I'm only going to say this once - XML is *a* solution to the trivial problem of syntax;
Syntax is not a trivial problem between advanced applications. Designing a protocol that is expandable, be it for config files or interprocess communication is a pain in the neck. We started using SGML for config files and interprocess communication in 1995. You have no idea how much work we saved by using standard parsers and a protocol that wouldn't break if we added a column to the message...
it does not, however, assault the intractable problems of semantics.
True enough. XML only gets you half-way there. Isn't that still better than nothing? Moreover, since the semantics is intractable, as you well state, the solution is to manually provide syntactic markings through, you guessed it... XML tagging.
After 6 years of adminning systems, I feel more secure with each daemon checking their config files than passing it onto an independent parser daemon.
Reality check. What is likelier to be buggy: a one-off parsing routine or a well established and universally tested parser such as SAX?
We know the answer to that one since the whole open source movement is predicated on it. The publicly available routine will be less buggy.
The world is moving en-masse to XML. Yes, it is overhyped (just as high-level languages, structured programming, OOP and Java were in their time). Yet all of those were clear steps forward in computing.
Same goes with XML. Linux can be either ahead of the curve, or behind it, always destined to be a late copy of a thirty-year old operating system.
Now that Linux is stable and of amazing quality it is time to start looking towards the future and make sure a good operating system becomes hands down the best.
ACLs are a must have in the corporate network environment, where Linux will ultimately live or die.
Linux reflects the needs of its developers (their itches, to use Eric Raymond's terms). High security and restricted access are not common needs for people who enjoy hacking code, and thus the extreme weakness of the Linux security model, of which ACLs are but one component.
It is pretty obvious that ICANN went for the safe choices. This is quite alright as long as they revisit this issue not too far into the distant feature.
Real life often require moving conservatively on issues which could have a major impact, and it is a tried and true engineering technique.
Hopefully in a year from ICANN will aprove a few dozen more domains based on the experience from things such as.museum.
The W3C is fairly irrelevant. XML came from a group outside W3C. It was first proposed at the SGML conference as a reaction to the losing battle between SGML and the awfully designed HTML.
The W3C is corporate driven. You have to pay a hefty membership sum to be able to contribute. This is anathema to the way the Internet grew, where people became net.goods by contributing code.
It also suffers from the standards-body slow-as-molasses syndrome. I have participated in several working groups and they sem to work under a mandate to make as few changes as possible.
An effort such as HTML++ got delayed time and time again.
To reiterate, XML is here only because it originated outside the W3C umbrella.
They added Javascript and Java to the browser specifically to enable this.
Oh boy, how soon do they forget.
Netscape added Javascript in an attempt to forestall Sun's deployment of Java. Java got added only when Sun and Netscape signed a truce along the lines: "we, Sun, allow you, netscape, to use the Java moniker on your completely unrelated scripting language. In exchange you agree to support Java on your browser".
They created the first (or one of the first) web application servers - NAS, now iPlanet.
Another one... The first server was NCSA httpd, and many web sites transition from NCSA httpd to Apache without ever going through NSA/iPlanet.
People didn't see the web as a platform at first.
True, but by the time Netscape was talking about it, others had also grasped it, not least of which was Sun with their Java language, whose existance was made public in 1994.
It was Netscape that pushed this idea, evangelized it.
This much I can give them credit for. James Clark drove the point over and over until Bill Gates finally got it, and turned "hardcore on the internet". Gates immediately promoted the underlings who had spent the previous year trying to explain this to him.
To deny that is to overlook Netscape's primary contribution to the internet.
Netscape's primary contribution to the internet is developing a cool and fast web browser which was made available for free.
Netscape's goal was to turn the web into a platform. They succeeded in that goal.
Netscape had nothing to do this. The web is popular as a platform because it takes the fabled client-server architecture to the masses.
The benefits of a client-server architecture became apparent to all in the late 80s, but until the web appeared, writing a decent client-server application either required an advanced degree on networking and distributed systems, or the purchase of a closed-platform solution.
