" the right to use pictures and/or voice recordings of employees in advertisements."
That's what I figured they wanted it for. Or so they can shoot video of whatever and use it for promotional material without asking permission of everyone who gets into a clip. Unfortunately they didn't specify a use other than _whatever they want_. Although I don't think it was exclusive. Lawyers always seem to try to include everything just to cover what they want. In this case, they lost someone - and HR was throwing more money at me than I expected, so they must have thought they wanted me.
I make it a point to discuss IP agreements up front. I once inteviewed at EDS, and the HR person wanted me to look specifically at the IP one and some other document. There were many, but they singled out these two. Upon reading the "all your inventions (past,present,shower) are belong to us" and the "right to use my voice and likeness for any purpose" among other things, I asked if they could be changed. No. I left before the technical part of the interview. I had no job at the time, but the IP stuff was too restrictive for someone who writes code on the side.
I now work for a company where the IP agreement explicitly _excludes_ anything not directly related to my work for the company. Reading this one was a breath of fresh air. Still working there 4.5 years later.
"The FSF requires detailed changelogs for their own projects."
They also require copyright attribution for any significant changes. That's nice because they can then just say it's all their work. They could conceivably relicense the code any way they see fit - GPL 3.0 for example.
FC3 allowed me to rip CDs to FLAC, only to discover there was no player included that could handle the format. No patents either, just a glaring inconsistency.
BTW, I assume the patent issue is also why there isn't a mpeg video player, but I do understand that.
So KDE shouldn't do an official release until the translations are done. Obviously you consider translations a critical feature, and I'd agree with you. There should be some translations of 3.4.0 available from day one (IMHO). Perhaps with more to follow, but certainly with the same version number for a major release. Don't blame a distro for putting in the latest "official" release.
"they are required to keep the original copyright attributions in the code.
I've often wondered about that. If I keep the original copyright notice in there, and make modifications (perhaps significant) and release my changes under the GPL, how do I label it? I have to keep a copyright notice that attributes the code to someone else? What about my contributions? What if it's a complex mix where I can't nicely label just my parts? The GPL tells you what you're allowed to do and what your obligations are, it doesn't tell you how to do this in practice. Does the FSF have documentation on this? I haven't seen it.
You're quite right about the differences between rasterizing and ray tracing. In my RT library, polygon (and other object) intersection tests are not the limiting factor, not even close. Using a proper scene subdivision structure you generally only do a couple intersections per pixel, and then only one shading operation. This means a 1 megapixel display running at 60 FPS need only do 60M shadings per second (barring reflections etc). The bottleneck in my code is a recursive tree traversal. Unfortunately once you optimize that, you still have a very strange mix of tree traversal, intersection test, shading, and recursive rays. It's not a problem that makes for easy hardware pipelining.
The library is actually much faster now than that old release of the game. If moores law continues by giving us multicore chips, you'll have realtime raytrace FPS, and our chess game will be slugish with photon mapping both in about 7-8 years.
When I was a kid and saw my dad programming, he said "I think I can use this to make games." Boring strange looking text on the screen was suddenly a way to do something I was interested in. Software isn't an end unto itself (for most people) but a means to something they're interested in. Games are universally interesting to kids, though you may find another interesting topic and regardless, you should tell them how software can also be applied to all sorts of other things they may be interested in.
I assumed better would mean it takes longer - and hence more money. Your "bus" example shows that this is not always true, which leads make me suggest "more effective" examining (I'm trying to get at - review that doesn't let stupid stuff through). Either way, a patent on a bus should be rejected which will result in less money to the PTO.
I find it disturbing that on appeal the patent gets reviewed by people who are less knowlegable in the field. That's like at a big company where managers make technical decisions on stuff they aren't qualified to make. Bummer.
First, there is no law prohibiting people from disclosing company information (except securities laws concerning releases during certain times) so they have not committed a crime. What they probably did is violate terms of some non-disclosure agreement. That's more like breaking a contract, but it's not a "crime". You can't really say someone violated their NDA until you know who leaked the information and produce the signed NDA. This judge is probably trying to reinforce the smackdown on the press after the CIA identity leak - which probably was a crime. This is not the same thing.
The US PTO is has been a profit center for the government for many years now. That's part of the problem, no one wants to cut a revenue source. Doing better examination would reduce profits, and eliminating any class of patent would do the same.
