Well, my point was not to differentiate the big players (Microsoft, Cisco, et al) from the smaller firms, but rather the commercial firms from the open source groups.
To splinter the discussion, I do agree, innovation does come from the small firms. They know they need to come up with something unique because they can't compete with a saturated market with limited resources compared to the other companies in the market.
Those are nice but are really minuscule in comparison to the innovation I'm talking about. I'm talking about GUI's and The Internet, that revolutionized the world and those examples haven't.
I never said Microsoft did real innovation, but since you mentioned it, I'm waiting for what Longhorn and the next iteration of Windows. I saw a video of a Microsoft programmer demonstrating Avalon/WinFX and it was pretty nice. I read somewhere that Microsoft plans to incorporate Avalon in everything so it'll make everything Microsoft/Apple/Linux has look 2D and obsolete. I don't see either Apple or Linux folks with something like that which does make it innovative.
Sure Apple has an eye-candy interface, but I've used OSX and OSX doesn't have anything near Avalon. Linux has it's various window managers, which you can take how you want.
I'm not a Microsoft fanboy, I'm waiting to try it first hand.
I agree with you that there have been incremental improvements recently, better filesystem here, better version of a popular software product there, but nothing as groundbreaking as a graphical user interface or the internet itself has come along in a long time. Earlier today, I watched Pirates of Silicon Valley and I was left to wonder what it would be like to be on the cliff of innovation like Gates and Jobs were. My next thought was "when is the next instance of revolutionary technology going to hit?"
My point is that it's the commercial software firms are making the innovations, albeit small, because they are being paid to. OSS alternatives are being written for lots of reasons, mainly because they believe in OSS and the idea of free software. They aren't in any hurry because they know they won't be the first to market so they run at their own pace.
This is the fundamental difference between pioneering commercial software firms and groups of talented programmers write software so they can give it away and say "see, you don't have to pay for it".
All those are simply nice features within an application. For example, some Mozilla guy looked at Mozilla and then to RSS feeds and had a "eureka" moment. That's hardly innovation, it's just combining an existing program that intersects with something else enough so it makes sense to combine the two. My argument was aimed at the notion of an entire application.
The problem is that OSS will always be catching up. I have yet to see popular OSS software show real innovation. They are just alternatives to existing applications. While that itself isn't bad, the innovators that pioneer original applications will continue working and breaking new ground and making money until free competitors catch up to them.
I never said novels suck, they are useful to some people. For you, dimension A is filled by a novel, but for me, dimension A is filled with technical knowledge. I have an imagination but not in the way most people know. I mentioned LotR because traditionally that book series and other fantasy novels were touted as the only way one can use their imagination, but it's just not true.
I disagree because novels don't provide mind expansion. If I pick up a book, I don't have this overcoming urge to fill my head with it's knowledge.
I have tried to read in the past and was unsuccessful. In high school, teachers saw potential in me and put me in advanced math and english classes (which called for summer reading). I did fine in advanced math but it took half the summer to read one book and there were 3 to read.
I don't see that as any type of shortcoming because I know I have imagination. I can visualize technical things like shifting bits in a compression algorithm. I can also explain characteristics of the algorithm based on input it's been given and the output it produces. I just can't visualize a forest while I might read Lord of the Rings.
Maybe books in their current hardcover forms are becoming obsolete, maybe only bibliophiles are the only ones that want to keep books around. I am tired at how someone like yourself condescends down to anyone that doesn't absolutely treasure the same classic literature you love citing some imaginary lack of intelligence. Do I condescend down upon you if you don't like my kind of video games?
It'll work but it's not a finished product without the CS guys. Likewise, the CS guys wouldn't have a job if it weren't for us EE/ECE guys to provide new technology for the market.
It depends on what you're making too. I interviewed for a popular internship with a company called Leitch. Their local office created a PowerPC-based multichannel video communication system for use in hospitals and anyone that wants to make a television signal with a custom channel lineup. The product used basically only EE/ECE people because there would be an extremely minimal interface for the end users.
heh, you shouldn't, I just envision technology to be an assembly line.
1. Someone has an idea. 2. The EE's and I fabricate it. 3. You CS guys make it perform it's purpose. 4. IT guys make sure it keeps performing it's purpose.
I'm in Computer Engineering myself (graduating in May myself). I don't think I'll be frying bacon or anything. I've looked at jobs and companies like Intel and Siemens are looking for fresh college grads, as well as smaller firms.
I've been looking for jobs in the Boston area and they have the same aforementioned jobs with Intel, Siemens and such, however they also have smaller firms that were put together by MIT graduates that are also looking for EE/ECE graduates fresh out of college.
