Carver Mead at Caltech used to say that transistor logic is digital by design not by nature.
What he meant is that transistors are inherently analog devices. We just run them at full saturation levels (almost) all the time, so that the output is flat.
He went on to demonstrate that rather impressively by building the first neural networks in VLSI, by using CMOS transistors operating in the near-linear response range. (This was more than 10 years ago now, so I can't remember exactly when that was.)
Check out Squeak, the free, portable Smalltalk machine. Like all Smalltalks, Squeak runs in an "image". The image is your entire language, programming environment, and execution environment, all at once.
The interpreter, programming tools, and even the GUI all exist as long-lived objects in this large (sometimes very, VERY large) memory space. When you aren't using Squeak, the image gets stored as a file on disk.
There are also projects to run Squeak on bare metal--no intermediate operating system like Windows or Linux. Squeak itself becomes the operating system.
This memory technology would be ideal for a Squeak machine. The image would always live in NVRAM. In such a case, there isn't a distinction between the operating system as it exists in static form (files on disk) and executing form (code in memory). There are always just objects in memory. Very elegant.
I did read the article. I chose to make a politically pointed, satirical remark. (For your education, satire often uses exaggeration and rhetorical devices to make a point.) This obviously exceeded your capacity.
As far as I can see, the biggest problem with/. is the utter lack of a sense of humor among people like yourself. The second biggest would be a complete lack of courtesy.
For that matter, what qualifications do you have to judge any of the following? a) My intelligence b) My fields of expertise c) Fields in which I lack expertise d) Your own moral and editorial superiority
Eclipse brings in some features from VisualAge for Java that I've always loved. First, every single time I hit "Save", it compiles. I get immediate feedback. Second, the scrapbook rules. Being able to try a bit of Java code at any time is great.
I'm quite happy with the painless CVS integration. Eclipse has the easiest CVS/SSH setup I've ever seen. I'm able to reference multiple CVS repositories from my workspace without even thinking about it.
The different perspectives take getting used to. I still get lost from time to time. I don't know which one I'm in and I don't know it until I get the "wrong" context menu. It doesn't help that all the perspective icons look alike.
Overall, I'm fully supporting Eclipse--even to the point of recommending it to my clients.
I've often observed exactly the same thing. I've seen issue logs and defect reports kept in spreadsheets on a network drive, so they can be shared (imperfectly). Likewise, product data, vendor information, and lots and lots of project schedules.
I know that spreadsheets are not the ideal tool for any of these purposes. The point, though is that they are being used for these purposes. Rather than shrug and say, "Well, users are goofy." I'd like to figure out what makes spreadsheets appealing.
Most people use it as a kind of schema-less database-lite. Do away with all the troublesome database stuff like referential integrity and rigid schemas and it becomes simple enough for the average non-abstract-thinker. It's infinitely malleable, they don't have to go through ranks of DBAs just to add a column.
New column? Insert!
New table? Add a sheet.
The main usage pattern spreadsheets don't support well is sharing. Instead of sharing a single spreadsheet, most sharing is through copying. More data lives in spreadsheets in email inboxes than in the knowledge repositories. (Of course, these usually get printed out as soon as they are received, and used as reference long after the original has been revised.)
I've often thought that this usage "loose database" would be well served by some kind of multi-user spreadsheet--a common space where people working on a project could share semi-structured data.
I have the same reaction to this as I do to the articles about quantum entanglement.
How the heck are you going to get a single photon to go large distances without getting absorbed? Even in space, if the photon hits a single atom, it will get absorbed, causing an electron to be excited. When the electron "leaps" back to a ground state, emitting a photon, isn't this a new photon?
I would think that this would lose any previously known polarization. If I'm wrong, please explain how a photon retains its "identity" even after being absorbed.
Imagine that this isn't in space, but in the atmosphere. Plenty of matter to interfere with long-range transmission of individual photons. Fiber-optic cables? I dunno.
If Linux wants to be on alot of desktops then this type of memo isn't going to get it too far.
Danger! Anthropomorphism dead ahead!
"Linux" isn't an animate being. It doesn't want anything.
Even "The Linux Community" is a just a convenient label for a large group of individuals who happen to share an attribute. (And a relatively unimportant one at that.) It is a mistake to assign the ambitions some members of that group to every member of that group.
Each and every developer, user, and contributor is free to work under their own terms. Nobody can force the Abiword developers to work more hours or provide more end-user support. They'd just stop.
At the same time, if you are really committed to the idea of commerical-level product support for end-users, you are free to start a company, fork the code, and have your own Abiword.