"Don't let them make you back down. The possibility of needing to protect yourself from or overthrow the current or future government is another completely reasonable reason to own a firearm."
Yes, that's another reason people should be allowed to have guns and is often cited as something on the minds of those who wrote the constitution. However, it is an unusual reason and many people don't buy the arguement and consider it radical or nutty - once you mention it, they'll write you off as a kook. If you can win a pro-gun argument without bringing this up I think you'll be better off.
Are you saying you need to be a PE to label yourself an engineer? Is your dads company doing civil engineering? I think you're right in that case. There was talk about requiring it for software and electrical engineers, but I thought that all died down. So my question is: who does this apply to?
I suspect they never tested at full capacity because it might not be real safe to land with that much fuel. At 85 percent fuel by weight it'd be dangerous to land, and the handling probably isn't really great fully loaded. The only way to test with full load then would be to fly for 3 days. What's wrong with dumping the excess fuel before landing? I dunno. The web site also has a story about some problems weighing it before takeoff - not that they connect the two.
Has he left Japan yet?
I was pleased to read that Williams International (who made the engine) is right here in Michigan, not too far for me to work - unfortunately they had no postings for software people:-(
IMHO, X-Files went to crap because they started focusing on all the things that made the show great. In the old days, most episodes were unrelated cases with this occasional alien/government/conspiracy in the background and a bit of sexual tension between the main characters. These elements in moderation made it more interesting. Toward the end the conspiracy became the ONLY thing they focused on and since resolving the mystery would end the show, they were stuck with the awkward situation of chasing the conspiracy ad-nausium. Other shows have failed for similar reason: Ally McBeal IMHO was entertaining with these sometime quirky characters. When they decided that's what made the show good, they concentrated on those aspects until the characters were all a bunch of freaks. Star Trek gets on shakey ground whenever they focus too much on nifty "tech" or ongoing story lines. And of course Happy Days failed the same way by focusing on Fonzies coolness by having him "jump the shark". Anyway, it's not how long a show runs, IMHO it's when they start focusing on the good elements too much that it goes to crap. I suppose that happens when the writers can't think of any new story ideas. Hmmm.
"Why then shouldn't they go ahead and pursue a patent attack strategy in order to crush what they see as the competition?"
Perhaps they should. However, a common theme on slashdot is how broken the patent system is. Trivial things are patented every day that demonstrate the system not working the way it was intended. Sometimes companies fight over silly patents and it can be fun to watch one big company screw with another one (which might have done the same if it could) and comment on the system. Free software (and OSS too) are usually not corporate developments, a full Linux distribution is the product of thousands of people working for over a decade to develop. FLOSS represents a lot of different things to a lot of people, so to see it crushed by MS utilizing the broken patent system would be a travesty on a global scale.
My guess is that's one reason MS hasn't tried to actually play the patent card against linux. It wouldn't be money attacking money, it would be very big money attacking the people (really bad PR). This would be unprecedented, so there is great uncertainty with it. Also, with governments and business around the world considering OSS, this kind of attack would make legislators question the very patent system such an attack would rely upon. Imagine using patents to go after Linux, with the result that the rules change - or get repealed - so that you can't use them against other companies either. No one knows what would happen if MS attacked Linux with patents, but all effects other than defeating Linux would likely be negative.
Either that, or they are waiting for legalization of software patents in Europe...
"You have to be able to hire smart people, make sure they know what they're doing"
The problem with that is evaluating the people you hire. How can you say a guy knows what he's doing if YOU don't know what he's doing? Not to say it's impossible, but it can be difficult. From what I've read, the most successful companies in the fortune 500 have top people who were promoted from within. They know how the company operates and what it's capable of. The CEO of XOM for example started there as a chemist - there's a lot more to running the company than that, but he knows what they do and understands how it's done and what's possible.
You must do some profiling. Most people on slashdot have never done what I consider embedded code (PC 104 boards don't count). I write for micro controllers and DSPs, and the compilers vary a LOT. Even compilers from the same vendor with different target achitectures can have different optimization capability. When I do write a really strange looking expression, I often leave the equivalent readable code commented out or give a good description of what I did for whoever works on it next.
From the sound of your post, I'm assuming you already understand the bit about premature optimization and are now at the point where you must do it anyway. Slashdot is probably not the best forum for this, as they all have great hardware and compilers and wouldn't understand how someone could actually care about saving 1/2 microsecond of execution time on a 60MHz machine.
