At first I thought "rich media" meant Jack Valenti, but now I see that, as parent says, this patent covers editing something like a Flash script over the Internet, with some network application server doing the actual work. However, if you had an X-Windows application that edited such a Flash script, & you exported its display to another terminal, then you're clearly violating this patent, even though you could have done this in the 90s, maybe the 80s.
Even if we ignored that, this patent is obvious: it's a principle of CS that anything you can do on your own box, you can do remotely, thus if the local application isn't patented, then the web application can't be patented either, because it's obvious. I think if the USPTO realized this (although they're systemically disinclined to understand anything, since their revenue comes from *approving* patents), many assinine web patents would go away.
A few weeks ago we saw an article advocating "patents lite", in which the USPTO checks that the patent covers something patentable, but does *not* check for prior art, usefulness, or non-obviousness. The patent is much shorter, like 3 years. The first time it's challenged, the burden is now on the patent-holder to show that it's useful, novel, & non-obvious.
It turns out that de facto, we have patents lite. Clearly, the Patent Office checked nothing before it granted this patent, so we're just waiting for the first lawsuit. We have patents lite, & the system still doesn't work. It's time to end this.
Yeah, but Joel's an ass. Have any of his worshippers here on/. actually *used* something written by Fog Creek or whatever? FogBUGZ, a web-based bugtracker, seems to be his one claim to fame, & it's terribly mediocre. I mean, mostly it works, but the search function doesn't, the UI is inconsistent, & while you can define filters (such as, "my open priority-1 bugs"), you can't share them, which makes them nearly useless. Joel writes a good spiel, but when it comes to coding, his company ain't the shit.
Plus, he argues passionately for paying programmers well & giving them exciting projects, but in at least two cases he hired interns to start his company's most interesting apps.
He wasn't kidding. It sure went downhill fast after 1999. His other opening lines, "we have permanently entered a new economy", and "Novell has again come to be seen as a worthy competitor to Microsoft", are not exactly prophetic (quick check on Yahoo stocks shows Novell's price has ended up pretty much where it was five years ago).
Other disagreements: "most of them would probably turn out to be terrible managers". I strongly disagree. Of the 5 managers I interact with weekly, the 3 who have running code in our systems (i.e., they're promoted developers) dress the worst and manage the best: they tell me my deadlines and my priorities, they ask me what support I need to write code, and they leave me the fuck alone. The 2 who don't have code running on our servers, who were first hired as managers, like to reorganize our hierarchy, introduce burdensome reporting requirements so the execs have more Social Science Numbers to look at, and want to transform our nice offices with *real* *doors* into a miserable cubefarm. I say, promote geeks! Even if they don't want it! I totally agree when he says "you can tell them what to do, but you can't tell them how to do it" (this is far from an original thought of his), but unless your managers are geeks, this approach will leave them feeling powerless and threatened. Managers meddle, it's what they're trained for.
If you want an insightful, thorough, and applicable discussion of all these ideas, as well as many more, some of them *original*, read the Scrum Handbook.
First of all, I resent the vaguely cultural-studies post-structuralist jargon of the article: "Here the heterarchy transcends the firewall and pressure can be applied from without." What's a heterarchy? Is that firewall a metaphorical one? I, for one, do NOT welcome our Foucault-reading post-modern academic overlords.
As for corporate blogging, the most useful blogs I've come across are from important developers in Microsoft (in particular) & also Google, Netscape, Python, etc. A number of times I've been investigating a fairly obscure question about some Microsoft API (shut up, it's my job), & found an excellent answer in a Microsoftie's blog. E.g., some feature seems blatantly missing, I'm searching for it, & the developer mentions in his blog that the feature IS indeed missing but he hopes to implement it in version 3.
This has nothing to do with marketing. I'm not sure what you'd call it in suitspeak, but it's sort of a conversational style of customer support & community-building.
Bhutan is a devout BUDDHIST country, you hoser. And they have TVs now, although when I first read about that (more than 5 years ago), the Bhutanese were quite reasonably shy about appearing on it. Hence they had to cajole some poor sod to read the national news in a pained monologue. If you see some of the recent Bhutanese movies, though, it appears things are changing fast. I have no idea how they weigh their TVs and Buddhas, though.
