I'm the project leader for H2O. I'm not sure where Jonathan got the $1mil figure -- my estimate is well less than half of that, which is really not much money at all for professional software development. That money has paid for about about 30 man months of development time, basically 2-3 senior engineers working for about a year to go from conception to a well tested code release + publicly available site.
More importantly, the Rotisserie is far, far from a glorified message board. It is, in fact, one of the very few true, recent innovations in online discussions. It implements a radically different approach to online discussion that solves many of the problems that people generally make about online discussions -- that the quality of the posts is often very poor, that boards are more often than not balkanized into narrow interest groups that merely agree with one another, and that many more people lurk than participate in discussions. The tool uses a combination of techniques to combat these issues, which are especially important for facilitating meaningful academic discouse but are also vital for conducting any thoughtful, productive discussion online.
First, the system slows down the discussion into semi-synchronous rounds. Every discussion is broken up into a series of discrete rounds, and no rotisserie post is published until the end of the given round. This structure encourages people to take the time to put considerable thought into their posts rather than trying to post as quickly as possible to garner the most attention.
Second, the system democratizes the discussion by automagically routing posts between users for further response. After the first round is over, each first round post is assigned to a specific other user for further response. The discussion can continue in this way for as many rounds as the discussion creator desires. This structure encourages everyone to participate equally in the discussion, allowing smaller voices equal weight to large ones. This structure also encourages more careful response, since every post has a very good chance of getting a response (posters are encouraged by the likelihood of a careful critique, in both the carrot and the stick senses).
Last, the system allows for discussion between different projects (projects are loosely analogous to courses, though they can also be less formal, ongoing centers for discussions around a common topic). This combats the balkanization problem by bringing genuinely differing views into play for a given discussion. On a less idealistic scale, it allows for different courses within a given school or, even better, different courses at different schools to discuss with one another, giving students the opportunity to get exposed to potentially radically differing frameworks of thought than those taught by their own professors.
We have been using some form of a rotisserie tool at the Berkman Center for several years now and have been using this particular incarnation of the tool for almost a year now. Both teachers and students report great success in using the tool. It really works. It's our hope to encourage its use beyond the academy for any group that wants to create the kind of productive, meaningful discourse that is difficult with traditional threaded messaging systems.
The further plans for the project are also potentially groundbreaking. We plan to create a collaborative course development system that will allow teachers freely to share their syllabi with one another and easily connect with other courses exploring similar topics. Teachers are currently limited pretty strictly to their own local resources and those of the propietary text book companies when creating course content. H2O will make it possible for teachers to participate in the same kind of community based production demonstrated by the free software world and by the bevy of other such successful efforts (cddb, wikipedia, kuro5hin, etc)
There are still solutions that allow no meaningful information to be sent. For example, why not have the client just ask for new updates since a given date and cache the rest? That took me all of about 15 seconds to think up and would result in far less bandwidth use than sending the user every upgrade applicable to her system every time she connects.
Either 1) privacy is just not a factor for the folks at all or 2) they want the data for other uses. Most likely it's the former, but the fact that the makers of the 95% market share OS don't care enough about privacy to make it even a small concern when designing systems like this is Really Scary, maybe scarier than them purposefully collecting my data, because at least then there's the possibility that they'll be careful with my data once they've got it.
It's very easy to do this -- just have the server send a list of all of the available updates to the client and have the client figure out what it needs to update. It requires a little more bandwidth, but this extra bandwidth of downloading the list of a few hundred or so updates is miniscule compared to the 30M service packs downloads. It certainly requires less bandwidth than uploading a list of every piece of software installed on the client.
The use of this kind of spyware is why I refuse to use redhat's up2date program, which also sends client information to the server during updates. Microsoft and redhat made the same conscious design decision to make their update software spy on their users (though at least redhat is honest about it).
For an example of a program that does the Right Thing, see autoupdate for redhat, which just downloads the list of available updates and checks those against the ones installed on your computer. All it requires to work is some client-side logic and access to an FTP server with the upgrade packages -- no spying necessary.
Actually, I think it's not such a good thing. The fact is that our grandmothers have been buying mostly unserviceable cars for years now, and they have been pretty happy with them. This situation means that the answer to the traditional free software question "would you buy a car with the hood welded shut" is now "yes."