All that changed with the arrival of the web. You could flush all of netscape's buggy code down the toilet and people would still have developed for the web. Is the sensible thing to do in most cases.
AOL and Netscape deserve scorn for claiming the high moral ground of standards and openess only when they are losing. As soon as the have a dominant position they piss on them, such as with the blink tag.
Sure, this is a good idea, but the problem with many of the patents is not prior art. It's that they shouldn't have been issued because they're obvious to anyone working in the field, not because someone else did it earlier.
I agree. We implemented at least one of the patents before the date at a company I used to work for (Database copying) without ever considering it anything less than the obvious way to go and at least two other I talked about with a business partner as an obvious way of how to leverage the internet for business (Double click, Sight sound).
One-click falls squarely in this category. Once you are logged in (via password or cookie) making a sale without request of further information is obvious. Access any intranet site. Are you asked time and time again for your password? Or rather the system keeps track of who you are and shows you only those things you can look at? This is esentially undistinguishable from One-click, and the fact that money is exchanged at the end of the transaction is not non-obvious in any way.
Does the first ticket have to be a full-fare ticket?
No. They twin any ticket bought from selected aircarriers (AA, Delta and some others, if I remember correctly). I understand the plan changed this year with the talk about mergers and all, so read the fine print before enrolling...
FYI, Platinum's no longer a big deal--they hand them out to anyone willing to pay the $300/yr., pretty much... Don't be a sucker, switch back to green and save a couple hundred bucks a year.
Read the fine print. "Free companion ticket anywhere in the world". That is worth more than $300, and the sole reason why I upgraded to platinum six years ago...
While long distance stopped being a monopoly since the breakup of AT&T, it seems that a lot of the "screw the customer" attitude borne from those years is still with us and furthermore managed to make its way to other new carriers...
Just yesterday I had an incident with a Sprint "customer advocate" which refused to allow international calls from my digital cell phone.
Almost no business would turn away a platinum american express, six-figure-income programmer as a risky and undesirable customer, yet that is what Sprint did.
I think those are attitudes derived from times when the corporation, not the customer was king...
With an unemployment rate of 4.5% there is a shortage of all type of workers not only highly skilled ones. Anyone trying to argue otherwise should be granted honorary membership on the flat earth society.
I don't know if 180K H1B visas is too few or too many, but facts are facts, and it bothers me that so many moderators rank the "no such shortage" posts so highly, when actual objective measures to the contrary are readily available...
The NYT wants to have it both ways. First claims in one article that there is no IT shortage, next it writes in another that the shortage is so severe that IT professionals are skipping college.
That, my friends, is how news are manufactured just to stir up controversial issues and sell more copies. I wonder how many/.ers registered today in the NYT web site. The chief NYT editor should go to bed happy today.
When I was young, the advice from grown ups was "do not sign anything you don't have to, be it a contract, a letter, a memo, anything. If you sign it it means that you meant it, if you don't is just idle chatter".
So,/.ers out there: how about it, why do you sign your e-mail letters?
I pressume most of/. readers are producers of intelectual property in one way or another. Some do it for a fixed salary (aka employees) and others sell the product of their work on the market.
Surprisingly, the majority of the opinion here seems to be against copyright protection for music and patent protection for new inventions.
In real life, most peoples wages depend on copyright. Even RMS uses copyright protection to enforce its GPL terms.
That doesn't excuse stupid patents from the USPTO or judges that issue brain dead judgements covering all benefits from copyright, which is not what the law says....
At least (most) of the Linux users are honest about the shortcomings of their preferred OS,
Yeah right. They are not called religious wars for nothing.
Did you see the "honest" reaction to Miguel de Icaza's criticism of Linux? The juvenile puns on M$ that quickly get moderated to 4 Funny? How long did it take to the Linux community to admit that its UI sucks (thus LUIGI)? How much longer before they dump the retarded X-windows architecture?
I won't take sides, but here are a few useful observations.
It is too early to tell. Win2K has been out only for a short period of time, so the big flaws are yet to be found. Most of them get fixed in the first service patch.
