anyone buy Internet access from Bell Atlantic. I don't know if other telcos are as bad, but this experience was terrible.
The Nitro was kind of the legendary holy grail of NeXT computing.
Here's a site with good info on the Nitro board:
http://www.channelu.com/NeXT/Black/Nitro/
Sam Goldberger of Spherical Solutions hacked up a 60Mhz board called the Pyro that he was selling from his site for a long time. I don't know where he is now, orb.com seems to have been taken over by the aggressively commercial orb.net.
NetBSD should run on your cube, and there is a 68k linux project out there, but I'm not sure if they are supporting the NeXT hardware or not.
I have used my 030 cube for ages, it made a great web server, AppleShare file server, and everything else I ever asked it to do.
Considering that the text file with the CC numbers was posted to usenet repeatedly around that time, by your logic male sopranos should be making a comeback in opera companies, right?
He was never found to have used any of those credit card numbers.
Reading through this thread, I'm really surprised that so many slashdot readers are not familiar with the details of Mitnick's case. If you read the trial documents, you will see that what he was charged with was *lying on the telephone*.
That's right, the Master Hacker's uber-leet k-rad tactic consisted primarily of calling up the help desk and saying, "Hi this is Bob. I forgot my password, could you remind me again?"
He then used the passwords to get onto internal servers and download source code. As far as anyone knows, he never redistributed the source or used it for personal gain.
Yes, he misrepresented himself and took source code that was not available to the public. He downloaded the source to Solaris and some cell phone switching software. If he had instead enrolled in any community college, Sun would have sold him a student source code license for $100. If he had waited until now, he could have downloaded the source for free!
And for this he gets five years in prison? He agreed to the plea bargain because he was still in jail two years after his sentence was up and it looked like the government was planning on keeping him in jail indefinitely until he admitted to something, anything.
I'm not in favor of malicious hacking, or destructive behavior in general, but I am favor of some sense of justice in a justice system. What Kevin did was unethical and wrong, but should he really have spent more time in jail than many people convicted of armed robbery, rape, or murder?
Interesting precedent. If Markoff and Shimomura were convicted of entrapment, would that mean that they would have to pay back the million bucks they got for Takedown?
I think they are planning this already. Ever since Gates decided to step down as CEO and become "chief software architect". It looks to me like they are anticipating being broken in three along the following lines:
%micro_split {
'windows' => 'gates'
'office' => 'ballmer'
'internet' => 'beluzzo'
}
They know it will be illegal to collaborate *after* the split, so my guess is that the busy beavers in Redmond are working like crazy collaborating in anticipation of the split. It would not surprise me if they have already written business plans (and marketing plans, coming from MS) to last them the next ten years.
Now that Microsoft is officially an illegal monopoly, and we now know that they intentionally put this back door in their software, it only reinforces that consumers actually are hurt by Microsoft's poor judgement.
If we can demonstrate harm done from this, I think that MS just might find themselves facing a huge class action lawsuit.
Slashdot (et al.) v. Microsoft, anyone?
No kidding, we don't those ciminals with their sommelier's spoons, white towels, and silver crumbers corrupting the Internet!
There was a huge thread about this on one of the security mailing lists recently. There was a lot of debate about whether or not a honeypot was entrapment. The short answer is maybe. In some areas, and for some agencies (like the US military) honeypots are considered an illegal form of entrapment. If you are thinking at all about implementing one, check with you local authorities first.
You also need to take a good look at your security policy and determine what your security goals are. For most businesses, trying to catch the person who rooted your box is a secondary goal at best. The most important thing is to get the systems back on line and minimize the downtime. A honeypot only makes sense if you are trying to gather information that you will use to try to prosecute the attacker.
Getting decent evidence that will be admissable in court is extremely diffcult, so many people don't really try. For more information on gathering forensic evidence, check out this PDF from a recent SANS conference.
http://www.sans.org/TALKS/KRUSE.PDF
YMMV, but in my own opinion, the time and effort you put into a honeypot would be better spent securing your actual boxes.
Yuck, I hate to blow my own horn, but this is exactly the thing I was trying to get across in the Katz article from yesterday (http://slashdot.org/comments.pl?sid=00/03/17/1662 37&cid=66).
A micropay per play model is the best thing I have come up with yet to allow people to freely distribute music files and guarantee payment to the musicians.Just selling MP3 singles doesn't do much good if the files are going to be widely copied and distributed.
I will be the first one to say that relying on a continual stream of micropayments is not perfect, and there are things about it that I strongly dislike. If anyone has come up with a better way of resolving this problem, I would really like to hear about it.
I'm the person quoted here in Katz's article. One thing that he didn't say (probably because I didn't make it clear enough (or because babelfish still lacks a geek to reporter translater)), is that this was a thought experiment, not a business plan.