If the car is cheaper and more reliable, the evidence shows that people will definitely buy it with the hood welded shut, because the vast majority of folks don't know how to service a car these days anyway. And, more importantly, most folks are simply short sightedly selfish -- they will generally choose to save a couple of bucks on a car now rather than hold out for a serviceable car that will save them money in service fees in the long term. And they certainly will not spend even a few extra dollars to buy a serviceable car on the principle that it's important that people be able to understand the basic technologies through which they navigate the world.
You need to recount your CDs or get a new encoder. If your 100 CDs are taking up 30G, you are using 300M per CD. That's about the size the CD data would be if you just gzipped it. 100 CDs at 192Kb should take more like 10G of space.
Cable fraud is a federal criminal offense. You can go to jail for a long time if you commit the offense. Moreover, it is most likely not a breach of contract, since the folks who were stealing the cable likely never signed a contract saying "I will not try to access more programming than the programming for which I am paying." The cable company has no need to make people sign contracts like that precsiely b/c stealing cable access *is* a crime.
This is not such a good argument to make anymore, since most new cars sold these days are so reliant on sophisticated computer systems that only manufacturer licensed technicians with expensive diagnostic equipment from the manufacturer can do anything productive beyond change basic fluids. Even though this situation is not quite the same as having the hood of the car legally welded shut, one can certainly imagine a time in the near future at which car manufacturers will prevent consumers from fiddling under the hood due to Intellectual Property concerns (already, it is likely illegal to muddle with the software that is the brain of the car).
I would argue that this state is a bad one, but the fact is that most people these days do choose to buy cars that have very heavy restrictions on who can service them. So this analogy is probably not a good one to use anymore.
Speech is expressive insofar as it expresses ideas or information about politics, society, religion, etc. The problem with source code is that the courts have viewed it as more functional than expressive as often as not -- in other words, they have found that its purpose is more to make a computer do something than to express an idea.
In some cases, this view is correct. If I send a solitaire program to a friend with a note that says "try out this game. it's fun !", then the code is merely functional. The friend is likely not even to read it before he compiles and runs it. If, however, I send him the same code with a note that says "check out how the renderer in this game implements the Zbuffer !", then I am sending the code in order to express some ideas about how the technology behind the game works. In this sense, source code is clearly expressive of technological / scientific ideas insofar as it is meant to explain those ideas. Clearly, DeCSS can be seen in this light -- it is the clearest possible explanation of exactly how the DVD encryption mechanism works.
In order for the DeCSS code to deserve protection, however, we must show that the ideas it expresses have direct relevance to political and social ideas. Merely explaining one's view of how the law of gravity works should not be protected nearly so strongly as explaining one's view of how a democracy should be run. In this case, however, the technological ideas behind DeCSS are directly related to how freedom of speech evolves in our new digital world.
The movie industry has proposed the DVD encryption format as the principle mechanism for distributing digital movies for the next decade at least. As consumers, we should have the right to make a free choice about whether or not we agree with how their mechanism works. Otherwise, we are abdicating the great power of the digital age (the power of many-to-many broadcast communication) back into the hands of the few movie studio heads. Without a proper understanding of how the mechanism works, however, we can't even begin to make an informed decision about whether to support the mechanism in the marketplace, much less enter into a discussion about whether to support the mechanism.
The same DMCA law that prevents us from speaking about how the DVD encryption mechanism works has stopped Professor Felten and his team of researchers from speaking about how the SDMI protection mechanism works. They intended only to publish a plain English paper, but were afraid of the consequences of former court decisions that ruled that any a plaing English description of how code works is merely functional, and therefore not protected speech.
As it currently stands, the DMCA does not even allow a consumer to test a product that he has bought to make sure that it works as it claims to. I cannot at this time buy a version of Windows2000 and test to make sure that it encrypts passwords as it claims to, unless I get explicit permission from Microsoft to do so. If I did perform those tests, I certainly could not publish my findings. This situation is like forbidding Consumer Reports from testing whether or not a child-safety lock works and then suing them if they had the audacity to publish a report describing how the lock might fail !