On topic, I think it's great that they're using Avocado's Number to define the kilo. So let's see, 6.022 x 10^23 avocados would weight ~1.8 x 10^23 kilos, or 28 Earths (this measurement based on the standard platinum-iridium avocado that was shipped to me from Paris, which is significantly larger than a plum, but was free to ship, since any attempt to weigh it would cause fatal recursion).
Government: FOSS, Slashdot, Wikipedia. These aren't replacing regional governments, obviously, they're just small examples of groups negotiating resources in new, interesting ways.
Religion: Subgenius? Forget it.
My argument's based on Third-World Internet users growing faster than US users. So getting their attention should be a major priority for US sites at some point. Besides, if poor countries set up their own DNSes, they could just translate foo.com to foo.co.us, so now I really don't see what all the fuss is about.
Does anyone know what about ICANN's administration the plaintiffs don't like?
I'm reading The Checkbook and the Cruise Missle, in which Arundhati Roy says injustices increase as decision-makers are geographically separated from those affected by the decision. She cites the World Bank in Geneva, and the IMF, the WTO, as examples.
So Third-World countries want power over names? And they think they can accomplish that by moving the naming committee to UN Headquarters in New York? The UN didn't work for poor people in Iraq, or Palestine. Why will it work in the case of Internet names?
This is the first case I know of where software standards have reached the level of world politics. (It's different from software patents in Europe.) I don't think they ever belong there. Software standards have developed reasonably well under Darwinian conditions: it may take decades, but eventually everyone switches to open standards because there's an advantage to being able to communicate. E.g., everyone uses TCP/IP now, not IPX or any other proprietary network protocol. I know, I know, we're still fighting this battle daily, but you can see the positive trend, & it's happening without any legislation or government enforcement.
What I'm getting at is Third World countries should just set up their own root DNS servers. Whatever it is they want -- get rid of the 3-letter root domains? So instead of.com, US sites will have to use.co.us like everyone else? That seems reasonable. If they just set up root DNS servers that don't answer requests for.com (or.org,.gov, etc.), those servers will be more convenient to client hosts in their region. Software will get patched to check both authorities, since it's an easy fix, & US sites will register both types of domains to maximize their availability. Then, over an excruciating number of years, while everyone has to support both naming styles,.com & the other 3-letter domains will die out, & the plaintiffs will have their way.
I'm gonna sound like a Wired columnist, but here goes: The Internet is suggesting new kinds of economics, government, maybe religion.... We should stick with what works, instead of imposing traditional kinds of governance onto the Internet.
And in article, they wrote it beamed the updated attendance record to a teacher's handheld. A slick piece of work, that, & hard to make robust (I have some experience in this area). My intuition has always been that evil people are bad programmers. Something about how their twisted brains can never exude straightforward code....
I bet they never got it working. The article implies they hardly used it before shutting it off.
And, he doesn't even understand himself. Here he is, trying to legislate what we call Linux ("That's GNU-slash-Linux"), as if he owned it. One of the things you have to give up, if you develop open-source (or free, or whatever) software, is the right to be credited as you'd wish. Someone may grab your code, rename & repackage it, strip many of your identifying marks from it, & sell it for a million dollars. That is precisely the freedom I thank Stallman for inspiring us to achieve, & it's exactly what Stallman, now that he's been eclipsed, wants to take away from us.
That said (or ranted), us Slashtrolls' reaction to this is too predictable. Why is the OSS community, on the whole, so antipathetic toward RMS? Is it because he's become such a dogmatic preacher? Is it that he's always been so, but now that we're nearly on top, we'd rather not be reminded of our moral obligations? Is it that we only respect the one with the latest Freshmeat entry?
The reason the web appears so much more vulnerable in this study than in previous studies and general opinion is that they focus on something different from the usual. They're looking at the web, not the internet.
They're looking at the web topologically, as usual, but rather than measuring distance from site A to site B by the minimal number of router hops required, they're measuring the number of clickable links from A to B. In other words, if you started at A.com and had no keyboard, could you click your way to B.com?
The results were that topological diameter was 19 links. Anyone know the diameter of the internet (average traceroute hops from any site to any other)? Furthermore, the overall connectedness is low, so if you took out (e.g.) Yahoo and MSN, you might not be able to click from someone's panda hentai page at Geocities to my Jar-Jar hate site. I can't seem to find in the articles whether this only deals with static linking, or if search engines are accounted for somehow.