Stability is mostly a red herring. In fact in the early days WinNT was more stable than Linux (I had a friend who made money selling cards to automatically reboot linux when the HD went idle for over 40 seconds). Nowadays Linux has a small edge on stability over NT, yet my development NT machine goes on for months at a time without power down/reboots.
Beware of anybody that claims to be balanced yet fails to admit that (1) Ken Cutler designed the NT kernel architecture and he's an OS god, and (2) Linux is a based on a very good, but very old operating system (UNIX). This does not automatically make NT good and Linux bad. However, most of the good things about NT and the bad things about Linux derive from those facts.
What for? If I were running a web server, IMHO Linux is the clear winner. Free Apache server, tons of free tools for server maintenance, high stability, low overhead. If you plan on exchanging memos with the secretary next door, you pretty much need an M$ compatible OS.
Who is it for? Linux is not yet ready to be hoisted upon my sixty year old parents. OTOH a CS major should be running Linux at home.
Beware of GUI arguments. Paraphrasing Churchill, the NT GUI is the worst possible, except for all others which are even worse. Linux is catching up quickly, and MacOS X seems likely to overtake both, but currently the least bad GUI of the two (linux-M$) is the Win95-Win2k line.
PayPal is pretty much unassailable as the de facto Internet payment system.
The net has over 100 million users, the big three credit card companies have nearly a combined billion cardholders, yet according to Cringely the 3.3 million customers of PayPal makes them "unassailable".
In fact, I'd say that 3.3 million users is a very small set of net users, and that the real challenge is to reach those who are not geeks nor addicted to online auctions.
Having been a Mac aficionado back in the early 90s (with a pair of PowerBooks), I still don't see why Apple is considered such the "nice guy" in these debates.
Apple has legendarily screwed their customers, starting with making the Mac incompatible with the Apple ][.
But somehow Apple always comes accross as the good guys because they use a rainbow logo....
Can someone please explain to me when we all turned away from Linux Advocacy, in which the strengths/benefits of Linux are placed above marketing and mudslinging?
And eight months later Forrester says Microsoft is doomed? I didn't take them seriously in the slightest then and I don't now.
More importantly who reads Wired nowadays? The artsies took over about four years ago, overhyping mumbo-jumbo stuff such as Idealab and the Santa Fe institute. The internet savvy gave up on wired a long time ago. They now read nerve, suck, salon, redherring and of course/. All of them on-line.
Re:The Author Doesn't Know What An OS is...
on
Is UNIX An OS?
·
· Score: 1
Anybody with a degree in CS (or anyone who's ever taken a college level intro to computing class) knows that this is not the definition of an OS. To put it simply an operating system is that is initially loaded when the computer is booted and manages the system resources as well as the other programs running on the computer.
That definition is old and incorrect. Trust me, I have three degrees in Computer Science. Some old OS textbooks still quote that antiquated old definition. The modern definition of OS is a system that provides all commonly required services.
Notice that this is a moving target.
To quote John Ousterhoot (of Tcl fame) back when he was teaching the OS course at Berkeley (1980s): The essence of an operating system lies in the services it provides to user programs
Are hard drives common? Yes, and therefore the drivers are part of the OS. Are hand-held wireless scanners common? No, and thus their drivers do not come with the OS.
Saying that an OS is a program loader + file manager is tantamount to saying that a computer is a collection of vacuum tubes hardwired together to perform a computation. At some point in the past these definitions were correct. Today they are outdated and have been updated by lead researchers in the field (such as Ousterhoot). Problem is they haven't made it into some texts. But it has into others. Silberschatz writes in Operating Systems Concepts:
It is easier to define operating systmes by what they do than by what they are. The primary goal of an operating system is convenience for the user.
This definition is closer to the McWeek one that to the one "from anybody who has a degree in CS".
By the way, I think the author of the article reads slashdot. Viz. from the McWeek article: It would be like calling a motor, transmission and a suspension a car; there's a lot more to making a car (or an operating system) nowadays. About a week before that yours truly wrote in slashdot:
Defining an operating system is similar to defining a car. Strictly speaking four wheels and an engine make a
car. In practice, you wouldn't even consider purchasing something without air bags, seat belts, cup holders,
padded seats, cargo space, windshield wipers... you get the point.