If the relevant section of the DMCA is allowed to stand, we will have taken away from the average citizen his most basic rights to make intelligent decisions about and join in discussions over how speech in the new digital world works. In doing so, we will have ruined the one thing that is the Internet's greatest potential strength -- its potential for taking the power of speech out of the hands of the few wards of the current heads of the media congolmerates and back into the hands of the masses of people in the country. That harm is a prototypical example of the harm done by restricting expressive speech.
To preface, I use oracle extensively at work for exactly the reasons that the author explains. I need things like transactions, triggers, and constraints. However, support is *not* a reason you should choose oracle over mysql.
I've had problems with oracle crashing and tried to use the oracle support, for which we pay about $10,000/year.
I ended spending about three hours on the phone, in between waiting on hold and talking to the clueless tech support person. After three weeks of calling and pestering them to fix the problem, they finally came back with a work around that required me to disable a small but significant piece of functionality of the DB to stop it from crashing.
In contrast, I've had problems with mysql three times. Each time ended up being triggered by bad SQL code on my part. Regardless, I always had a response within half a day and a patch and an explanation within two days (usually within a day!). I repeat: I invariably had a patch to address my specific problem within two days of reporting it. And I don't pay the mysql folks squat.
According to this week's Time Magazine, the Human Genome Project is now estimating that they will finish in November of this year, unless I grossly misunderstood the article.
The Newton 2000 was the last Newton model. It was really more of a sub-company notebook than a PDA (it featured a much larger screen and a portable keyboard and was made for tasks like word processing).
I was just pointing out that the release of the alta vista search engine would be Really Cool, b/c there are lots of uses for it that don't involve buying millions of dollars of machinery and setting up Yet Another Web Index. Your 'topologigraphic' map project just furthers my point.
The vast majority of people who use search engines like altavista don't want to index the whole web. They just want to index either their web sites or some database content. For instance, my company builds intranet libraries for financial services companies. Part of the work involves receiving large feeds of stories from various sources, and our customers want to do free text searches on those stories. We use oracle for our main database, but oracle is very bad at solving this particular problem, so we need to use another database (such as the alta vista search engine) to do the text indexing.
For people like me, who are the vast majority of people who want to use the alta vista search engine, the open sourcing of the product (if they will be open sourcing it) is terrific news.
I think you misread the part about valid web sites. That part refers to the Affiliate Network program. In other words, you have to have a valid web site to join the Affiliate Network program.
The article actually doesn't expand on the source code freedom beyond the mention in the title, which is more than a little frustrating.
Many of the rankings they gave were based not on actual tests, but on the claimed features of the products. The two most glaring were the rankings for security and stability. I was floored to see that win2k scored better in both of these areas until I found out that they didn't actually test them. They simply looked at the feature list. win2k has better security becuase it now supports kerberos authentication and better stability b/c it has support for RAID (both of which RedHat has, but that's another matter).
Regardless of how many features win2k has (and perhaps especially due to its number of features), the most important thing is how well they are implemented. It doesn't matter how many features RAID system has if the server blue screens once a day !
The core of your argument seems to be 'TV is stupid, Tivo just enhances TV, so Tivo is also stupid'.
Well, duh. If you don't like TV, you are not going to like the Tivo.
FYI, some people like TV for perfectly valid reasons that you just can't fathom. Really. You are in no position to judge whether TV improves the quality of my or anyone else's life.
Err, if the stock price stays at its current price, it will be the biggest IPO increase ever. I wouldn't call that going "how _every_ recent IPO goes."
Nope. It opened at $300, which means that the first share traded over the public exchange cost $300. $30 was the pre-IPO price, which means that's what people paid before the stock was publically avaliable.:)
I was, but I obviously misunderstood. I apologize.
I obviously have some lingering resentment over the redhat IPO, in which lots of people reported getting letters from redhat after only submitting a bug report or two, if that.
There's obviously no perfect way to do a thing like this, and I do think making an honest attempt at doing it fairly is better than no attempt at all. But I'm still left feeling a bit like I've been slapped in the face a couple of times now. Doubtless lots of other people are feeling just the same.
I've written a fairly well used web application to which I've dedicated ridiculous amounts of time over the past year and a half. I also taught a free, semester long course on Linux, which was a huge effort on my part. I didn't do either to get rich, and both efforts have been rewarding in and of themselves. But I do get a little bitter when I read about people like the above, who've done either nothing or very little to futher free software and are receiving great benefits nonetheless.