This is sort of an odd way to look at the web. Most people don't start from their home page and start clicking until they find something interesting. You start at some place you type in, do a search, make a huge leap into a topologically distant area, then start moving around connected nodes, then make another huge leap. If MSN died and their routers stayed up, would the web be geometrically less useful, as they claim, or just linearly less useful?
I'd say that this focus on the topology of links is really vieaux chapeau now that most people use interactive services to grab information. The web isn't a static, well, "web", anymore.
Although the jobs are currently a little scarce (everyone wants to work for RedHat, I'd imagine, but RH doesn't have that kind of budget yet), working for an Open Source / Free Software company would be a great way to get everything you want.
You asked to make money, do something interesting, and contribute to the world. All of the suggestions above are great for making a contribution to society, but most of them (notably being a high school sysadmin, or going to Africa), won't make you much money and probably won't be technically challenging. Apply to RedHat, IBM, Corel, SGI, SuSE, 3Dfx, etc.--most companies with lucrative and interesting jobs are still at least 35% evil, but if you can get into the cool parts you can contribute to making Free Software the dominant business model, and get a good salary hacking interesting problems.
It's always surprising when a relatively arcane piece of computer science becomes practical.
The Byzantine Generals problem deals with exactly what McNett needs:
"There has to be a security model that is very easy, that doesn't allow a client machine to gain more insight than it should on the nature of a task and that can assure that no one client machine has enough grasp of the project that it can adversely affect the result."
The Byzantine generals problem is formulated similarly. One formulation (the closest to this) is: N generals are on a hilltop, about to attack a city. K are traitors, who will interfere with any protocol in the most damaging way possible. They must agree on some piece of data (the time to attack the city) reliably. Here is a link with some explanations and implementations of the solution.
A commercial "Distributed.com" would have a simpler problem, because they can reliably a) authenticate a computer's identity, so they know if two messages come from the same computer, and b) they can assume that the server isn't a traitor. This will severely reduce the level of redundancy necessary. Still, they must deal with truly malicious nodes, whereas Distributed.net has only had to deal with faulty ones.
As for granulating the data so that K traitorous nodes cannot glean something useful from the data, this should be interesting information theory. I would think that adding some garbage data to calculate from, along with the real stuff, might be a decent cost/security trade-off.
Not that this story really deserves a serious, even semi-serious reply, but...
Researchers have often "determined" whether a drug is chemically addictive by testing whether e.g. a rat will self-administer the drug. Hook the beast up to an IV, demonstrate to it that pushing a lever will give it a drop of smack, and watch. Rats will self-administer opiates and quickly become massively addicted. Caffeine, nicotine, cocaine; the usual suspects. Rats don't seem to like alcohol, THC (pot), or LSD.
So, in light of recent worries about web addiction, will a parrot self-administer the web?
Woah, check out my keywords...Echelon's gonna love this post.
This doesn't count the TCO of the pieces of crap you get. My mother got an eMachine a year ago. It had a fan that sounded like a Harrier, so we sent it back. Every time we sent it back (paying $25 one-way shipping) we got a new machine, meaning we had to set up all the software again, and a new hardware problem always surfaced in the space of two weeks. Going on newsgroups, I discovered that most of these problems are endemic throughout their product line (particularly the unreasonably noisy fan). The FTC should require total cost of ownership disclosures--that machine must have cost us $500 in shipping and time alone.
This is where it's at, folks--cheap space travel. The article claims that the ISRO claims they can launch a moon-orbiter for Rs 350 crore (a crore is 100,000, so 350,000,000 rupees ~= $11 million), which is cheaper than I would have thought possible. Even guessing that they'll go over budget by a factor of three, it'll still be phenomenally efficient.
Of course, the space sation is already five years overdue and over budget by a factor of something like ten, without being more than a quarter completed. But the situation with that was politically and technically quite different.
Sure, the superpowers may say 'been there, done that', but one advance this mission might show us (as well as crazy rocket guy's mission) is how to do space travel cheaply and on short notice. If NASA could do missions this cheaply, they could just send up three at a time to boost their success rate. =)
Reliable, cheap, turn-key space travel is what will bring the future here. Like computers: moving from ENIAC to the Vaio laptop.