Wow! What does your company do with that beast? What kind of reviews & presentations benefit from that kind of horsepower? I'm guessing it's not for displaying PowerPoint with 4-foot high bullets.
I hear ya. I got the worst feeling when I realized that everything he said makes sense... from a certain twisted point of view.
It's like this statement I heard about Microsoft once. I can't remember it exactly, but it basically said that every single decision made about the design, features, software, whatever all comes down to one question, "How does this increase market share?"
It's incredibly depressing to hear somebody say that aesthetically pleasing, technically adept code will always lose to dreck and hackery. It's even worse when you hear yourself start to agree. Bleagh. I'm gonna go take a shower.
This is not really intended to shut SONICblue down. As the article states, the defendants and the plaintiffs are also negotiating a business deal. This lawsuit is nothing more than a pressure tactic designed to get a more favorable deal.
Actually, when Raymond first wrote tCatB, the "Cathedral" he referred to was GCC! It wasn't originally about closed vs. open source. It was examining why Linux was developing and evolving much "faster" than GCC. His conclusion was that the centrally designed and planned approach of the GCC team (a fairly closed group of developers, even though their software is free) led to slower development than the relative free-for-all of the Linux kernel.
It was only after the initial publication of his paper that people started to think Cathedral == closed source.
Professor Goodstein happens to be very active with the Caltech drama club and the Pasadena Playhouse. (Or at least he was when I was there a few years back.)
I actually saw him performing Shakespeare.
I attended Caltech, where Professor Goodstein taught freshman physics. He was one of those rarities: a tenured professor that still loved to teach. I believe that he is honestly motivated by reaching out to people and watching the spark of knowledge kindle.
At that time, he was also active with the L.A. county school system, trying to improve science education for the entire system. (And that's in addition to being Vice-Provost.)
You might consider learning something about an individual before you stereotype them.
I wondered about that, too. But, you know, AC probably wasn't even trying to find a powerful venue. I read this as a personal decision--a moral stand. I don't think he expects his own decision to single-handedly change the situation. He just will not be a part of it.
I have heard stories of professors forcing their students to release coursework under GPL... As a derivative work, all the resulting coursework inherited GPL.
Thanks for bringing this up... There's an important point here that deserves illustration. The students' code would indeed be considered a "derivative work". Therefore, if they distributed binary versions of their software, they would have to make the source code available under the terms of the GPL.
The distinction is that nothing would force the students to release anything whatsoever. The GPL doesn't force you to release your derivative works. It says that if you do distribute your works, you must also provide the same benefits that you got.
I think this is a commonly misunderstood point. In-house software, web server software, etc. never gets distributed. Therefore, even if you use GPL code to create a web site (for example), you need not release your source code.
Now, in the case of the students you mentioned, it's not clear whether they got significant benefit from using the header file (other than on their grade). I'm more than a little leery of being FORCED to adopt the GPL to pass the class.
(It begins to sound like the political correctness movement.)
The idea that the three branches of government consist of the Executive, Legislative, and Judiciary is a common misconception. It was propagated by a bunch of textbooks written in the early 70's, and it just isn't correct.
The checks and balances built into our Constitution say nothing about "judicial review" or the Supreme Court striking down laws. In fact, the three branches that were meant to be checked and balanced were the Executive, Senate, and House of Representatives. This situation where both Houses act as a single unit is an aberration of bicameralism.
The whole principle of "judicial review" was invented out of whole cloth in 1803. In Marbury v. Madison the Court declared a law passed by Congress to be unconstitutional. The reasoning went like this (from the person of Chief Justice John Marshall):
I took an oath of office in which I swore to uphold and defend the Constitution.
This law conflicts with the provisions of the Constitution.
I cannot uphold both this law and my personal oath of office.
This law is an act of man; my oath was sworn before God.
Therefore, I find that this law cannot be upheld.
Sort of a nice bit of logic used to expand the powers of the court. If memory serves, I think all of our military officers also swear to uphold and defend the Constitution against all threats foreign or domestic. You don't see them going about striking down laws. (Or if they do, nobody else takes that as authority!)
Absolutely right! The vendors have made their choices and now we consumers get to make ours.
Assuming, of course, that we know this is being done. A single consumer shopping at the site would have no reason to suspect that he is getting a different price than the next bloke.
That's why items like this are worth posting, so everybody can make their (supposedly) rational decisions armed with complete information.
On a cluetrain sort of note, this also points out why dynamic pricing doesn't work when the marks can talk amongst themselves.
