What's really needed is something that an engineer at a random hardware manufacturer can pick up and run with--not so much the why or even the what of Linux devices drivers as the how. Although Linux Device Drivers is a good start, it's getting a little long in the tooth and will rapidly get outdated with some of the resource management changes that will appear in 2.4 (and are already in the development 2.3 version). New device types like USB (a whole suite of issues here), video capture, and so on, need to be treated, along with better coverage of issues like the SCSI interface.
Ideally, all an engineer should need to do is find a driver for the device most similar to his/her new device, and follow some general steps in indentifying and making the necessary adaptations. In a perfect world, the engineer would understand the Linux kernel inside and out, but chances are he/she will be completely Linux-ignorant, and will have a matter of a few weeks in which to produce the driver. Thus it would be helpful to draw parallels to Windows drivers (which are likely to be already available or at least more familiar to the engineer), and perhaps cookbook methods for translating them for Linux. Put a limit on theory, history, and other details that don't help the engineer get the job done, fast.
I know that this attitude might be a bit controversial, but in the real world, non- support for a prefered device is a frequent show-stopper for Linux. Let the open-source community work on making the drivers work well--but there has to be a driver in the first place.
Let's hope Alessandro Rubini, or someone of his knowledge and abilities, can produce such a book. It's one of the more crucial issues blocking Linux World Domination(tm).
Excluding evidence obtained by illegal means is often the only way of "punishing" such constitutional violations. Unless the violation is exceedingly flagrant or notorious, it's damn hard to get a civil judgement against law enforcement for such violations, much less a criminal judgement.
Since the media will generally only report the most spectacular cases, both of excluded evidence and of constitutional abuse, it's easy to get a distorted view. The fact is that very few serious criminals "get off on technicalities," and that petty (and occasionally not-so-petty) violations of constitutionality are pretty common in some jurisdictions. You'll only hear about the cases that sell newspapers or raise ratings, of course...
No one vendor--even MS--will make you profitable. I'll make some suggestions, but first an obeservation:
Let's see, just how profitable is Microsoft's portal business? Hint: it isn't. Compare them with Yahoo!, which is profitable (and has been for some time). Guess which one uses Unix? (FreeBSD, to be exact.) Guess which one is located in the SF Bay Area?
Here's my list:
Run lean. Hire self-managing people of proven experience who add directly to your product. If you need to, hire a few managers to make sure these people received what they need in terms of resources and coordination. Don't "bulk up" management until you absolutely have to. Leanness reduces friction, and means you can respond more quickly to opportunities--or problems.
Listen to your customers. They will be a major source of your ideas, and the first to tell you if you're doing something wrong. Respond to them and personalize service, if you can. One reason there is so little brand loyalty is that customer service on the Internet is almost uniformly terrible. But even a fickle customer can give you a really good idea in the brief time they are around. The Internet makes it easier to listen; use this an an edge.
Plan ahead. Establish goals, but don't be afraid to change them as your business and its environment change. Revisit your progress frequently as you go along. "How am I doing today?" "Where will this take me six months from now? A year?" Things like investments or an IPO are steps along the road, things that enable you to reach goals. But focusing exclusively or even primarily on them is a sure way to screw up your business.
Manage risks. Establish contigency plans for when things fail. If you find yourself spending more than a little time in reactive mode (responding to problems as they appear and dealing with their aftermath) you aren't spending enough effort on mapping out the areas of greatest risk to your business and acting preemptively.
Failing in any one of these areas--hiring, customer relations, planning, risk management--can doom you no matter how good your marketing plan is. Yet the overwhelming number of Internet startups I've seen have come up short in one of these areas--and more than a few are already gone.
I think Katz should be sentenced to watching so-called children's television every day for a month-- and I don't mean the stuff on PBS. The intensity of mass marketing in what is otherwise a creative wasteland is truly awesome, and crushingly mind-numbing: non-stop, aimless excitment filled with every product tie-in imaginable.