I'd definitely feel better if I knew how these decisions are being made. Perhaps even a place I could say "Hey I've done a lot of work for Linux and free software. How do I get considered ?"
Slashdot Mirror
I'm the project leader for H2O. I'm not sure where Jonathan got the $1mil figure -- my estimate is well less than half of that, which is really not much money at all for professional software development. That money has paid for about about 30 man months of development time, basically 2-3 senior engineers working for about a year to go from conception to a well tested code release + publicly available site.
More importantly, the Rotisserie is far, far from a glorified message board. It is, in fact, one of the very few true, recent innovations in online discussions. It implements a radically different approach to online discussion that solves many of the problems that people generally make about online discussions -- that the quality of the posts is often very poor, that boards are more often than not balkanized into narrow interest groups that merely agree with one another, and that many more people lurk than participate in discussions. The tool uses a combination of techniques to combat these issues, which are especially important for facilitating meaningful academic discouse but are also vital for conducting any thoughtful, productive discussion online.
First, the system slows down the discussion into semi-synchronous rounds. Every discussion is broken up into a series of discrete rounds, and no rotisserie post is published until the end of the given round. This structure encourages people to take the time to put considerable thought into their posts rather than trying to post as quickly as possible to garner the most attention.
Second, the system democratizes the discussion by automagically routing posts between users for further response. After the first round is over, each first round post is assigned to a specific other user for further response. The discussion can continue in this way for as many rounds as the discussion creator desires. This structure encourages everyone to participate equally in the discussion, allowing smaller voices equal weight to large ones. This structure also encourages more careful response, since every post has a very good chance of getting a response (posters are encouraged by the likelihood of a careful critique, in both the carrot and the stick senses).
Last, the system allows for discussion between different projects (projects are loosely analogous to courses, though they can also be less formal, ongoing centers for discussions around a common topic). This combats the balkanization problem by bringing genuinely differing views into play for a given discussion. On a less idealistic scale, it allows for different courses within a given school or, even better, different courses at different schools to discuss with one another, giving students the opportunity to get exposed to potentially radically differing frameworks of thought than those taught by their own professors.
We have been using some form of a rotisserie tool at the Berkman Center for several years now and have been using this particular incarnation of the tool for almost a year now. Both teachers and students report great success in using the tool. It really works. It's our hope to encourage its use beyond the academy for any group that wants to create the kind of productive, meaningful discourse that is difficult with traditional threaded messaging systems.
The further plans for the project are also potentially groundbreaking. We plan to create a collaborative course development system that will allow teachers freely to share their syllabi with one another and easily connect with other courses exploring similar topics. Teachers are currently limited pretty strictly to their own local resources and those of the propietary text book companies when creating course content. H2O will make it possible for teachers to participate in the same kind of community based production demonstrated by the free software world and by the bevy of other such successful efforts (cddb, wikipedia, kuro5hin, etc)
There are still solutions that allow no meaningful information to be sent. For example, why not have the client just ask for new updates since a given date and cache the rest? That took me all of about 15 seconds to think up and would result in far less bandwidth use than sending the user every upgrade applicable to her system every time she connects.
Either 1) privacy is just not a factor for the folks at all or 2) they want the data for other uses. Most likely it's the former, but the fact that the makers of the 95% market share OS don't care enough about privacy to make it even a small concern when designing systems like this is Really Scary, maybe scarier than them purposefully collecting my data, because at least then there's the possibility that they'll be careful with my data once they've got it.
It's very easy to do this -- just have the server send a list of all of the available updates to the client and have the client figure out what it needs to update. It requires a little more bandwidth, but this extra bandwidth of downloading the list of a few hundred or so updates is miniscule compared to the 30M service packs downloads. It certainly requires less bandwidth than uploading a list of every piece of software installed on the client.
The use of this kind of spyware is why I refuse to use redhat's up2date program, which also sends client information to the server during updates. Microsoft and redhat made the same conscious design decision to make their update software spy on their users (though at least redhat is honest about it).