Linux took a giant step when Linus finally clued into this a while ago. He wrote in Communications of the ACM in April 1999: The most esciting developments for Linuz will happen in user space, not kernel space. He sees the OS as a program environment encompasing from kernel all the way to user guis.
One economist researched productivity gains in the 80's from computer purchases and found there were none. This has become known as the productivity paradox, and is now accepted mantra in popular culture, as well as some academic circles.
Let us consider the example of a department store I happen to be familiar with. Sometime around the late 70's they became fully computerized in their accounting department. At that time they let go a staff of 40 accountants and over 100 clerks.
The productivity gains are obvious: 140 people taking home a combined $5M a year in wages replaced with a computer system costing $1M.
So why did these productivity gains do not show in the bottom line (and thus in the economist study)?
Simple, it turns out that once the accounting system was installed, the managers didn't just sit back and wait until the savings rolled in.
To the contrary. They noticed the gains and told their IT people "wow! the savings are enormous. Is there any other such savings lying around?".
The IT people replied "indeed, we could computerize your point-of-sales registers, manage your inventory and payroll as well as upgrade your accounting package to have better tracking of overdue accounts and many other such things".
To which the managers replied "take this $4M dollars we just saved, and go get more programmers, software and hardware so you can carry on!". The savings from these were again reinvested into more programs and more hardware. Further creating more savings and more reinvestment.
As you can imagine, this cannot go on forever. At one point or another the IT department will say "at long last, after thirty years of hard work, there is no obvious process left to computerize".
That day the company stops replacing computers and software every year and moves to a slower replacement, upgrade and development cycle. Suddenly there are all these savings that start hitting the bottom line and we go from a "productivity paradox" to the "surprising productivity gains of the new economy".
There is truly no reason to be surprised about either of the two phenomena, as long as you understand the concept of reinvesting 100% of your productivity gains.
Slashdot Mirror
Syntax is not a trivial problem between advanced applications. Designing a protocol that is expandable, be it for config files or interprocess communication is a pain in the neck. We started using SGML for config files and interprocess communication in 1995. You have no idea how much work we saved by using standard parsers and a protocol that wouldn't break if we added a column to the message...
it does not, however, assault the intractable problems of semantics.
True enough. XML only gets you half-way there. Isn't that still better than nothing? Moreover, since the semantics is intractable, as you well state, the solution is to manually provide syntactic markings through, you guessed it... XML tagging.
After 6 years of adminning systems, I feel more secure with each daemon checking their config files than passing it onto an independent parser daemon.
Reality check. What is likelier to be buggy: a one-off parsing routine or a well established and universally tested parser such as SAX?
We know the answer to that one since the whole open source movement is predicated on it. The publicly available routine will be less buggy.
The world is moving en-masse to XML. Yes, it is overhyped (just as high-level languages, structured programming, OOP and Java were in their time). Yet all of those were clear steps forward in computing.
Same goes with XML. Linux can be either ahead of the curve, or behind it, always destined to be a late copy of a thirty-year old operating system.
Now that Linux is stable and of amazing quality it is time to start looking towards the future and make sure a good operating system becomes hands down the best.
Yeap, I might lose Karma on this one, but hey I'll die for "the greater good".
Linux reflects the needs of its developers (their itches, to use Eric Raymond's terms). High security and restricted access are not common needs for people who enjoy hacking code, and thus the extreme weakness of the Linux security model, of which ACLs are but one component.
Real life often require moving conservatively on issues which could have a major impact, and it is a tried and true engineering technique.
Hopefully in a year from ICANN will aprove a few dozen more domains based on the experience from things such as .museum.
The W3C is corporate driven. You have to pay a hefty membership sum to be able to contribute. This is anathema to the way the Internet grew, where people became net.goods by contributing code.
It also suffers from the standards-body slow-as-molasses syndrome. I have participated in several working groups and they sem to work under a mandate to make as few changes as possible.
An effort such as HTML++ got delayed time and time again.