Slashdot Mirror
They also bought Delano and Viant. They're probably in the multiple 100's, possibly in the 1000's.
Clearly, the way to get laws passed to protect privacy rights is to get a drippy movie about it on Lifetime.
Carver Mead at Caltech used to say that transistor logic is digital by design not by nature.
What he meant is that transistors are inherently analog devices. We just run them at full saturation levels (almost) all the time, so that the output is flat.
He went on to demonstrate that rather impressively by building the first neural networks in VLSI, by using CMOS transistors operating in the near-linear response range. (This was more than 10 years ago now, so I can't remember exactly when that was.)
Check out Squeak, the free, portable Smalltalk machine. Like all Smalltalks, Squeak runs in an "image". The image is your entire language, programming environment, and execution environment, all at once.
The interpreter, programming tools, and even the GUI all exist as long-lived objects in this large (sometimes very, VERY large) memory space. When you aren't using Squeak, the image gets stored as a file on disk.
There are also projects to run Squeak on bare metal--no intermediate operating system like Windows or Linux. Squeak itself becomes the operating system.
This memory technology would be ideal for a Squeak machine. The image would always live in NVRAM. In such a case, there isn't a distinction between the operating system as it exists in static form (files on disk) and executing form (code in memory). There are always just objects in memory. Very elegant.
I did read the article. I chose to make a politically pointed, satirical remark. (For your education, satire often uses exaggeration and rhetorical devices to make a point.) This obviously exceeded your capacity.
/. is the utter lack of a sense of humor among people like yourself. The second biggest would be a complete lack of courtesy.
As far as I can see, the biggest problem with
For that matter, what qualifications do you have to judge any of the following?
a) My intelligence
b) My fields of expertise
c) Fields in which I lack expertise
d) Your own moral and editorial superiority
So in a nation where parents can only have 1 child, they are cloning humans...
Yeah, that makes sense.
There isn't really a difference. All wars are economic.
Heh-heh.
.exe's anyway!"
I do like the statement, however, that linux users are less likely to open unknown attachments. Says quite a lot about our community right there.
<sarcasm>
It says, "Pine makes it really frickin' hard to run a binary, and all my mother-in-law sends me are
</sarcasm>
Eclipse brings in some features from VisualAge for Java that I've always loved. First, every single time I hit "Save", it compiles. I get immediate feedback. Second, the scrapbook rules. Being able to try a bit of Java code at any time is great.
I'm quite happy with the painless CVS integration. Eclipse has the easiest CVS/SSH setup I've ever seen. I'm able to reference multiple CVS repositories from my workspace without even thinking about it.
The different perspectives take getting used to. I still get lost from time to time. I don't know which one I'm in and I don't know it until I get the "wrong" context menu. It doesn't help that all the perspective icons look alike.
Overall, I'm fully supporting Eclipse--even to the point of recommending it to my clients.
I've often observed exactly the same thing. I've seen issue logs and defect reports kept in spreadsheets on a network drive, so they can be shared (imperfectly). Likewise, product data, vendor information, and lots and lots of project schedules.
I know that spreadsheets are not the ideal tool for any of these purposes. The point, though is that they are being used for these purposes. Rather than shrug and say, "Well, users are goofy." I'd like to figure out what makes spreadsheets appealing.
Most people use it as a kind of schema-less database-lite. Do away with all the troublesome database stuff like referential integrity and rigid schemas and it becomes simple enough for the average non-abstract-thinker. It's infinitely malleable, they don't have to go through ranks of DBAs just to add a column.
New column? Insert!
New table? Add a sheet.
The main usage pattern spreadsheets don't support well is sharing. Instead of sharing a single spreadsheet, most sharing is through copying. More data lives in spreadsheets in email inboxes than in the knowledge repositories. (Of course, these usually get printed out as soon as they are received, and used as reference long after the original has been revised.)
I've often thought that this usage "loose database" would be well served by some kind of multi-user spreadsheet--a common space where people working on a project could share semi-structured data.
I have the same reaction to this as I do to the articles about quantum entanglement.
How the heck are you going to get a single photon to go large distances without getting absorbed? Even in space, if the photon hits a single atom, it will get absorbed, causing an electron to be excited. When the electron "leaps" back to a ground state, emitting a photon, isn't this a new photon?
I would think that this would lose any previously known polarization. If I'm wrong, please explain how a photon retains its "identity" even after being absorbed.