Gosh! After the lenient early 90s that saw such monstrosities as MSNBC emerge from the muck, it's good to see the DOJ remembering its duties again. They let AOLTimeWarner merge, of course, but at least they're paying attention. A few more breakups and it will almost feel like the Roosevelt/Taft days again. Well, maybe not....
The Patent & Trademark Office could take a few pointers from the DOJ.
I was expecting something insidiously 'innovative' from C#, given that the COOL vaporware a year or two ago was supposed to be an architecturally as well as syntactically different language from Java, C, and C++. However, C# just looks like a vague mish-mash of Java and C++. Indeed, all but a few features could be implemented in C++ with operators and macros. Syntactically, the only really interesting thing I can see is that
int[,,] a3;// 3-dimensional array of int
is valid syntax, which should remove a few headaches. Nevertheless, I was overall expecting something a little more...more.
A lot of things that are good about the Java platform helped it take off, and lot of things that are bad about the platform keep it circling at low altitude. What really allowed Java to take off as a language, in addition to the platform, was the library and how it is keeping up with recent ideas about 'design patterns'. Java's symbiosis with design patterns, and the fact that Java GUI programming is some of the easiest there is, have been a great asset. What really remains to be seen is how C#'s libraries are accepted by the developer community, because these days it's often the standard libraries that really matter.
It's a shame that C# will be so embroiled in politics from the get-go that we'll never get to see how good (or bad) it actually is. Microsoft has some smart people, and I think the world has a niche for a simplified, compiled, reasonably-performing, garbage-collecting Java/C++. But C# is the child of a tyrant--it can never succeed or fail on its own merits.
This new website, firstgov.gov, will be created at no cost to the government [much deserved snip] in 90 days or less. It will uphold the highest standards for protecting the privacy of its users.
Bill's quite quick on the uptake, you know? He has already mastered the art of vaporware. I envision a brave new world of E-Government(TM), where the irrelevant announcements of new Microsoft products and the pointless promises of politicians will go hand in hand!
...every on-line resource offered by the federal government...
Ah hah--now I see why it will be free, and take only 90 days to build. The http://firstgov.gov/ will simply be a big picture of the IRS's middle finger. But it will take three months for the Feds' first finger to figure out what their middle finger is doing.
...one-stop shopping for government services...
Excuse me, Webmaster, what's the going rate on senators, and is it possible to buy one online?
Wladawsky-Berger seemed to be a very smart, non-marketeering, down-in-the-trenches kind of guy. He spoke with enough easy jargon that C|Net, even in a technical article, had to insert parentheticals explaining what he said. He mentioned SGI (a direct competitor) in an extremely positive way, and didn't take the opportunity to bash Sun. If IBM has people like this working at its highest level ("vice president of technology and strategy"), I have great hopes for their continuing wonderfulness.
How in the world did IBM, famous for its entrenched monopolist corporate culture, manage to turn itself around so quickly and fundamentally?
Does this fall under the category of obligatory Beowulf comment? Does that mean we can dispense with it for this story? =)
I believe the problem with long-range asteroid prediction is not computational power. The N-body problem with only a few dozen good-sized gravity wells in the system is easily simulated with a desktop machine. The problem is the accuracy of our measurements. The magnitude of the error term dominates the significant result pretty quickly as you extrapolate measurements centuries into the future.
What are we to make of the lack of public response to the problem of protecting against meteor collisions? With--what?--3 movies two summers ago about major collisions (okay, one was about a comet, which I don't think anyone would survive), still no one has much of a reaction. I don't know whether to revive my faith in humanity, since those execrable had so little effect on the national consciousness, or to leave it dead, because we never wake up and try to save ourselves until it's too late.
An AC just pointed out something which makes this more interesting. Since the AC will get moderated into oblivion (for good reason, given the needless fucking profanity), I'll repeat his/her point: RedHat and Cygnus are now the same company. Wasabi is distinguishing between oranges and oranges.
We're trying to be Cygnus of this space, rather than the RedHat.
It's interesting that he made a strong destinction (perhaps with a note of derogation?). I suppose the difference is that Cygnus builds tools (many of them proprietary, incidentally) for developing on Unix, for porting between Win32 and Unix, and they customize open source tools and OSen. On the other hand RedHat doesn't customize Linux for specific clients, rather it focusses on the consumer model of one-size-fits-all (or 'a-few-sizes-fit-all').