So in conclusion, I'm not worried. I'm aware I won't have a glorious 6-figure salary in July, I know that, I know I need to pay my dues as far as job quality, but it's not like Computer Engineering will be obsolete in 5 months. I'm so far upstream past the IT/CS people that I can land practically anywhere in computing I want, but I'd rather be in computer engineering because the work is more engaging than anything else.
Don't worry about your degree, brother/sister, you won't "make it big" instantly after you graduate but you'll be just fine.
You are correct, but I am speaking from the computer engineering curriculum and XOR's are very important to us because we use them alot and the increase in hardware cost and footprint for having to substitute I mentioned in the grandparent instead of a single XOR gate is unacceptable. However, yes, a NAND in a more general logic design is pretty useful.
I was given a Playstation 2 by my sister because of DRE's. She used it to read DVD's most of the time and that's the prime reason why DRE's occur.
I opened it and tried working with the adjustment wheel but it didn't fix the DRE. I read another fix is that the laser isn't receiving enough power and didn't want to handle all that. I saw the lense was dirty, cleaned it and it works fine. Since then my sister bought another Playstation 2 so I got a free Playstation 2.
Well, that's if it's under warranty which the first generation models are well past the warranty period. The cost to fix the Playstation by Sony is close to the cost of a new Playstation. I assure you that you are the exception and not the rule.
First generations aren't always the best, look at the Sony Playstation 2.
Do a google search for the terms "disc read error" and find results about Sony Playstation 2's having Disc Read Errors. This is because the first models used plastic pieces handling the laser but the later models used metal pieces that don't fair after a while. It's not quite "water under the bridge" when so many results come up with a relatively unspecific search terms all point to one problem.
A guy I talked to arranged basically an digital EQ on his roof.
He took green, yellow and red lights and arranged them in 16 rows of 16 colums where the bottom 8 light blocks of each column were green, the last 8 were divided between 4 yellow and 4 red. The columns divided among the frequency range he was going for. He built a piece of hardware to control on and off and mated it with some software to take Winamp music and read the frequency and power data and adjusted the lights accordingly to make his roof lights change to show an EQ. He also put some speakers outdoors to play the music the EQ was showing to people that stopped by. I had some pictures but they are unfortunately gone, that was pretty hot too.
That was true too, it wasn't just one issue that did DEC in. DEC had a great thing but marketing flops and the partnership you described was their downfall.
Alpha was plenty better than the Motorola 68x architectures out at the time (and better than Intel's but that should be given) but DEC just didn't know how to do anything else except design the hardware. When they finished and showed off the Alpha at conventions, other companies had equally complex hardware and software to demonstrate with. DEC had a much better hardware design but maybe 2 half-ass programs to show the thing turns on and does a couple things. Then Compaq bought them but didn't take the line any further, just wanted to bleed out what was left of the Alpha.
I think you misunderstood me, the carry flag in the PSW or CCR (or whatever you call it) would be set if it should be. If we are on the assembly level, you'd have to employ add/sub with carry/borrow instead of vanilla add/sub. The compiler would have to reasonably account for this.
I don't know about your first 2 reasons, never dealt with that. However, on point 3, you can't dismiss overflows just because you might not think they will happen often. You have to accomodate for all special cases otherwise the design is only partially correct.
I know homemade arch's should do it in a single clock, but I'm slicing the issue into smaller amounts of time. It's a better practice to make it as fast as possible as soon as possible because if you start pipelining your architecture and find out your ALU add is the slowest (unlikely since memory is the big bottleneck), then you have to rework it anyway.
It'll work for any bus length as long as the XOR operation itself is long enough to support the bus. If A or B is greater than 2^(32) - 1, that's the fault of the previous instruction to cause an overflow. The XOR trick certainely can't be held responsible for problems with code before it.
That's a feature of C then. That's the thing, I'm being taught to become a processor designer. We are taught to not presume anything. The question originally posed to me from my teacher was that we had 2 registers and an ALU that were connected in a loop. ALU's that we deal with don't inherently have that capability. Even if the ALU could detect a carry, the added conditional statements would balloon the instructions enough to make adding/subtracting unreasonable.
We look at ways of getting things done as quick as possible on as many architectures as possible, from the very big (x86, 68k, etc...) to very small (homemade architectures, 6811, etc...).
Slashdot Mirror
Well, my point was not to differentiate the big players (Microsoft, Cisco, et al) from the smaller firms, but rather the commercial firms from the open source groups.
To splinter the discussion, I do agree, innovation does come from the small firms. They know they need to come up with something unique because they can't compete with a saturated market with limited resources compared to the other companies in the market.