My last round of optimization was in a complex control loop running in a 20KHz interrupt on a 60 MHz part. This code was eating about half the available processor time, and reducing it by a few microseconds freed enough time to add 15 percent to the background code - which is quite substantial. It was a series of 13 different improvements slightly more complex than the example you gave. I don't recommend doing this stuff all over the place, but when you need to the only way to be sure is to do real timing tests - sometimes things you think will help don't and things you think are the same are better... And it mostly depends on the compiler you've got.
" it would in fact be unusual to present a false color image, since radio data are more commonly illustrated using contour maps."
Great, so where's the map? Perhaps the news article just chose not to show it. That's fine, but it really looks stupid to show a picture of empty space and claim to know there's something there without showing the evidence. Dark matter is supposed to be undetectable (directly), so to claim it's there requires some indirect proof. They claim it's shaped like a galaxy, so there should be some nice visual that indicates that.
"By totally misunderstanding every aspect of the story, you have effctively stood up in front of the/. community, and loudly proclaimed 'I'm dumb as shit'. Congratulations."
I understood the article just fine thanks. It didn't present the detail you claim these guys have. Without that, I still say it's just as likely to be fiction. Sure you can interpret my post as me being "dumb as shit". Sometimes rather than asking "where's the picture that indicates something is actually there?" I like to make it more interesting. This is at the risk of being modded down and called "dumb as shit" as happened in this case. I really don't have a problem with that - my life does not depend on the impressions of/. readers.
Just don't go to those sites that annoy you. For example, slashdot doesn't have these things. Nor does google news, the times, any of the other sites I visit regularly. The fact that so many people are annoyed by these things seems to point to the fact that they can't stop searching for [celebrity] porn. No one ever points out that the problems are with the dodgy sites.
Another possibility is that people are not good at finding the more legitimate stuff they want and end up clicking links to dodgy sites. That's just user error. You'd think with errors being so annoying people would learn.
How do you know your buddy surfs porn? Ask him if popups or floaters are a big problem for him.
I always laugh at this word. We used to have an intern that said one of here sorority sisters was nicknamed "Hoover" - and she said "It's for the reason you think". Hah! One day my buddy looked over and just said "ya, I wanna meet that girl 'Hoover'". ROFL
It's funny. Geeks will donate $300,000 to advertise thier web browser... But they can't organize a lobbying group to represent them in washington. Sure, the EFF tries to do good things, but mostly after there is a problem. The FSF only tries to stay afloat and stop GPL violators. Where is the free software political action in America? The money is there.
They claim to have detected a large amount of Hydrogen using radio a radio telescope. So where is the false color image of this galaxy? How do they know it's rotating like a galaxy? The only way you can claim the presense dark matter is by its effects on real matter. They haven't shown any sort of evidence of the real matter they claim to have detected. To post a picture of empty space and say it's full of dark matter is just stupid. I think the only dark matter this article shows is in the astronomers head.
"It should really be noted that some of the low-level tools from sysinternals use very similar techniques to what a rootkit would do, just that they do it for monitoring and not with falsification of data as intent."
I can see it now. The future Microsoft product (which might come free with the OS) will say this other tool is a rootkit and remove it. This area of security should be very interesting to watch.
I have shuttle SN95G5 - nforce3 socket 939. The LAN controller on my box has some bad VPD data inside, and Linux performance was suffering in a bad way (the system would freeze every second or so) Several people have reported similar problems on the net, but Shuttle has refused to fix the problem because it's not a problem for Windows (you can find error informatio in Win, but I guess it's not a performance problem). ASus has a board with a similar problem, and their patch has done wonders for my Shuttle. I just wish Shuttle supported their products (and Linux) a little better. That aside, I really really like my SN95G5 now that Asus has provided a fix for it;-)
Slashdot Mirror
That's what I figured they wanted it for. Or so they can shoot video of whatever and use it for promotional material without asking permission of everyone who gets into a clip. Unfortunately they didn't specify a use other than _whatever they want_. Although I don't think it was exclusive. Lawyers always seem to try to include everything just to cover what they want. In this case, they lost someone - and HR was throwing more money at me than I expected, so they must have thought they wanted me.