Note that Cygnus makes most of its revenue from distributing proprietary tools.
N: If clients approach you for NetBSD development are you making sure that it's going to be released under the BSD license?
P: Everything that we can we will. There will be instances where clients come to us for work that will be used in house, or is uninteresting. But we're unequivocably an open-source company, and we want to release virtually everything we do as open source.
They're not going to GPL everything. Cygnus' revenue model relies on proprietary products. There's a lack of information on how Wasabi will actually make money (if neither support nor beach balls seems like a good revenue source). This makes me wonder how well they'll do, and how much work they can give back to the community. Frankly, if a company is committed to GPLing every line of code they produce, RedHat seems like a better business model to emulate.
Slashdot Mirror
At first I thought "rich media" meant Jack Valenti, but now I see that, as parent says, this patent covers editing something like a Flash script over the Internet, with some network application server doing the actual work. However, if you had an X-Windows application that edited such a Flash script, & you exported its display to another terminal, then you're clearly violating this patent, even though you could have done this in the 90s, maybe the 80s.
Even if we ignored that, this patent is obvious: it's a principle of CS that anything you can do on your own box, you can do remotely, thus if the local application isn't patented, then the web application can't be patented either, because it's obvious. I think if the USPTO realized this (although they're systemically disinclined to understand anything, since their revenue comes from *approving* patents), many assinine web patents would go away.
A few weeks ago we saw an article advocating "patents lite", in which the USPTO checks that the patent covers something patentable, but does *not* check for prior art, usefulness, or non-obviousness. The patent is much shorter, like 3 years. The first time it's challenged, the burden is now on the patent-holder to show that it's useful, novel, & non-obvious.
It turns out that de facto, we have patents lite. Clearly, the Patent Office checked nothing before it granted this patent, so we're just waiting for the first lawsuit. We have patents lite, & the system still doesn't work. It's time to end this.
Google had participated in the discussions and it was possible that its search technology would be used in the project.
Meaning, patent examiners will now Google the phrase "customer review" before saying, "Gosh, what an original idea!"
In other news, patent examiners' computers will now have web access...
Yeah, but Joel's an ass. Have any of his worshippers here on /. actually *used* something written by Fog Creek or whatever? FogBUGZ, a web-based bugtracker, seems to be his one claim to fame, & it's terribly mediocre. I mean, mostly it works, but the search function doesn't, the UI is inconsistent, & while you can define filters (such as, "my open priority-1 bugs"), you can't share them, which makes them nearly useless. Joel writes a good spiel, but when it comes to coding, his company ain't the shit.
Plus, he argues passionately for paying programmers well & giving them exciting projects, but in at least two cases he hired interns to start his company's most interesting apps.
Dude needs to work on his street cred.
From TFA:
"This is a golden era for geeks"
He wasn't kidding. It sure went downhill fast after 1999. His other opening lines, "we have permanently entered a new economy", and "Novell has again come to be seen as a worthy competitor to Microsoft", are not exactly prophetic (quick check on Yahoo stocks shows Novell's price has ended up pretty much where it was five years ago).
Other disagreements: "most of them would probably turn out to be terrible managers". I strongly disagree. Of the 5 managers I interact with weekly, the 3 who have running code in our systems (i.e., they're promoted developers) dress the worst and manage the best: they tell me my deadlines and my priorities, they ask me what support I need to write code, and they leave me the fuck alone. The 2 who don't have code running on our servers, who were first hired as managers, like to reorganize our hierarchy, introduce burdensome reporting requirements so the execs have more Social Science Numbers to look at, and want to transform our nice offices with *real* *doors* into a miserable cubefarm. I say, promote geeks! Even if they don't want it! I totally agree when he says "you can tell them what to do, but you can't tell them how to do it" (this is far from an original thought of his), but unless your managers are geeks, this approach will leave them feeling powerless and threatened. Managers meddle, it's what they're trained for.
If you want an insightful, thorough, and applicable discussion of all these ideas, as well as many more, some of them *original*, read the Scrum Handbook.