Pediatricians are saying that this stuff isn't healthy, and I for one agree with 'em. Parents have the right--and a duty--to set limits for their kids. It's no more outrageous to suggest they limit kid's watching of television than to suggest they limit kid's consumption of candy.
No, the USPS is a government corporation, like, for example, the FDIC (Federal Deposit Insurance Corporation), formed by an act of Congress. It's quasi-independent in terms of its operations, but it still must refer back to Congress for significant changes (such as postal rates).
Note for folks who slept through civics class: the Post Office is actually mandated by the US constitution. Another tidbit: because it is legally required to charge the same rate for first-class mail, regardless of costs, the USPS has been given an exclusive right to carry such mail. Given that the US Senate would have to approve a change in this, and that high-cost states have as many senators (though far fewer people) than low-cost states, you aren't likely to see it happen any time soon.
Yahoo has generally made a profit for the past two and a half years; it would have again last quarter except for the money it spent on acquiring GeoCities and Broadcast.com. (Since those were one-time charges, technically it was profitable.)
Of course, Yahoo's stock valuation is a few hundred times its earnings--much higher than almost any non-Internet firm. And certainly much higher than RedHat's will be, assuming it actually does prove profitable. That is, unless investors begin seeing as strong a potential for growth in Linux and open-source as they do in the Internet.
This was my beef with the article: it said that RedHat was less risky than an Internet IPO (which I agree with) but never quite explained why. They aren't primarily an Internet company, nor are they a traditional OS company, yet those are the two things the article compared them to. Like so many articles of its ilk, it never quite found a clue as to just what RedHat is.
I'd hardly call the article in-depth. It danced around, saying it's a "much safer bet" than the standard Internet IPO without really explaining why, then makes a comparison to Be, Inc which ignores the fact that OS sales plays a relatively small role in Red Hat's future business plans. Overall, it isn't even a particulary good specimen of punditry.
As for moderation issues--I'm sure Rob is working on improving things. Don't forget that this sort of public moderation is quite new, and there will be a lot to observe and learn in Getting It Right.
It's a backlit LCD, folks, not a searchlight. This isn't a Hoover vacuum, ya know.
The typical rack full of servers connected to a CPU switch is a real bear to check while standing in front of it--switching to each server in turn takes time. It's much nicer to have each server displaying its load and other health info, even if it only has a 4x20-char display. Then use the CPU switch to check out the one with a load of 395.7, or forget the switch (and video card in each server) and just plug into (or switch to) the appropriate serial port to find out more.
The word is "perquisite." I've seen it abbreviated as both "perq" and "perk."
Perqs I'd like:
A reasonably-sized desk with drawers
A bookshelf (or two)
A window
A door
Quiet
A decent computer (large monitor, fast CPU, large memory, good network, good OS)
A decent chair
I'll handle the rest--no need to add all that other stuff just to make me forget that my actual working environment is just a cushioned warehouse with garish fluorescent lights.
"Linux is 30 year old technology. It's just a notch above Luddism." Bob Metcalfe
Electronic computers are a 55-year-old technology. What's Metcalfe doing still messing around with them?
As others have mentioned before, WinNT is based on Dec's VMS, an OS that is of the same vintage as Unix. In fact, Microsoft themselves proudly point to this when touting WinNT's enterprise credentials. Why isn't Metcalfe diss'ing WinNT as "30-year-old technology?"
Claims such as Metcalfe's assume that those who hear them are ignorant of Unix's history. In 1978, when I first encountered Unix, it had maybe a tenth the functionality it does today, even though it was already nine years old. It was 16-bit, it only supported 64MB filesystems, 20 open files was a kernel-enforced limit, file names could only be 14 characters-- it had no networking, no demand paging or other virtual memory features, no threads, and so on. What it did have was the modularity and elegance that its successors, including Linux, have emulated with greater or lesser success as they have extended its limits and added all the modern features of a 1990's OS.