I now work for a company where the IP agreement explicitly _excludes_ anything not directly related to my work for the company. Reading this one was a breath of fresh air. Still working there 4.5 years later.
They also require copyright attribution for any significant changes. That's nice because they can then just say it's all their work. They could conceivably relicense the code any way they see fit - GPL 3.0 for example.
BTW, I assume the patent issue is also why there isn't a mpeg video player, but I do understand that.
So KDE shouldn't do an official release until the translations are done. Obviously you consider translations a critical feature, and I'd agree with you. There should be some translations of 3.4.0 available from day one (IMHO). Perhaps with more to follow, but certainly with the same version number for a major release. Don't blame a distro for putting in the latest "official" release.
I've often wondered about that. If I keep the original copyright notice in there, and make modifications (perhaps significant) and release my changes under the GPL, how do I label it? I have to keep a copyright notice that attributes the code to someone else? What about my contributions? What if it's a complex mix where I can't nicely label just my parts? The GPL tells you what you're allowed to do and what your obligations are, it doesn't tell you how to do this in practice. Does the FSF have documentation on this? I haven't seen it.
Shameless plug: rtChess
The library is actually much faster now than that old release of the game. If moores law continues by giving us multicore chips, you'll have realtime raytrace FPS, and our chess game will be slugish with photon mapping both in about 7-8 years.
When I was a kid and saw my dad programming, he said "I think I can use this to make games." Boring strange looking text on the screen was suddenly a way to do something I was interested in. Software isn't an end unto itself (for most people) but a means to something they're interested in. Games are universally interesting to kids, though you may find another interesting topic and regardless, you should tell them how software can also be applied to all sorts of other things they may be interested in.
I find it disturbing that on appeal the patent gets reviewed by people who are less knowlegable in the field. That's like at a big company where managers make technical decisions on stuff they aren't qualified to make. Bummer.
That's not to say I condone violating your NDA.
The US PTO is has been a profit center for the government for many years now. That's part of the problem, no one wants to cut a revenue source. Doing better examination would reduce profits, and eliminating any class of patent would do the same.
Yes, that's another reason people should be allowed to have guns and is often cited as something on the minds of those who wrote the constitution. However, it is an unusual reason and many people don't buy the arguement and consider it radical or nutty - once you mention it, they'll write you off as a kook. If you can win a pro-gun argument without bringing this up I think you'll be better off.
Are you saying you need to be a PE to label yourself an engineer? Is your dads company doing civil engineering? I think you're right in that case. There was talk about requiring it for software and electrical engineers, but I thought that all died down. So my question is: who does this apply to?
Has he left Japan yet?
I was pleased to read that Williams International (who made the engine) is right here in Michigan, not too far for me to work - unfortunately they had no postings for software people :-(
IMHO, X-Files went to crap because they started focusing on all the things that made the show great. In the old days, most episodes were unrelated cases with this occasional alien/government/conspiracy in the background and a bit of sexual tension between the main characters. These elements in moderation made it more interesting. Toward the end the conspiracy became the ONLY thing they focused on and since resolving the mystery would end the show, they were stuck with the awkward situation of chasing the conspiracy ad-nausium. Other shows have failed for similar reason: Ally McBeal IMHO was entertaining with these sometime quirky characters. When they decided that's what made the show good, they concentrated on those aspects until the characters were all a bunch of freaks. Star Trek gets on shakey ground whenever they focus too much on nifty "tech" or ongoing story lines. And of course Happy Days failed the same way by focusing on Fonzies coolness by having him "jump the shark". Anyway, it's not how long a show runs, IMHO it's when they start focusing on the good elements too much that it goes to crap. I suppose that happens when the writers can't think of any new story ideas. Hmmm.
Perhaps they should. However, a common theme on slashdot is how broken the patent system is. Trivial things are patented every day that demonstrate the system not working the way it was intended. Sometimes companies fight over silly patents and it can be fun to watch one big company screw with another one (which might have done the same if it could) and comment on the system. Free software (and OSS too) are usually not corporate developments, a full Linux distribution is the product of thousands of people working for over a decade to develop. FLOSS represents a lot of different things to a lot of people, so to see it crushed by MS utilizing the broken patent system would be a travesty on a global scale.