What I was trying to do was imagine a system where musicians who chose to do so could distribute music online without using a record company and without fear of losing royalties. I know that these are not the terms everyone works under, but these were the givens in the thought experiment.
I came up with the idea of using micropaymets every time the recording is played. This way, musicians would encourage the use of napster and trading files since playing the file itself would generate the revenue. Both the musician and the listener would benefit.
The more I thought about it, I liked some aspects and strongly disliked others. The idea of cumulative payments appeals to me. A song I really like is worth a lot to me and I am happy to see the performer get paid in direct proportion to my appreciation (this is another variation of the street performer idea, you throw more in the hat for a song you like).
I did not like the fact that it would require breaking the MP3 format and would open up a bunch of security issues. The security problems might not be that bad, as I imagined that the debit accounts would be very small; a $5 deposit would get you 5000 plays. At worst, if someone cracked the player we would only be back where I am.
Sorry about the repetition, but I wanted to make this part perfectly clear: I have no plans to wrap myself in an open source, copyleft, or any other kind of flag and I have absolutely no intention of taking this idea any further than an intellectual exercise.
I do consider the open source and copyleft ideals very important to me, so thank you 348 for being so quick to defend them!
Here is the quote from Lego's web site:
The company and its products now take on the name LEGO, which is formed from the Danish words "LEg GOdt" ("play well"). Later, it is realised that in Latin the word means "I study", "I put together".
And here's the link:
http://www.lego.com/info/history/hist3259.asp
I guess this means we are both right, and I lambast you with a pedantic "neener neener nee-ner".
The plural of Lego is Legamus. Lego comes from the Latin verb "I play". Legamus "we play" is the natural plural.
Maybe it just doesn't look as good on the box?
I had looked for tools to use in algorithmic composition for years before finding keykit. It works quite differently than some commercial MIDI packages, but for the experimenter/MIDI hacker it is a very useful tool
I think you're on to something. What about these services that offer small payments in exchange for surfing the net and letting some company record the clickstream?
I don't use these services because for me, the $20/mo or whatever is not worth the loss of privacy (plus they don't run on Linux ), but what about in a place where the average income is $20/mo? You could get the equivalent of a full time wage for sitting at home surfing the net, learning about programming or whatever to bootstrap your skills to the next level.
This has potential. It could really change the life of some aspiring young geek.
I actually thought about this quite a bit back in the early 90s when I was seriously thinking about opening an ISP in Mongolia. The nice people at Magic Net (http://www.magic.mn/) beat me to it though.
At the time, my plan was to initially sell access to the government, foreign corporations with offices in the country, and scientific research stations. I even had some prototype packet radio units that could be used on fossil digs.
The surprising thing is that many of these ISP that have popped up in places you least expect are thriving. They keep their equipment costs low and they don't have the super high customer acquisition costs that burn so much money for providers in the US. I don't know how profitable they are, but I think that they are probably doing better than most ISPs in developed countries.
Starting an ISP in a country with no telephone or network infrastructure is not for the easily intimidated. Clearly there are many places in the world where people are more concerned with survival than the color scheme in the YRO section of Slashdot. Still, starting an ISP in a developing country may provide a positive jump start in a way that would not happen otherwise.
I read the other day that Charles Schwab has installed 6 mainframes in the past year and that they are taking new ones as fast as the vendors can deliver them.
There are still some places where the dirt cheap PC style boxes have not infiltrated. As many people have pointed out, mainframes can handle much higher transaction volumes and have much better uptime figures than any other kind of system.
Also, remember that many of the outfits that use mainframes have huge amounts of legacy code in production that might not port so easily to a beowolf cluster, or anything else.
The costs are huge but you need to put them in perspective. I was talking with one of the local Sun techs in my area and he had recently been out to work on an Starfire that a client kept as a hot spare. This is a $2 million dollar machine and it's just the backup! The tech thought it was overkill but the client, a major brokerage house, does $4 million an hour in transactions. The spare machine would pay for itself ini preventing only 30 minutes of downtime.
For the organizations that really need them, the big iron is still worth every penny.
>2) Have software that warns you about every change to your system.
Is there anything equivalent to tripwire or tcpwrappers for Windows? For *nix based systems, these are indispensible tools to track the integrity of your systems.
At a securty conference I went to about a year ago, everyone was complaining that they were not available for win32 systems. Has that changed since then?
Isn't an affiliate program really just a commission sales model? The idea is that if the affiliates deliver customers who buy merchandise, they get paid.
Seems to me you wouldn't need to turn over too many rocks to come up with prior art on that one.
I think a standard desktop could be a real detriment to Linux. We are seeing Linux adopted in embedded systems, handhelds, supercomputers, and desktop machines. An interface that works for one of these applications may not work at all for another.