So Metcalfe should know better. And so should you.
It isn't even law yet--just proposed law. Each state's legislature has to adopt it. Typically, they have adopted such codes in the past, but it's not a done deal yet. It can still be fought on a state-by-state level.
I mean the idea that the company is 1) planning on loosing money,
Heard of any good IPO's for companies that weren't loosing money? The issue is whether they'll use their indebtedness to improve their business in such a way as to make more money later on.
2) any body can take their product,
That's what open-source is about. But that doesn't keep them from making money through services (which, for businesses, make up a much larger outlay than software purchases).
3) their biggest competitor is M$
Yep, and the fact that Microsoft sees Linux as a serious threat actually adds credibility. It's a Good Thing(tm).
These investors are just looking to see if they can make money off of this.
Yep. And they've sat through enough rah-rah glossy presentations to be pretty jaded concerning those "comforting things" you talk about. They'll go through the business plan with a microscope (and an accountant) before investing. The purpose of these presentations is to get them interested enough to read it. No one says "Great talk, Bob!" and whips out their wallet...
RH doesn't need investers playing "buzzword bingo." They need investers who will take the time to see just how this rather unusual business model would work, and who won't bolt the first time open source or Linux gets diss'ed in the trades. Honesty and a lack of pretentiousness is in their interest.
The only thing standing in the way of other ISPs carrying some of the load--hosting instant messaging for their customers--is the proprietary nature of the system. That's the crux of the issue! AOL pays because they insist on keeping their system closed! With an open standard--such as the one AOL originally proposed and then reneged on--an AOL client could message someone on another ISP through that ISP's server. Everyone would "flip the bill," like email. But AOL wants a monopoly on instant messaging, so taking a page from another "Bill," they are trying to make the technology closed and proprietary.
The main difference is your mail client (Outlook, Netscape, Eudora, whatever) is not directly linked to a competing service and simply leeching off the existing market leader's users and network resources to gain share.
You're leaching off of a competetor's mail server when you send email messages to their members. Hundreds of millions of non-AOL email messages use AOL's mail servers every month. You don't hear about them changing SMTP to block email from competitors. Why should instant messages be any different?
"Direct connection" is a red herring-- if I send email to someone at AOL, my mail agent is making a "direct connection" to AOL's mail servers. The only substantive difference is that they queue the message for later retrieval rather than generating a screen pop.
Lets see; my company has a rather severe policy that people not release their email password to anyone. They can fire someone for sharing their password. Yet people glibly type their passwords into Eudora, Outlook, Netscape (or Pine). What's more, my company turns a blind eye to this practice--and it's a good thing, too, since no one would ever read their email otherwise.
There is little difference between a mail agent and an instant messager. Both request passwords to pass to the underlying server. You aren't "giving your password to Microsoft" by using their messager any more than you are "giving your password to Microsoft" whenever you use Outlook. (All bets are off if your mail agent or messager is a trojan, of course.)
I can't believe that someone here actually fell for AOL's obvious FUD.
Symbios cards work well with Linux. Both my Alpha and my P-II Linux boxen have 53C875-based controllers, and both perform outstandingly. So even if Mylex/BusLogic were to go over to The Dark Side, there is still a good alternative to Adaptec.
You'll note that IBM has become a pretty big player in commodity disk drives. This acquisition gives them one more piece of the low-end server.
MacOS has the advantage that the same company which controls the OS controls the hardware it runs on. It would be pretty simple for Linux (and, admittedly, Windows) to reliably auto-install on a half-dozen pre-defined hardware configurations. Instead, Linux is faced with billions of potential hardware configurations, with different CPU's, BIOS's, glue logic, disk controllers, video controllers, mice, keyboards, and so on. It's amazing it does as well as it does...