First of all, I resent the vaguely cultural-studies post-structuralist jargon of the article: "Here the heterarchy transcends the firewall and pressure can be applied from without." What's a heterarchy? Is that firewall a metaphorical one? I, for one, do NOT welcome our Foucault-reading post-modern academic overlords.
As for corporate blogging, the most useful blogs I've come across are from important developers in Microsoft (in particular) & also Google, Netscape, Python, etc. A number of times I've been investigating a fairly obscure question about some Microsoft API (shut up, it's my job), & found an excellent answer in a Microsoftie's blog. E.g., some feature seems blatantly missing, I'm searching for it, & the developer mentions in his blog that the feature IS indeed missing but he hopes to implement it in version 3.
This has nothing to do with marketing. I'm not sure what you'd call it in suitspeak, but it's sort of a conversational style of customer support & community-building.
!!
Bhutan is a devout BUDDHIST country, you hoser. And they have TVs now, although when I first read about that (more than 5 years ago), the Bhutanese were quite reasonably shy about appearing on it. Hence they had to cajole some poor sod to read the national news in a pained monologue. If you see some of the recent Bhutanese movies, though, it appears things are changing fast. I have no idea how they weigh their TVs and Buddhas, though.
On topic, I think it's great that they're using Avocado's Number to define the kilo. So let's see, 6.022 x 10^23 avocados would weight ~1.8 x 10^23 kilos, or 28 Earths (this measurement based on the standard platinum-iridium avocado that was shipped to me from Paris, which is significantly larger than a plum, but was free to ship, since any attempt to weigh it would cause fatal recursion).
> sound like a Wired columnist.
Heh.
Economics: FOSS.
Government: FOSS, Slashdot, Wikipedia. These aren't replacing regional governments, obviously, they're just small examples of groups negotiating resources in new, interesting ways.
Religion: Subgenius? Forget it.
My argument's based on Third-World Internet users growing faster than US users. So getting their attention should be a major priority for US sites at some point. Besides, if poor countries set up their own DNSes, they could just translate foo.com to foo.co.us, so now I really don't see what all the fuss is about.
Does anyone know what about ICANN's administration the plaintiffs don't like?
I'm reading The Checkbook and the Cruise Missle, in which Arundhati Roy says injustices increase as decision-makers are geographically separated from those affected by the decision. She cites the World Bank in Geneva, and the IMF, the WTO, as examples.
.com, US sites will have to use .co.us like everyone else? That seems reasonable. If they just set up root DNS servers that don't answer requests for .com (or .org, .gov, etc.), those servers will be more convenient to client hosts in their region. Software will get patched to check both authorities, since it's an easy fix, & US sites will register both types of domains to maximize their availability. Then, over an excruciating number of years, while everyone has to support both naming styles, .com & the other 3-letter domains will die out, & the plaintiffs will have their way.
So Third-World countries want power over names? And they think they can accomplish that by moving the naming committee to UN Headquarters in New York? The UN didn't work for poor people in Iraq, or Palestine. Why will it work in the case of Internet names?
This is the first case I know of where software standards have reached the level of world politics. (It's different from software patents in Europe.) I don't think they ever belong there. Software standards have developed reasonably well under Darwinian conditions: it may take decades, but eventually everyone switches to open standards because there's an advantage to being able to communicate. E.g., everyone uses TCP/IP now, not IPX or any other proprietary network protocol. I know, I know, we're still fighting this battle daily, but you can see the positive trend, & it's happening without any legislation or government enforcement.
What I'm getting at is Third World countries should just set up their own root DNS servers. Whatever it is they want -- get rid of the 3-letter root domains? So instead of
I'm gonna sound like a Wired columnist, but here goes: The Internet is suggesting new kinds of economics, government, maybe religion.... We should stick with what works, instead of imposing traditional kinds of governance onto the Internet.
You're kidding.... ....right?
And in article, they wrote it beamed the updated attendance record to a teacher's handheld. A slick piece of work, that, & hard to make robust (I have some experience in this area). My intuition has always been that evil people are bad programmers. Something about how their twisted brains can never exude straightforward code....
I bet they never got it working. The article implies they hardly used it before shutting it off.