That's the whole reason behind the failing of WinCE and Windows for Pen-Based Computing before it. The Windows interface does not transfer well to a handheld and that frustrates users much more than learning an "unfamiliar" interface like the Palm OS.
Having some apps that work with KDE and others that work only with Gnome can be a pain. However, I think we should be careful not to tie Linux too closely to a single interface paradigm in case it causes problems for the whole range of Linux devices.
I have wondered about this as well. It seems to me that one of the real strengths of the open source model is that it frees developers from having to meet artificial ship dates in order to sell a bunch of upgrades to inflate the quarterly sales figures.
Yes, having a nifty one click installer would be realy nice. So would having a kernel that scales well above four processors, a free database with a full SQL implementation, a journaling file system, or any of a thousand other technical goals.
The real difference between Linux and a commercial OS is that Linux developers have the freedom to get the tech right. MS or any other commercial software house needs to ship releases quickly to keep the operation going, even if that means releasing half baked early beta quality software as a 1.0 release.
Easy installs and all the rest are worthy goals, but I don't think that technical goals should be sacrificed for the sake of aesthetic ones.
While I have certainly seen my share of ugly exchanges between Linux users, a lot of what comes off as anger is ideological zeal.
Some people who use and develop Linux do so because it is the right tool for the job, or because it is efficient, or because it doesn't cost anything, or other equally valid reasons. You cannot forget that there is also a large population of people who use and code for Linux because they feel it is the *right* thing to do.
Other platforms don't have this, and it is a huge difference. Can you imagine a Windows user saying that running Windows is morally superior to other types of computers? Even smug Mac users (and I speak as a longtime smug Mac user) feel a deep technical superiority and a sense of oppression from the MS juggernaut, but I don't think any of them thinks using a Mac makes one a better person.
Some of the people who have slogged away for years bringing Linux to its current level have this sense of occupying the moral high ground. RMS is the obvious example.
Linux means a lot to these people, much more than Windows does to an MCSE. It is unavoidable that they are going to have strong feelings and that, sometimes, this will look like spleen venting. Maybe there is a parallel with religion. Some religious people might respond to a precieved insult to their beliefs very badly. It doesn't excuse rude behaviour but the motivation is much deeper than a simple flame.
The part I was most impressed by was the fact that BootX is not required. I think BootX is a great tool (and that Ben Herrenschmidt is One Cool Guy for writing it), but I like the idea of not needing a Mac OS partition.
It looks like IBM is finally getting their POP board spec finished. With any luck we should be seeing dirt cheap non-Apple branded hardware before too long.
LinuxPPC has been working with IBM on software for the POP systems, so it is not too surprising to see their distro targeted towards these machines. However, YDL has some kind of a contract with IBM, so I would not be surprised to see them supporting the new hardware pretty soon as well.
Well done, LinuxPPC! If I weren't downloading an ISO of mkLinux right now, I'd be grabbing the new release already.
I think Katz has it all backwards. Yes, many geeks were picked on in school and some may not have Emily Post approved social skills, but this is not what makes up geeks. Reading the first few thousand words of the excerpt he seems to be arguing that social isolation is what causes geekdom. In fact the opposite is true. It is the focus on technology that causes the distance from "normal" society.
When I was 12 years old, the only thing I cared about was hacking my TRS-80. I wrote thousands and thousands of lines of code for that beast because it was what I loved to do. It took precedence over playing sports or going to dances at school or other things that some other kids were into. But the fact that I was not interested in their Lord of the Flies style social Darwinism did not make me code better.
I think Katz fails to appreciate that we are differenct because we are different. The societal pressure didn't make us learn assembly language. This is the same reason he wants to lump artists and other nonconformists together with geeks. He sees the external similarities between groups of social outcasts, but misses the essence of geekdom entirely.
I don't want to seem to harsh on him, though. At least Katz is putting forward the effort to try and understand who geeks are and what makes us tick. He's not there yet, but at least he's trying, and that's a lot more than I can say about most non-geeks I encounter.
Think of the orbit as a bicycle wheel. The earth is at the hub and the orbits are at different points along the spokes.
As the wheel spins, the parts that are farther away from the hub havce to travel farther to make a complete revolution, but since the wheel turns as a unit, the outer points have to turn faster than the inner points.
The difference between a satellite in orbit and a bicycle wheel is that the satellite is being pulled toward the Earth. To calculate a geosynchronous orbit, you find the point where the speed needed to keep from crashing back to earth is the same as the speed needed to match the Earth's rotation. It turns out that the point where these two match is at an altitude of about 30,000km.
At 675km, the gravitational attraction is much stronger and a satellite needs to orbit at a higher speed to keep aloft. If you slowed the orbital speed to match the rotation of the Earth you would need to constantly burn fuel to counteract the force of gravity.
That's pretty much how I would explain it to my four year old. I hope it makes sense.