To reiterate, XML is here only because it originated outside the W3C umbrella.
Oh boy, how soon do they forget.
Netscape added Javascript in an attempt to forestall Sun's deployment of Java. Java got added only when Sun and Netscape signed a truce along the lines: "we, Sun, allow you, netscape, to use the Java moniker on your completely unrelated scripting language. In exchange you agree to support Java on your browser".
They created the first (or one of the first) web application servers - NAS, now iPlanet.
Another one... The first server was NCSA httpd, and many web sites transition from NCSA httpd to Apache without ever going through NSA/iPlanet.
People didn't see the web as a platform at first.
True, but by the time Netscape was talking about it, others had also grasped it, not least of which was Sun with their Java language, whose existance was made public in 1994.
It was Netscape that pushed this idea, evangelized it.
This much I can give them credit for. James Clark drove the point over and over until Bill Gates finally got it, and turned "hardcore on the internet". Gates immediately promoted the underlings who had spent the previous year trying to explain this to him.
To deny that is to overlook Netscape's primary contribution to the internet.
Netscape's primary contribution to the internet is developing a cool and fast web browser which was made available for free.
Netscape had nothing to do this. The web is popular as a platform because it takes the fabled client-server architecture to the masses.
The benefits of a client-server architecture became apparent to all in the late 80s, but until the web appeared, writing a decent client-server application either required an advanced degree on networking and distributed systems, or the purchase of a closed-platform solution.
All that changed with the arrival of the web. You could flush all of netscape's buggy code down the toilet and people would still have developed for the web. Is the sensible thing to do in most cases.
AOL and Netscape deserve scorn for claiming the high moral ground of standards and openess only when they are losing. As soon as the have a dominant position they piss on them, such as with the blink tag.
I agree. We implemented at least one of the patents before the date at a company I used to work for (Database copying) without ever considering it anything less than the obvious way to go and at least two other I talked about with a business partner as an obvious way of how to leverage the internet for business (Double click, Sight sound).
One-click falls squarely in this category. Once you are logged in (via password or cookie) making a sale without request of further information is obvious. Access any intranet site. Are you asked time and time again for your password? Or rather the system keeps track of who you are and shows you only those things you can look at? This is esentially undistinguishable from One-click, and the fact that money is exchanged at the end of the transaction is not non-obvious in any way.
No. They twin any ticket bought from selected aircarriers (AA, Delta and some others, if I remember correctly). I understand the plan changed this year with the talk about mergers and all, so read the fine print before enrolling...
Read the fine print. "Free companion ticket anywhere in the world". That is worth more than $300, and the sole reason why I upgraded to platinum six years ago...
Just yesterday I had an incident with a Sprint "customer advocate" which refused to allow international calls from my digital cell phone.
Almost no business would turn away a platinum american express, six-figure-income programmer as a risky and undesirable customer, yet that is what Sprint did.
I think those are attitudes derived from times when the corporation, not the customer was king...
I don't know if 180K H1B visas is too few or too many, but facts are facts, and it bothers me that so many moderators rank the "no such shortage" posts so highly, when actual objective measures to the contrary are readily available...
Once again Apple demoes a supper cool operating system that is to be released "next year". This story is getting pretty old with me...
That, my friends, is how news are manufactured just to stir up controversial issues and sell more copies. I wonder how many /.ers registered today in the NYT web site. The chief NYT editor should go to bed happy today.
Why are people signing their e-mail with PGP/GPG?
When I was young, the advice from grown ups was "do not sign anything you don't have to, be it a contract, a letter, a memo, anything. If you sign it it means that you meant it, if you don't is just idle chatter".
So, /.ers out there: how about it, why do you sign your e-mail letters?
Surprisingly, the majority of the opinion here seems to be against copyright protection for music and patent protection for new inventions.
In real life, most peoples wages depend on copyright. Even RMS uses copyright protection to enforce its GPL terms.
That doesn't excuse stupid patents from the USPTO or judges that issue brain dead judgements covering all benefits from copyright, which is not what the law says....
Yeah right. They are not called religious wars for nothing.