Imagine that this isn't in space, but in the atmosphere. Plenty of matter to interfere with long-range transmission of individual photons. Fiber-optic cables? I dunno.
It's like the man said, the average user only hits 20% of the features. The kicker is that each user hits a different 20%.
If Linux wants to be on alot of desktops then this type of memo isn't going to get it too far.
Danger! Anthropomorphism dead ahead!
"Linux" isn't an animate being. It doesn't want anything.
Even "The Linux Community" is a just a convenient label for a large group of individuals who happen to share an attribute. (And a relatively unimportant one at that.) It is a mistake to assign the ambitions some members of that group to every member of that group.
Each and every developer, user, and contributor is free to work under their own terms. Nobody can force the Abiword developers to work more hours or provide more end-user support. They'd just stop.
At the same time, if you are really committed to the idea of commerical-level product support for end-users, you are free to start a company, fork the code, and have your own Abiword.
Wow! What does your company do with that beast? What kind of reviews & presentations benefit from that kind of horsepower? I'm guessing it's not for displaying PowerPoint with 4-foot high bullets.
It's like this statement I heard about Microsoft once. I can't remember it exactly, but it basically said that every single decision made about the design, features, software, whatever all comes down to one question, "How does this increase market share?"
It's incredibly depressing to hear somebody say that aesthetically pleasing, technically adept code will always lose to dreck and hackery. It's even worse when you hear yourself start to agree. Bleagh. I'm gonna go take a shower.
This is not really intended to shut SONICblue down. As the article states, the defendants and the plaintiffs are also negotiating a business deal. This lawsuit is nothing more than a pressure tactic designed to get a more favorable deal.
It was only after the initial publication of his paper that people started to think Cathedral == closed source.
Mine too. I may change it to October 11th.
Professor Goodstein happens to be very active with the Caltech drama club and the Pasadena Playhouse. (Or at least he was when I was there a few years back.)
I actually saw him performing Shakespeare.
I attended Caltech, where Professor Goodstein taught freshman physics. He was one of those rarities: a tenured professor that still loved to teach. I believe that he is honestly motivated by reaching out to people and watching the spark of knowledge kindle.
At that time, he was also active with the L.A. county school system, trying to improve science education for the entire system. (And that's in addition to being Vice-Provost.)
You might consider learning something about an individual before you stereotype them.
Not everything is a freedom issue. This is just a request.
Thanks for bringing this up... There's an important point here that deserves illustration. The students' code would indeed be considered a "derivative work". Therefore, if they distributed binary versions of their software, they would have to make the source code available under the terms of the GPL.
The distinction is that nothing would force the students to release anything whatsoever. The GPL doesn't force you to release your derivative works. It says that if you do distribute your works, you must also provide the same benefits that you got.
I think this is a commonly misunderstood point. In-house software, web server software, etc. never gets distributed. Therefore, even if you use GPL code to create a web site (for example), you need not release your source code.
Now, in the case of the students you mentioned, it's not clear whether they got significant benefit from using the header file (other than on their grade). I'm more than a little leery of being FORCED to adopt the GPL to pass the class. (It begins to sound like the political correctness movement.)
The checks and balances built into our Constitution say nothing about "judicial review" or the Supreme Court striking down laws. In fact, the three branches that were meant to be checked and balanced were the Executive, Senate, and House of Representatives. This situation where both Houses act as a single unit is an aberration of bicameralism.
The whole principle of "judicial review" was invented out of whole cloth in 1803. In Marbury v. Madison the Court declared a law passed by Congress to be unconstitutional. The reasoning went like this (from the person of Chief Justice John Marshall):
- I took an oath of office in which I swore to uphold and defend the Constitution.
- This law conflicts with the provisions of the Constitution.
- I cannot uphold both this law and my personal oath of office.
- This law is an act of man; my oath was sworn before God.
- Therefore, I find that this law cannot be upheld.
Sort of a nice bit of logic used to expand the powers of the court. If memory serves, I think all of our military officers also swear to uphold and defend the Constitution against all threats foreign or domestic. You don't see them going about striking down laws. (Or if they do, nobody else takes that as authority!)Cheers,
--Mike
Absolutely right! The vendors have made their choices and now we consumers get to make ours.
Assuming, of course, that we know this is being done. A single consumer shopping at the site would have no reason to suspect that he is getting a different price than the next bloke.
That's why items like this are worth posting, so everybody can make their (supposedly) rational decisions armed with complete information.
On a cluetrain sort of note, this also points out why dynamic pricing doesn't work when the marks can talk amongst themselves.