One of the first (if not the first) ARPANET mailing lists was HUMAN-NETS, formed before many Slashdot readers were even born. Many of the discussions there anticipated the Internet of today. It's a precurser to USENET, and to discussion sites like this. Unfortunately, although some of the last HUMAN-NETS postings are archived in http://communication.ucsd.edu/A-News/FA.human-nets /FA.human-nets-index.html, I don't know where earlier issues are archived.
Such an archive would be a record of the earliest wide-ranging (not just technical) discussion on a computer network.
When ICANN was formed, its bylaws stated that it would fund itself through domain registration. Just how this would be done was left unsaid, but it was tacitly assumed that some sort of a fee would be arranged. When ICANN did the obvious thing with the $1/domain fee, NSI turned its FUD-generating PR apparatus loose, and swore publicly that they'd never pay ICANN a dime. As a result, ICANN--which was to help broker between the several domain authorities--is in serious debt, and thus pretty much unable to do much of anything.
As for Esther Dyson, she's one of those people who has long been working toward intra-industry coordination. She sits on the boards of a number of organizations Slashdotters might know about, such as the EFF and Cygnus Solutions. (She chaired the EFF until stepping down to head ICANN.) I'm mystified why some folks here want to demonize her or put her in the same catagory as magazine pundits.
TENEX? Multics? That was back in NCP days--the ARPANET didn't use TCP/IP until the mid-'80s. You've been around long enough to remember the Great Change, Guy. I guess you're technically correct, though, if we ignore your reference to "TCP/IP"-- NCP was IP in its day. (And you folks were wondering why IPv4 wasn't IPv1... well, it turns out that the original ARPANET first used something other than TCP. Back just before TCP appeared, DEC-10's were the kings of the net.)
IIRC, there were simultaneous efforts to build IPv4 stacks on TOPS-20 and others, but the first full implementation was done at Berkeley and incorporated into the Berkeley Standard Distibution for VAX and then PDP-11--and DARPA paid good money for it.
For those of you in the US who slept during civics class:
The power to create the postoffice is explicitly mentioned in the US constitution (unlike much of what the government does these days). The founders felt that universal mail delivery was one of the foundations of a democracy. Thus any postal system--even if it consisted of a bunch of private companies--would wind up heavily regulated, and thus expensive to run. The current quasi-indendent corporation is probably the best solution, given this mandate.
Currently, only the USPS can carry normal letters. The argument is that if competition were allowed, competitors would "cream skim" the profitable routes and services such that the USPS could no longer be self-sustaining (since it must, by law, deliver to each US resident), or alternatively the USPS would have to raise rates for outlying areas. Since congress has to approve rates, the latter simply isn't going to happen; there are too many sparsely populated states with two senators each.
AOL could have approached AT&T like businessmen and said, "Let's make a deal".
AT&T would have laughed in their faces. The issue here isn't one of paying for bandwidth. @HOME is a content-provider, not just an ISP. It competes directly with AOL for "eyeballs" (i.e. ad revenue). AT&T has no incentive to share bandwidth for just the cost of that bandwidth, since it means they no longer can force subscribers to use their software and come to their portal.
Said another way, AT&T wants to invade AOL's territory, but exclude AOL from theirs by buying control over access. At stake are tens of (US$) billions in projected revenue over the next few years.
I survived the Northridge earthquake (I live less than two miles from its epicenter). It was intense, but most houses in my neighborhood came through with little or no damage (as did mine). On the other hand, a frame house--and most other structures--have little chance against a tornado or flood.
My dentist's office was right across Tampa Blvd from Northridge Meadows. He suffered very little damage (some broken stuff in the lab, an X-ray head crashed into a wall), and he was back seeing patients in less than a week. (Given we were still having pretty strong aftershocks, I'm not sure I would have gone to see him at that time, though.)