Those are nice but are really minuscule in comparison to the innovation I'm talking about. I'm talking about GUI's and The Internet, that revolutionized the world and those examples haven't.
I never said Microsoft did real innovation, but since you mentioned it, I'm waiting for what Longhorn and the next iteration of Windows. I saw a video of a Microsoft programmer demonstrating Avalon/WinFX and it was pretty nice. I read somewhere that Microsoft plans to incorporate Avalon in everything so it'll make everything Microsoft/Apple/Linux has look 2D and obsolete. I don't see either Apple or Linux folks with something like that which does make it innovative.
Sure Apple has an eye-candy interface, but I've used OSX and OSX doesn't have anything near Avalon. Linux has it's various window managers, which you can take how you want.
I'm not a Microsoft fanboy, I'm waiting to try it first hand.
I agree with you that there have been incremental improvements recently, better filesystem here, better version of a popular software product there, but nothing as groundbreaking as a graphical user interface or the internet itself has come along in a long time. Earlier today, I watched Pirates of Silicon Valley and I was left to wonder what it would be like to be on the cliff of innovation like Gates and Jobs were. My next thought was "when is the next instance of revolutionary technology going to hit?"
My point is that it's the commercial software firms are making the innovations, albeit small, because they are being paid to. OSS alternatives are being written for lots of reasons, mainly because they believe in OSS and the idea of free software. They aren't in any hurry because they know they won't be the first to market so they run at their own pace.
This is the fundamental difference between pioneering commercial software firms and groups of talented programmers write software so they can give it away and say "see, you don't have to pay for it".
All those are simply nice features within an application. For example, some Mozilla guy looked at Mozilla and then to RSS feeds and had a "eureka" moment. That's hardly innovation, it's just combining an existing program that intersects with something else enough so it makes sense to combine the two. My argument was aimed at the notion of an entire application.
The problem is that OSS will always be catching up. I have yet to see popular OSS software show real innovation. They are just alternatives to existing applications. While that itself isn't bad, the innovators that pioneer original applications will continue working and breaking new ground and making money until free competitors catch up to them.
I never said novels suck, they are useful to some people. For you, dimension A is filled by a novel, but for me, dimension A is filled with technical knowledge. I have an imagination but not in the way most people know. I mentioned LotR because traditionally that book series and other fantasy novels were touted as the only way one can use their imagination, but it's just not true.
I disagree because novels don't provide mind expansion. If I pick up a book, I don't have this overcoming urge to fill my head with it's knowledge.
I have tried to read in the past and was unsuccessful. In high school, teachers saw potential in me and put me in advanced math and english classes (which called for summer reading). I did fine in advanced math but it took half the summer to read one book and there were 3 to read.
I don't see that as any type of shortcoming because I know I have imagination. I can visualize technical things like shifting bits in a compression algorithm. I can also explain characteristics of the algorithm based on input it's been given and the output it produces. I just can't visualize a forest while I might read Lord of the Rings.
Maybe books in their current hardcover forms are becoming obsolete, maybe only bibliophiles are the only ones that want to keep books around. I am tired at how someone like yourself condescends down to anyone that doesn't absolutely treasure the same classic literature you love citing some imaginary lack of intelligence. Do I condescend down upon you if you don't like my kind of video games?
Not so fast...
It'll work but it's not a finished product without the CS guys. Likewise, the CS guys wouldn't have a job if it weren't for us EE/ECE guys to provide new technology for the market.
It depends on what you're making too. I interviewed for a popular internship with a company called Leitch. Their local office created a PowerPC-based multichannel video communication system for use in hospitals and anyone that wants to make a television signal with a custom channel lineup. The product used basically only EE/ECE people because there would be an extremely minimal interface for the end users.
heh, you shouldn't, I just envision technology to be an assembly line.
1. Someone has an idea.
2. The EE's and I fabricate it.
3. You CS guys make it perform it's purpose.
4. IT guys make sure it keeps performing it's purpose.
I'm in Computer Engineering myself (graduating in May myself). I don't think I'll be frying bacon or anything. I've looked at jobs and companies like Intel and Siemens are looking for fresh college grads, as well as smaller firms.
I've been looking for jobs in the Boston area and they have the same aforementioned jobs with Intel, Siemens and such, however they also have smaller firms that were put together by MIT graduates that are also looking for EE/ECE graduates fresh out of college.