And, he doesn't even understand himself. Here he is, trying to legislate what we call Linux ("That's GNU-slash-Linux"), as if he owned it. One of the things you have to give up, if you develop open-source (or free, or whatever) software, is the right to be credited as you'd wish. Someone may grab your code, rename & repackage it, strip many of your identifying marks from it, & sell it for a million dollars. That is precisely the freedom I thank Stallman for inspiring us to achieve, & it's exactly what Stallman, now that he's been eclipsed, wants to take away from us.
That said (or ranted), us Slashtrolls' reaction to this is too predictable. Why is the OSS community, on the whole, so antipathetic toward RMS? Is it because he's become such a dogmatic preacher? Is it that he's always been so, but now that we're nearly on top, we'd rather not be reminded of our moral obligations? Is it that we only respect the one with the latest Freshmeat entry?
They're looking at the web topologically, as usual, but rather than measuring distance from site A to site B by the minimal number of router hops required, they're measuring the number of clickable links from A to B. In other words, if you started at A.com and had no keyboard, could you click your way to B.com?
The results were that topological diameter was 19 links. Anyone know the diameter of the internet (average traceroute hops from any site to any other)? Furthermore, the overall connectedness is low, so if you took out (e.g.) Yahoo and MSN, you might not be able to click from someone's panda hentai page at Geocities to my Jar-Jar hate site. I can't seem to find in the articles whether this only deals with static linking, or if search engines are accounted for somehow.
This is sort of an odd way to look at the web. Most people don't start from their home page and start clicking until they find something interesting. You start at some place you type in, do a search, make a huge leap into a topologically distant area, then start moving around connected nodes, then make another huge leap. If MSN died and their routers stayed up, would the web be geometrically less useful, as they claim, or just linearly less useful?
I'd say that this focus on the topology of links is really vieaux chapeau now that most people use interactive services to grab information. The web isn't a static, well, "web", anymore.
Oh, boy, here it comes...
You asked to make money, do something interesting, and contribute to the world. All of the suggestions above are great for making a contribution to society, but most of them (notably being a high school sysadmin, or going to Africa), won't make you much money and probably won't be technically challenging. Apply to RedHat, IBM, Corel, SGI, SuSE, 3Dfx, etc.--most companies with lucrative and interesting jobs are still at least 35% evil, but if you can get into the cool parts you can contribute to making Free Software the dominant business model, and get a good salary hacking interesting problems.
The Byzantine Generals problem deals with exactly what McNett needs:
The Byzantine generals problem is formulated similarly. One formulation (the closest to this) is: N generals are on a hilltop, about to attack a city. K are traitors, who will interfere with any protocol in the most damaging way possible. They must agree on some piece of data (the time to attack the city) reliably. Here is a link with some explanations and implementations of the solution.A commercial "Distributed.com" would have a simpler problem, because they can reliably a) authenticate a computer's identity, so they know if two messages come from the same computer, and b) they can assume that the server isn't a traitor. This will severely reduce the level of redundancy necessary. Still, they must deal with truly malicious nodes, whereas Distributed.net has only had to deal with faulty ones.
As for granulating the data so that K traitorous nodes cannot glean something useful from the data, this should be interesting information theory. I would think that adding some garbage data to calculate from, along with the real stuff, might be a decent cost/security trade-off.
Researchers have often "determined" whether a drug is chemically addictive by testing whether e.g. a rat will self-administer the drug. Hook the beast up to an IV, demonstrate to it that pushing a lever will give it a drop of smack, and watch. Rats will self-administer opiates and quickly become massively addicted. Caffeine, nicotine, cocaine; the usual suspects. Rats don't seem to like alcohol, THC (pot), or LSD.
So, in light of recent worries about web addiction, will a parrot self-administer the web?
Woah, check out my keywords...Echelon's gonna love this post.
This doesn't count the TCO of the pieces of crap you get. My mother got an eMachine a year ago. It had a fan that sounded like a Harrier, so we sent it back. Every time we sent it back (paying $25 one-way shipping) we got a new machine, meaning we had to set up all the software again, and a new hardware problem always surfaced in the space of two weeks. Going on newsgroups, I discovered that most of these problems are endemic throughout their product line (particularly the unreasonably noisy fan). The FTC should require total cost of ownership disclosures--that machine must have cost us $500 in shipping and time alone.
Of course, the space sation is already five years overdue and over budget by a factor of something like ten, without being more than a quarter completed. But the situation with that was politically and technically quite different.