Northridge Meadows was built in violation of building codes; not only was it missing a shear wall required by code, but structural members (lumber) and fasteners were not up to code either. It was cheap, shoddy construction. It was hardly unique--some other areas became ghosttowns because of heavily damaged apartment buildings--but it was, fortunately, the only such building that pancaked.
Rumors that the Richter number for the Northridge quake was "adjusted" are no more than that. Part of the reason for the damage it caused was the soft soil underlying much of the San Fernando Valley and the nature of the fault itself (thrust faults generate relatively more vertical energy than strike-slip faults like the San Andreas). There is no such FEMA rule, in any case...
Those are all web-development and E-commerce positions, except for the "office manager" (they'll need one of those anyhow). Perhaps they're fixing to move their portal operations out West, or create yet another on-line Linux store.
Slashdot Mirror
Ideally, all an engineer should need to do is find a driver for the device most similar to his/her new device, and follow some general steps in indentifying and making the necessary adaptations. In a perfect world, the engineer would understand the Linux kernel inside and out, but chances are he/she will be completely Linux-ignorant, and will have a matter of a few weeks in which to produce the driver. Thus it would be helpful to draw parallels to Windows drivers (which are likely to be already available or at least more familiar to the engineer), and perhaps cookbook methods for translating them for Linux. Put a limit on theory, history, and other details that don't help the engineer get the job done, fast.
I know that this attitude might be a bit controversial, but in the real world, non- support for a prefered device is a frequent show-stopper for Linux. Let the open-source community work on making the drivers work well--but there has to be a driver in the first place.
Let's hope Alessandro Rubini, or someone of his knowledge and abilities, can produce such a book. It's one of the more crucial issues blocking Linux World Domination(tm).
Excluding evidence obtained by illegal means is often the only way of "punishing" such constitutional violations. Unless the violation is exceedingly flagrant or notorious, it's damn hard to get a civil judgement against law enforcement for such violations, much less a criminal judgement.
Since the media will generally only report the most spectacular cases, both of excluded evidence and of constitutional abuse, it's easy to get a distorted view. The fact is that very few serious criminals "get off on technicalities," and that petty (and occasionally not-so-petty) violations of constitutionality are pretty common in some jurisdictions. You'll only hear about the cases that sell newspapers or raise ratings, of course...
No one vendor--even MS--will make you profitable. I'll make some suggestions, but first an obeservation:
Let's see, just how profitable is Microsoft's portal business? Hint: it isn't. Compare them with Yahoo!, which is profitable (and has been for some time). Guess which one uses Unix? (FreeBSD, to be exact.) Guess which one is located in the SF Bay Area?
Here's my list:
Failing in any one of these areas--hiring, customer relations, planning, risk management--can doom you no matter how good your marketing plan is. Yet the overwhelming number of Internet startups I've seen have come up short in one of these areas--and more than a few are already gone.
I think Katz should be sentenced to watching so-called children's television every day for a month-- and I don't mean the stuff on PBS. The intensity of mass marketing in what is otherwise a creative wasteland is truly awesome, and crushingly mind-numbing: non-stop, aimless excitment filled with every product tie-in imaginable.
Pediatricians are saying that this stuff isn't healthy, and I for one agree with 'em. Parents have the right--and a duty--to set limits for their kids. It's no more outrageous to suggest they limit kid's watching of television than to suggest they limit kid's consumption of candy.
No, the USPS is a government corporation, like, for example, the FDIC (Federal Deposit Insurance Corporation), formed by an act of Congress. It's quasi-independent in terms of its operations, but it still must refer back to Congress for significant changes (such as postal rates).
Note for folks who slept through civics class: the Post Office is actually mandated by the US constitution. Another tidbit: because it is legally required to charge the same rate for first-class mail, regardless of costs, the USPS has been given an exclusive right to carry such mail. Given that the US Senate would have to approve a change in this, and that high-cost states have as many senators (though far fewer people) than low-cost states, you aren't likely to see it happen any time soon.
Compare that with, say, Yahoo's losses per share.