My guess is that's one reason MS hasn't tried to actually play the patent card against linux. It wouldn't be money attacking money, it would be very big money attacking the people (really bad PR). This would be unprecedented, so there is great uncertainty with it. Also, with governments and business around the world considering OSS, this kind of attack would make legislators question the very patent system such an attack would rely upon. Imagine using patents to go after Linux, with the result that the rules change - or get repealed - so that you can't use them against other companies either. No one knows what would happen if MS attacked Linux with patents, but all effects other than defeating Linux would likely be negative.
Either that, or they are waiting for legalization of software patents in Europe...
The problem with that is evaluating the people you hire. How can you say a guy knows what he's doing if YOU don't know what he's doing? Not to say it's impossible, but it can be difficult. From what I've read, the most successful companies in the fortune 500 have top people who were promoted from within. They know how the company operates and what it's capable of. The CEO of XOM for example started there as a chemist - there's a lot more to running the company than that, but he knows what they do and understands how it's done and what's possible.
From the sound of your post, I'm assuming you already understand the bit about premature optimization and are now at the point where you must do it anyway. Slashdot is probably not the best forum for this, as they all have great hardware and compilers and wouldn't understand how someone could actually care about saving 1/2 microsecond of execution time on a 60MHz machine.
My last round of optimization was in a complex control loop running in a 20KHz interrupt on a 60 MHz part. This code was eating about half the available processor time, and reducing it by a few microseconds freed enough time to add 15 percent to the background code - which is quite substantial. It was a series of 13 different improvements slightly more complex than the example you gave. I don't recommend doing this stuff all over the place, but when you need to the only way to be sure is to do real timing tests - sometimes things you think will help don't and things you think are the same are better... And it mostly depends on the compiler you've got.
Great, so where's the map? Perhaps the news article just chose not to show it. That's fine, but it really looks stupid to show a picture of empty space and claim to know there's something there without showing the evidence. Dark matter is supposed to be undetectable (directly), so to claim it's there requires some indirect proof. They claim it's shaped like a galaxy, so there should be some nice visual that indicates that.
"By totally misunderstanding every aspect of the story, you have effctively stood up in front of the /. community, and loudly proclaimed 'I'm dumb as shit'. Congratulations."
I understood the article just fine thanks. It didn't present the detail you claim these guys have. Without that, I still say it's just as likely to be fiction. Sure you can interpret my post as me being "dumb as shit". Sometimes rather than asking "where's the picture that indicates something is actually there?" I like to make it more interesting. This is at the risk of being modded down and called "dumb as shit" as happened in this case. I really don't have a problem with that - my life does not depend on the impressions of /. readers.
Another possibility is that people are not good at finding the more legitimate stuff they want and end up clicking links to dodgy sites. That's just user error. You'd think with errors being so annoying people would learn.
How do you know your buddy surfs porn? Ask him if popups or floaters are a big problem for him.
I always laugh at this word. We used to have an intern that said one of here sorority sisters was nicknamed "Hoover" - and she said "It's for the reason you think". Hah! One day my buddy looked over and just said "ya, I wanna meet that girl 'Hoover'". ROFL
It's funny. Geeks will donate $300,000 to advertise thier web browser... But they can't organize a lobbying group to represent them in washington. Sure, the EFF tries to do good things, but mostly after there is a problem. The FSF only tries to stay afloat and stop GPL violators. Where is the free software political action in America? The money is there.
They claim to have detected a large amount of Hydrogen using radio a radio telescope. So where is the false color image of this galaxy? How do they know it's rotating like a galaxy? The only way you can claim the presense dark matter is by its effects on real matter. They haven't shown any sort of evidence of the real matter they claim to have detected. To post a picture of empty space and say it's full of dark matter is just stupid. I think the only dark matter this article shows is in the astronomers head.
I can see it now. The future Microsoft product (which might come free with the OS) will say this other tool is a rootkit and remove it. This area of security should be very interesting to watch.
I have shuttle SN95G5 - nforce3 socket 939. The LAN controller on my box has some bad VPD data inside, and Linux performance was suffering in a bad way (the system would freeze every second or so) Several people have reported similar problems on the net, but Shuttle has refused to fix the problem because it's not a problem for Windows (you can find error informatio in Win, but I guess it's not a performance problem). ASus has a board with a similar problem, and their patch has done wonders for my Shuttle. I just wish Shuttle supported their products (and Linux) a little better. That aside, I really really like my SN95G5 now that Asus has provided a fix for it ;-)