For an example of a program that does the Right Thing, see autoupdate for redhat, which just downloads the list of available updates and checks those against the ones installed on your computer. All it requires to work is some client-side logic and access to an FTP server with the upgrade packages -- no spying necessary.
http://www.bitstream.com/categories/products/fonts /vera/index.html
Actually, I think it's not such a good thing. The fact is that our grandmothers have been buying mostly unserviceable cars for years now, and they have been pretty happy with them. This situation means that the answer to the traditional free software question "would you buy a car with the hood welded shut" is now "yes."
If the car is cheaper and more reliable, the evidence shows that people will definitely buy it with the hood welded shut, because the vast majority of folks don't know how to service a car these days anyway. And, more importantly, most folks are simply short sightedly selfish -- they will generally choose to save a couple of bucks on a car now rather than hold out for a serviceable car that will save them money in service fees in the long term. And they certainly will not spend even a few extra dollars to buy a serviceable car on the principle that it's important that people be able to understand the basic technologies through which they navigate the world.
Also check out http://action.eff.org/tinseltown/.
You need to recount your CDs or get a new encoder. If your 100 CDs are taking up 30G, you are using 300M per CD. That's about the size the CD data would be if you just gzipped it. 100 CDs at 192Kb should take more like 10G of space.
http://www.ncta.com/pdf_files/statutes.pdf
errr, no.
Cable fraud is a federal criminal offense. You can go to jail for a long time if you commit the offense. Moreover, it is most likely not a breach of contract, since the folks who were stealing the cable likely never signed a contract saying "I will not try to access more programming than the programming for which I am paying." The cable company has no need to make people sign contracts like that precsiely b/c stealing cable access *is* a crime.
Here's an improved alphabetical index that prevents the annoyance of having to cross reference manually:
http://eon.law.harvard.edu/mscomment/
This is not such a good argument to make anymore, since most new cars sold these days are so reliant on sophisticated computer systems that only manufacturer licensed technicians with expensive diagnostic equipment from the manufacturer can do anything productive beyond change basic fluids. Even though this situation is not quite the same as having the hood of the car legally welded shut, one can certainly imagine a time in the near future at which car manufacturers will prevent consumers from fiddling under the hood due to Intellectual Property concerns (already, it is likely illegal to muddle with the software that is the brain of the car).
I would argue that this state is a bad one, but the fact is that most people these days do choose to buy cars that have very heavy restrictions on who can service them. So this analogy is probably not a good one to use anymore.
Speech is expressive insofar as it expresses ideas or information about politics, society, religion, etc. The problem with source code is that the courts have viewed it as more functional than expressive as often as not -- in other words, they have found that its purpose is more to make a computer do something than to express an idea.
In some cases, this view is correct. If I send a solitaire program to a friend with a note that says "try out this game. it's fun !", then the code is merely functional. The friend is likely not even to read it before he compiles and runs it. If, however, I send him the same code with a note that says "check out how the renderer in this game implements the Zbuffer !", then I am sending the code in order to express some ideas about how the technology behind the game works. In this sense, source code is clearly expressive of technological / scientific ideas insofar as it is meant to explain those ideas. Clearly, DeCSS can be seen in this light -- it is the clearest possible explanation of exactly how the DVD encryption mechanism works.
In order for the DeCSS code to deserve protection, however, we must show that the ideas it expresses have direct relevance to political and social ideas. Merely explaining one's view of how the law of gravity works should not be protected nearly so strongly as explaining one's view of how a democracy should be run. In this case, however, the technological ideas behind DeCSS are directly related to how freedom of speech evolves in our new digital world.
The movie industry has proposed the DVD encryption format as the principle mechanism for distributing digital movies for the next decade at least. As consumers, we should have the right to make a free choice about whether or not we agree with how their mechanism works. Otherwise, we are abdicating the great power of the digital age (the power of many-to-many broadcast communication) back into the hands of the few movie studio heads. Without a proper understanding of how the mechanism works, however, we can't even begin to make an informed decision about whether to support the mechanism in the marketplace, much less enter into a discussion about whether to support the mechanism.
The same DMCA law that prevents us from speaking about how the DVD encryption mechanism works has stopped Professor Felten and his team of researchers from speaking about how the SDMI protection mechanism works. They intended only to publish a plain English paper, but were afraid of the consequences of former court decisions that ruled that any a plaing English description of how code works is merely functional, and therefore not protected speech.