Yahoo has generally made a profit for the past two and a half years; it would have again last quarter except for the money it spent on acquiring GeoCities and Broadcast.com. (Since those were one-time charges, technically it was profitable.)
Of course, Yahoo's stock valuation is a few hundred times its earnings--much higher than almost any non-Internet firm. And certainly much higher than RedHat's will be, assuming it actually does prove profitable. That is, unless investors begin seeing as strong a potential for growth in Linux and open-source as they do in the Internet.
This was my beef with the article: it said that RedHat was less risky than an Internet IPO (which I agree with) but never quite explained why. They aren't primarily an Internet company, nor are they a traditional OS company, yet those are the two things the article compared them to. Like so many articles of its ilk, it never quite found a clue as to just what RedHat is.
Look at the byline--the article is from ZDNet.
I'd hardly call the article in-depth. It danced around, saying it's a "much safer bet" than the standard Internet IPO without really explaining why, then makes a comparison to Be, Inc which ignores the fact that OS sales plays a relatively small role in Red Hat's future business plans. Overall, it isn't even a particulary good specimen of punditry.
As for moderation issues--I'm sure Rob is working on improving things. Don't forget that this sort of public moderation is quite new, and there will be a lot to observe and learn in Getting It Right.
It's a backlit LCD, folks, not a searchlight. This isn't a Hoover vacuum, ya know.
The typical rack full of servers connected to a CPU switch is a real bear to check while standing in front of it--switching to each server in turn takes time. It's much nicer to have each server displaying its load and other health info, even if it only has a 4x20-char display. Then use the CPU switch to check out the one with a load of 395.7, or forget the switch (and video card in each server) and just plug into (or switch to) the appropriate serial port to find out more.
The word is "perquisite." I've seen it abbreviated as both "perq" and "perk."
Perqs I'd like:
- A reasonably-sized desk with drawers
- A bookshelf (or two)
- A window
- A door
- Quiet
- A decent computer (large monitor, fast CPU, large memory, good network, good OS)
- A decent chair
I'll handle the rest--no need to add all that other stuff just to make me forget that my actual working environment is just a cushioned warehouse with garish fluorescent lights."Linux is 30 year old technology. It's just a notch above Luddism."
Bob Metcalfe
Electronic computers are a 55-year-old technology. What's Metcalfe doing still messing around with them?
As others have mentioned before, WinNT is based on Dec's VMS, an OS that is of the same vintage as Unix. In fact, Microsoft themselves proudly point to this when touting WinNT's enterprise credentials. Why isn't Metcalfe diss'ing WinNT as "30-year-old technology?"
Claims such as Metcalfe's assume that those who hear them are ignorant of Unix's history. In 1978, when I first encountered Unix, it had maybe a tenth the functionality it does today, even though it was already nine years old. It was 16-bit, it only supported 64MB filesystems, 20 open files was a kernel-enforced limit, file names could only be 14 characters-- it had no networking, no demand paging or other virtual memory features, no threads, and so on. What it did have was the modularity and elegance that its successors, including Linux, have emulated with greater or lesser success as they have extended its limits and added all the modern features of a 1990's OS.
So Metcalfe should know better. And so should you.
can it be overturned?
It isn't even law yet--just proposed law. Each state's legislature has to adopt it. Typically, they have adopted such codes in the past, but it's not a done deal yet. It can still be fought on a state-by-state level.
I mean the idea that the company is 1) planning on loosing money,
Heard of any good IPO's for companies that weren't loosing money? The issue is whether they'll use their indebtedness to improve their business in such a way as to make more money later on.
2) any body can take their product,
That's what open-source is about. But that doesn't keep them from making money through services (which, for businesses, make up a much larger outlay than software purchases).
3) their biggest competitor is M$
Yep, and the fact that Microsoft sees Linux as a serious threat actually adds credibility. It's a Good Thing(tm).