Did you see the "honest" reaction to Miguel de Icaza's criticism of Linux? The juvenile puns on M$ that quickly get moderated to 4 Funny? How long did it take to the Linux community to admit that its UI sucks (thus LUIGI)? How much longer before they dump the retarded X-windows architecture?
The net has over 100 million users, the big three credit card companies have nearly a combined billion cardholders, yet according to Cringely the 3.3 million customers of PayPal makes them "unassailable".
In fact, I'd say that 3.3 million users is a very small set of net users, and that the real challenge is to reach those who are not geeks nor addicted to online auctions.
What side effects? Caffeine, like Aspirin seem to be among those rare drugs with almost no bad side effects, when consumed in moderation.
Apple has legendarily screwed their customers, starting with making the Mac incompatible with the Apple ][.
But somehow Apple always comes accross as the good guys because they use a rainbow logo....
Don't be so defensive.
Intelligent self-criticism is not mudslinging.
More importantly who reads Wired nowadays? The artsies took over about four years ago, overhyping mumbo-jumbo stuff such as Idealab and the Santa Fe institute. The internet savvy gave up on wired a long time ago. They now read nerve, suck, salon, redherring and of course /. All of them on-line.
That definition is old and incorrect. Trust me, I have three degrees in Computer Science. Some old OS textbooks still quote that antiquated old definition. The modern definition of OS is a system that provides all commonly required services. Notice that this is a moving target.
To quote John Ousterhoot (of Tcl fame) back when he was teaching the OS course at Berkeley (1980s): The essence of an operating system lies in the services it provides to user programs
Are hard drives common? Yes, and therefore the drivers are part of the OS. Are hand-held wireless scanners common? No, and thus their drivers do not come with the OS.
Saying that an OS is a program loader + file manager is tantamount to saying that a computer is a collection of vacuum tubes hardwired together to perform a computation. At some point in the past these definitions were correct. Today they are outdated and have been updated by lead researchers in the field (such as Ousterhoot). Problem is they haven't made it into some texts. But it has into others. Silberschatz writes in Operating Systems Concepts:
This definition is closer to the McWeek one that to the one "from anybody who has a degree in CS".By the way, I think the author of the article reads slashdot. Viz. from the McWeek article: It would be like calling a motor, transmission and a suspension a car; there's a lot more to making a car (or an operating system) nowadays. About a week before that yours truly wrote in slashdot: Defining an operating system is similar to defining a car. Strictly speaking four wheels and an engine make a car. In practice, you wouldn't even consider purchasing something without air bags, seat belts, cup holders, padded seats, cargo space, windshield wipers... you get the point.
Linux took a giant step when Linus finally clued into this a while ago. He wrote in Communications of the ACM in April 1999: The most esciting developments for Linuz will happen in user space, not kernel space. He sees the OS as a program environment encompasing from kernel all the way to user guis.
The productivity gains are obvious: 140 people taking home a combined $5M a year in wages replaced with a computer system costing $1M. So why did these productivity gains do not show in the bottom line (and thus in the economist study)?
Simple, it turns out that once the accounting system was installed, the managers didn't just sit back and wait until the savings rolled in. To the contrary. They noticed the gains and told their IT people "wow! the savings are enormous. Is there any other such savings lying around?".
The IT people replied "indeed, we could computerize your point-of-sales registers, manage your inventory and payroll as well as upgrade your accounting package to have better tracking of overdue accounts and many other such things".
To which the managers replied "take this $4M dollars we just saved, and go get more programmers, software and hardware so you can carry on!". The savings from these were again reinvested into more programs and more hardware. Further creating more savings and more reinvestment. As you can imagine, this cannot go on forever. At one point or another the IT department will say "at long last, after thirty years of hard work, there is no obvious process left to computerize".
That day the company stops replacing computers and software every year and moves to a slower replacement, upgrade and development cycle. Suddenly there are all these savings that start hitting the bottom line and we go from a "productivity paradox" to the "surprising productivity gains of the new economy".
There is truly no reason to be surprised about either of the two phenomena, as long as you understand the concept of reinvesting 100% of your productivity gains.