Sure, the superpowers may say 'been there, done that', but one advance this mission might show us (as well as crazy rocket guy's mission) is how to do space travel cheaply and on short notice. If NASA could do missions this cheaply, they could just send up three at a time to boost their success rate. =)
Reliable, cheap, turn-key space travel is what will bring the future here. Like computers: moving from ENIAC to the Vaio laptop.
The Patent & Trademark Office could take a few pointers from the DOJ.
int[,,] a3; // 3-dimensional array of int
is valid syntax, which should remove a few headaches. Nevertheless, I was overall expecting something a little more...more.
A lot of things that are good about the Java platform helped it take off, and lot of things that are bad about the platform keep it circling at low altitude. What really allowed Java to take off as a language, in addition to the platform, was the library and how it is keeping up with recent ideas about 'design patterns'. Java's symbiosis with design patterns, and the fact that Java GUI programming is some of the easiest there is, have been a great asset. What really remains to be seen is how C#'s libraries are accepted by the developer community, because these days it's often the standard libraries that really matter.
It's a shame that C# will be so embroiled in politics from the get-go that we'll never get to see how good (or bad) it actually is. Microsoft has some smart people, and I think the world has a niche for a simplified, compiled, reasonably-performing, garbage-collecting Java/C++. But C# is the child of a tyrant--it can never succeed or fail on its own merits.
Under the leadership of Vice President Gore...
...every on-line resource offered by the federal government...
...one-stop shopping for government services...
I can see already what this is about.
This new website, firstgov.gov, will be created at no cost to the government [much deserved snip] in 90 days or less. It will uphold the highest standards for protecting the privacy of its users.
Bill's quite quick on the uptake, you know? He has already mastered the art of vaporware. I envision a brave new world of E-Government(TM), where the irrelevant announcements of new Microsoft products and the pointless promises of politicians will go hand in hand!
Ah hah--now I see why it will be free, and take only 90 days to build. The http://firstgov.gov/ will simply be a big picture of the IRS's middle finger. But it will take three months for the Feds' first finger to figure out what their middle finger is doing.
Excuse me, Webmaster, what's the going rate on senators, and is it possible to buy one online?
How in the world did IBM, famous for its entrenched monopolist corporate culture, manage to turn itself around so quickly and fundamentally?
I believe the problem with long-range asteroid prediction is not computational power. The N-body problem with only a few dozen good-sized gravity wells in the system is easily simulated with a desktop machine. The problem is the accuracy of our measurements. The magnitude of the error term dominates the significant result pretty quickly as you extrapolate measurements centuries into the future.
What are we to make of the lack of public response to the problem of protecting against meteor collisions? With--what?--3 movies two summers ago about major collisions (okay, one was about a comet, which I don't think anyone would survive), still no one has much of a reaction. I don't know whether to revive my faith in humanity, since those execrable had so little effect on the national consciousness, or to leave it dead, because we never wake up and try to save ourselves until it's too late.
An AC just pointed out something which makes this more interesting. Since the AC will get moderated into oblivion (for good reason, given the needless fucking profanity), I'll repeat his/her point: RedHat and Cygnus are now the same company. Wasabi is distinguishing between oranges and oranges.
It's interesting that he made a strong destinction (perhaps with a note of derogation?). I suppose the difference is that Cygnus builds tools (many of them proprietary, incidentally) for developing on Unix, for porting between Win32 and Unix, and they customize open source tools and OSen. On the other hand RedHat doesn't customize Linux for specific clients, rather it focusses on the consumer model of one-size-fits-all (or 'a-few-sizes-fit-all').
Note that Cygnus makes most of its revenue from distributing proprietary tools.
N: If clients approach you for NetBSD development are you making sure that it's going to be released under the BSD license?
P: Everything that we can we will. There will be instances where clients come to us for work that will be used in house, or is uninteresting. But we're unequivocably an open-source company, and we want to release virtually everything we do as open source.
They're not going to GPL everything. Cygnus' revenue model relies on proprietary products. There's a lack of information on how Wasabi will actually make money (if neither support nor beach balls seems like a good revenue source). This makes me wonder how well they'll do, and how much work they can give back to the community. Frankly, if a company is committed to GPLing every line of code they produce, RedHat seems like a better business model to emulate.