These investors are just looking to see if they can make money off of this.
Yep. And they've sat through enough rah-rah glossy presentations to be pretty jaded concerning those "comforting things" you talk about. They'll go through the business plan with a microscope (and an accountant) before investing. The purpose of these presentations is to get them interested enough to read it. No one says "Great talk, Bob!" and whips out their wallet...
RH doesn't need investers playing "buzzword bingo." They need investers who will take the time to see just how this rather unusual business model would work, and who won't bolt the first time open source or Linux gets diss'ed in the trades. Honesty and a lack of pretentiousness is in their interest.
With AOL's AIM network only AOL flips the bill.
The only thing standing in the way of other ISPs carrying some of the load--hosting instant messaging for their customers--is the proprietary nature of the system. That's the crux of the issue! AOL pays because they insist on keeping their system closed! With an open standard--such as the one AOL originally proposed and then reneged on--an AOL client could message someone on another ISP through that ISP's server. Everyone would "flip the bill," like email. But AOL wants a monopoly on instant messaging, so taking a page from another "Bill," they are trying to make the technology closed and proprietary.
The main difference is your mail client (Outlook, Netscape, Eudora, whatever) is not directly linked to a competing service and simply leeching off the existing market leader's users and network resources to gain share.
You're leaching off of a competetor's mail server when you send email messages to their members. Hundreds of millions of non-AOL email messages use AOL's mail servers every month. You don't hear about them changing SMTP to block email from competitors. Why should instant messages be any different?
"Direct connection" is a red herring-- if I send email to someone at AOL, my mail agent is making a "direct connection" to AOL's mail servers. The only substantive difference is that they queue the message for later retrieval rather than generating a screen pop.
Lets see; my company has a rather severe policy that people not release their email password to anyone. They can fire someone for sharing their password. Yet people glibly type their passwords into Eudora, Outlook, Netscape (or Pine). What's more, my company turns a blind eye to this practice--and it's a good thing, too, since no one would ever read their email otherwise.
There is little difference between a mail agent and an instant messager. Both request passwords to pass to the underlying server. You aren't "giving your password to Microsoft" by using their messager any more than you are "giving your password to Microsoft" whenever you use Outlook. (All bets are off if your mail agent or messager is a trojan, of course.)
I can't believe that someone here actually fell for AOL's obvious FUD.
Symbios cards work well with Linux. Both my Alpha and my P-II Linux boxen have 53C875-based controllers, and both perform outstandingly. So even if Mylex/BusLogic were to go over to The Dark Side, there is still a good alternative to Adaptec.
You'll note that IBM has become a pretty big player in commodity disk drives. This acquisition gives them one more piece of the low-end server.
MacOS has the advantage that the same company which controls the OS controls the hardware it runs on. It would be pretty simple for Linux (and, admittedly, Windows) to reliably auto-install on a half-dozen pre-defined hardware configurations. Instead, Linux is faced with billions of potential hardware configurations, with different CPU's, BIOS's, glue logic, disk controllers, video controllers, mice, keyboards, and so on. It's amazing it does as well as it does...
One of the first (if not the first) ARPANET mailing lists was HUMAN-NETS, formed before many Slashdot readers were even born. Many of the discussions there anticipated the Internet of today. It's a precurser to USENET, and to discussion sites like this. Unfortunately, although some of the last HUMAN-NETS postings are archived in http://communication.ucsd.edu/A-News/FA.human-nets /FA.human-nets-index.html, I don't know where earlier issues are archived.
Such an archive would be a record of the earliest wide-ranging (not just technical) discussion on a computer network.
When ICANN was formed, its bylaws stated that it would fund itself through domain registration. Just how this would be done was left unsaid, but it was tacitly assumed that some sort of a fee would be arranged. When ICANN did the obvious thing with the $1/domain fee, NSI turned its FUD-generating PR apparatus loose, and swore publicly that they'd never pay ICANN a dime. As a result, ICANN--which was to help broker between the several domain authorities--is in serious debt, and thus pretty much unable to do much of anything.