As it currently stands, the DMCA does not even allow a consumer to test a product that he has bought to make sure that it works as it claims to. I cannot at this time buy a version of Windows2000 and test to make sure that it encrypts passwords as it claims to, unless I get explicit permission from Microsoft to do so. If I did perform those tests, I certainly could not publish my findings. This situation is like forbidding Consumer Reports from testing whether or not a child-safety lock works and then suing them if they had the audacity to publish a report describing how the lock might fail !
If the relevant section of the DMCA is allowed to stand, we will have taken away from the average citizen his most basic rights to make intelligent decisions about and join in discussions over how speech in the new digital world works. In doing so, we will have ruined the one thing that is the Internet's greatest potential strength -- its potential for taking the power of speech out of the hands of the few wards of the current heads of the media congolmerates and back into the hands of the masses of people in the country. That harm is a prototypical example of the harm done by restricting expressive speech.
I've had problems with oracle crashing and tried to use the oracle support, for which we pay about $10,000/year.
I ended spending about three hours on the phone, in between waiting on hold and talking to the clueless tech support person. After three weeks of calling and pestering them to fix the problem, they finally came back with a work around that required me to disable a small but significant piece of functionality of the DB to stop it from crashing.
In contrast, I've had problems with mysql three times. Each time ended up being triggered by bad SQL code on my part. Regardless, I always had a response within half a day and a patch and an explanation within two days (usually within a day!). I repeat: I invariably had a patch to address my specific problem within two days of reporting it. And I don't pay the mysql folks squat.
I'll take the mysql support any day.
According to this week's Time Magazine, the Human Genome Project is now estimating that they will finish in November of this year, unless I grossly misunderstood the article.
The Newton 2000 was the last Newton model. It was really more of a sub-company notebook than a PDA (it featured a much larger screen and a portable keyboard and was made for tasks like word processing).
I was just pointing out that the release of the alta vista search engine would be Really Cool, b/c there are lots of uses for it that don't involve buying millions of dollars of machinery and setting up Yet Another Web Index. Your 'topologigraphic' map project just furthers my point.
For people like me, who are the vast majority of people who want to use the alta vista search engine, the open sourcing of the product (if they will be open sourcing it) is terrific news.
The article actually doesn't expand on the source code freedom beyond the mention in the title, which is more than a little frustrating.
Regardless of how many features win2k has (and perhaps especially due to its number of features), the most important thing is how well they are implemented. It doesn't matter how many features RAID system has if the server blue screens once a day !
Well, duh. If you don't like TV, you are not going to like the Tivo.
FYI, some people like TV for perfectly valid reasons that you just can't fathom. Really. You are in no position to judge whether TV improves the quality of my or anyone else's life.
I think VA Linux is still holding about 40,000,000 shares, against the 4,400,000 they sold. So they only sold about 10% of the company. -hal
Err, if the stock price stays at its current price, it will be the biggest IPO increase ever. I wouldn't call that going "how _every_ recent IPO goes."
Nope. It opened at $300, which means that the first share traded over the public exchange cost $300. $30 was the pre-IPO price, which means that's what people paid before the stock was publically avaliable. :)
I was, but I obviously misunderstood. I apologize.
I obviously have some lingering resentment over the redhat IPO, in which lots of people reported getting letters from redhat after only submitting a bug report or two, if that.
There's obviously no perfect way to do a thing like this, and I do think making an honest attempt at doing it fairly is better than no attempt at all. But I'm still left feeling a bit like I've been slapped in the face a couple of times now. Doubtless lots of other people are feeling just the same.
Ditto that.
I've written a fairly well used web application to which I've dedicated ridiculous amounts of time over the past year and a half. I also taught a free, semester long course on Linux, which was a huge effort on my part. I didn't do either to get rich, and both efforts have been rewarding in and of themselves. But I do get a little bitter when I read about people like the above, who've done either nothing or very little to futher free software and are receiving great benefits nonetheless.
I'd definitely feel better if I knew how these decisions are being made. Perhaps even a place I could say "Hey I've done a lot of work for Linux and free software. How do I get considered ?"