So in conclusion, I'm not worried. I'm aware I won't have a glorious 6-figure salary in July, I know that, I know I need to pay my dues as far as job quality, but it's not like Computer Engineering will be obsolete in 5 months. I'm so far upstream past the IT/CS people that I can land practically anywhere in computing I want, but I'd rather be in computer engineering because the work is more engaging than anything else.
Don't worry about your degree, brother/sister, you won't "make it big" instantly after you graduate but you'll be just fine.
I'm still waiting for my "+1: Skank" mod to be approved.
Not funny, but sad in a conflict-of-interest sort of way. Whatever puts the 5 cents in their pocket I guess, integrity be damned.
You are correct, but I am speaking from the computer engineering curriculum and XOR's are very important to us because we use them alot and the increase in hardware cost and footprint for having to substitute I mentioned in the grandparent instead of a single XOR gate is unacceptable. However, yes, a NAND in a more general logic design is pretty useful.
NAND isn't even needed for XOR, just NOT, AND and OR:
A XOR B = ((NOT A) AND B) OR (A AND (NOT B))
Well, she plays Playstation 2 games too and I'm guessing the standalone DVD player doesn't understand Playstation 2 discs.
I was given a Playstation 2 by my sister because of DRE's. She used it to read DVD's most of the time and that's the prime reason why DRE's occur.
I opened it and tried working with the adjustment wheel but it didn't fix the DRE. I read another fix is that the laser isn't receiving enough power and didn't want to handle all that. I saw the lense was dirty, cleaned it and it works fine. Since then my sister bought another Playstation 2 so I got a free Playstation 2.
Well, that's if it's under warranty which the first generation models are well past the warranty period. The cost to fix the Playstation by Sony is close to the cost of a new Playstation. I assure you that you are the exception and not the rule.
First generations aren't always the best, look at the Sony Playstation 2.
Do a google search for the terms "disc read error" and find results about Sony Playstation 2's having Disc Read Errors. This is because the first models used plastic pieces handling the laser but the later models used metal pieces that don't fair after a while. It's not quite "water under the bridge" when so many results come up with a relatively unspecific search terms all point to one problem.
A guy I talked to arranged basically an digital EQ on his roof.
He took green, yellow and red lights and arranged them in 16 rows of 16 colums where the bottom 8 light blocks of each column were green, the last 8 were divided between 4 yellow and 4 red. The columns divided among the frequency range he was going for. He built a piece of hardware to control on and off and mated it with some software to take Winamp music and read the frequency and power data and adjusted the lights accordingly to make his roof lights change to show an EQ. He also put some speakers outdoors to play the music the EQ was showing to people that stopped by. I had some pictures but they are unfortunately gone, that was pretty hot too.
That was true too, it wasn't just one issue that did DEC in. DEC had a great thing but marketing flops and the partnership you described was their downfall.
Alpha was plenty better than the Motorola 68x architectures out at the time (and better than Intel's but that should be given) but DEC just didn't know how to do anything else except design the hardware. When they finished and showed off the Alpha at conventions, other companies had equally complex hardware and software to demonstrate with. DEC had a much better hardware design but maybe 2 half-ass programs to show the thing turns on and does a couple things. Then Compaq bought them but didn't take the line any further, just wanted to bleed out what was left of the Alpha.
I think you misunderstood me, the carry flag in the PSW or CCR (or whatever you call it) would be set if it should be. If we are on the assembly level, you'd have to employ add/sub with carry/borrow instead of vanilla add/sub. The compiler would have to reasonably account for this.
I don't know about your first 2 reasons, never dealt with that. However, on point 3, you can't dismiss overflows just because you might not think they will happen often. You have to accomodate for all special cases otherwise the design is only partially correct.
I know homemade arch's should do it in a single clock, but I'm slicing the issue into smaller amounts of time. It's a better practice to make it as fast as possible as soon as possible because if you start pipelining your architecture and find out your ALU add is the slowest (unlikely since memory is the big bottleneck), then you have to rework it anyway.
It'll work for any bus length as long as the XOR operation itself is long enough to support the bus. If A or B is greater than 2^(32) - 1, that's the fault of the previous instruction to cause an overflow. The XOR trick certainely can't be held responsible for problems with code before it.
That's a feature of C then. That's the thing, I'm being taught to become a processor designer. We are taught to not presume anything. The question originally posed to me from my teacher was that we had 2 registers and an ALU that were connected in a loop. ALU's that we deal with don't inherently have that capability. Even if the ALU could detect a carry, the added conditional statements would balloon the instructions enough to make adding/subtracting unreasonable.
We look at ways of getting things done as quick as possible on as many architectures as possible, from the very big (x86, 68k, etc...) to very small (homemade architectures, 6811, etc...).