As for Esther Dyson, she's one of those people who has long been working toward intra-industry coordination. She sits on the boards of a number of organizations Slashdotters might know about, such as the EFF and Cygnus Solutions. (She chaired the EFF until stepping down to head ICANN.) I'm mystified why some folks here want to demonize her or put her in the same catagory as magazine pundits.
This is just a normal LCD display with the front polarizer removed. Anyone else with a polarizer could see it too.
TENEX? Multics? That was back in NCP days--the ARPANET didn't use TCP/IP until the mid-'80s. You've been around long enough to remember the Great Change, Guy. I guess you're technically correct, though, if we ignore your reference to "TCP/IP"-- NCP was IP in its day. (And you folks were wondering why IPv4 wasn't IPv1... well, it turns out that the original ARPANET first used something other than TCP. Back just before TCP appeared, DEC-10's were the kings of the net.)
IIRC, there were simultaneous efforts to build IPv4 stacks on TOPS-20 and others, but the first full implementation was done at Berkeley and incorporated into the Berkeley Standard Distibution for VAX and then PDP-11--and DARPA paid good money for it.
For those of you in the US who slept during civics class:
The power to create the postoffice is explicitly mentioned in the US constitution (unlike much of what the government does these days). The founders felt that universal mail delivery was one of the foundations of a democracy. Thus any postal system--even if it consisted of a bunch of private companies--would wind up heavily regulated, and thus expensive to run. The current quasi-indendent corporation is probably the best solution, given this mandate.
Currently, only the USPS can carry normal letters. The argument is that if competition were allowed, competitors would "cream skim" the profitable routes and services such that the USPS could no longer be self-sustaining (since it must, by law, deliver to each US resident), or alternatively the USPS would have to raise rates for outlying areas. Since congress has to approve rates, the latter simply isn't going to happen; there are too many sparsely populated states with two senators each.
AOL could have approached AT&T like businessmen and said, "Let's make a deal".
AT&T would have laughed in their faces. The issue here isn't one of paying for bandwidth. @HOME is a content-provider, not just an ISP. It competes directly with AOL for "eyeballs" (i.e. ad revenue). AT&T has no incentive to share bandwidth for just the cost of that bandwidth, since it means they no longer can force subscribers to use their software and come to their portal.
Said another way, AT&T wants to invade AOL's territory, but exclude AOL from theirs by buying control over access. At stake are tens of (US$) billions in projected revenue over the next few years.
I survived the Northridge earthquake (I live less than two miles from its epicenter). It was intense, but most houses in my neighborhood came through with little or no damage (as did mine). On the other hand, a frame house--and most other structures--have little chance against a tornado or flood.
My dentist's office was right across Tampa Blvd from Northridge Meadows. He suffered very little damage (some broken stuff in the lab, an X-ray head crashed into a wall), and he was back seeing patients in less than a week. (Given we were still having pretty strong aftershocks, I'm not sure I would have gone to see him at that time, though.)
Northridge Meadows was built in violation of building codes; not only was it missing a shear wall required by code, but structural members (lumber) and fasteners were not up to code either. It was cheap, shoddy construction. It was hardly unique--some other areas became ghosttowns because of heavily damaged apartment buildings--but it was, fortunately, the only such building that pancaked.
Rumors that the Richter number for the Northridge quake was "adjusted" are no more than that. Part of the reason for the damage it caused was the soft soil underlying much of the San Fernando Valley and the nature of the fault itself (thrust faults generate relatively more vertical energy than strike-slip faults like the San Andreas). There is no such FEMA rule, in any case...
-EdThose are all web-development and E-commerce positions, except for the "office manager" (they'll need one of those anyhow). Perhaps they're fixing to move their portal operations out West, or create yet another on-line Linux store.