If a program as unique and popular as PacMan or VisiCalc was produced by an open source organization, what would happen? Would it be immediately pounced upon by imitators and perhaps even more interesting versions? I'm asking because I'm not sure. I tend to think it would be watered down somehow and I doubt anyone would get much credit or money from their inspiration.
Look at gcc or glibc or linux kernel or libstdc++. These are all hugely popular and used programs with no equally popular forks. It's not worth anyone;s time or effort to maintain a extremely divergant fork. Of course people add new features and code to these projects all of the time but by and large, most of the submissions are folded back into the main branch. Everyone benefits because noone gets a monopoly position with any of these tools.
The problem is a matter of perception IMHO. Games are easy and profitable to produce as closed source because they are one-time programs. kernels and word processors are much easier to produce as open source because they will be used over and over and over again and have more and more features added. You can make money from either. Just look at Redhat, Novell, etc...
This technology would make for awesome screens if they can scale it up to full resolution and make the glass larger.
of course, they're probably hitting some kind of physics barrier which prevents more pixels or large size glasses to be made. Otherwise, this would be very big screen news.
I doubt anyone's going to wear that attachment piece on their eye though, it just too big. Unless it's for work or something. Ben
Design, Design, Design is by far the most important aspect of a good computer scientist/engineer.
Why oh why don't universities teach this as the INTRODUCTORY CLASS??
If you cannot design, then you are a hacker/coder, not a engineer.
This applies if you want to do games or build the next big thing on the internet. It is even fundemental if you want to build your NES emulator.
Design is the art of taking a full level project specification and breaking it down into smaller, easier to handle parts. And then you break down those smaller parts until you have pieces which are easy to test and develop.
It's the ability to look at available packages and choose one based on the functionality it gives you.
Having good design skills is a lot like having good puzzle skills. You have to think about the problem you want to solve from different angles until a pattern jumps out at you and you figure out the basis needed to solve the puzzle. Everything after that is mechanics.
Ideally, you'll be able to design from scratch on every project. However, realistically, you'll have to design within the confines of previously built and running code.
Learn how to design/think abstractly and everything else is just details, Ben
PS. Algorithms and data structures are the tools you'll have when you design so it's good to learn the important ones in parallel. Operating systems and computer architectures are of secondly importance, mostly useful in optimization situations. (I'm coming from an app. developer's prespective)
PSPS. Reading lots of other code and understanding how it works helps a lot.
Duh. M$ is leveraging their desktop operating system monopoly to gain an advantage (cross-subsidizing from their monopoly) in console gaming. That may be illegal.
Perhapse. Mostly, I see it as a ploy to get kids to beg their parents "Oh, I can build games for the 360. Buy me a 360, please!!!!" If Sony was smart, they would go to a major games developer (EA?) and try the same thing for PS3. Or just wait for Mono to be ported to PS3 and then write drivers for it.
I too think this is an innovative thing because hopefully, M$ will develop a decent UI for managing small-scaled 3D games ala flash. It would be interesting to see their interface.
On the other hand, I can (and will) build games for my ipod on linux because 1) I can (gcc toolchain for arm is free and the tool C++ is standard) 2) no one is blocking me from doing so.
I think I've figured out the cruical difference between Creationists and Evolutionists.
Creationists support a static view of the human body and believe it has always retained the same form and structure. They admire the current form of the human body with all of its prefections and don't want to imagine it in a "lesser" state.
Evolutionists support a dynamic view of the human body, where it has changed over time. They admire the adaptions that allowed it to adapt to changing environments. They note the deficencies of the human body and work to discover how it can be improved and how/why it came to be.
I don't know how/why we are intelligent enough to concieve of a God and/or evolution but I'm more inclined to believe the arguments of the evolutionists because: a) they have more evidence b) they have the power to change my environment in a real physical manner (think drugs that defeat natural germs, etc...) where the creationists can only offer faith and comfort. c) the mind's imagination is an extremely powerful force for making the imaginary seem real.
I admire static beauty but admit that all things change. We grow up, we grow older, eventually we will die. Buildings are worn down, they are destroyed/fall down, and new ones are built. Same for trees, plants, seashores, forests, etc. Hopefully, our children will learn from our mistakes and change themselves/their environments so their children can live happier lives.
- the Perforce tools suck so badly. p4v is OK, but the command line tools are unacceptable. They're not too terrible. Of course, I don't care because p4v is amazing anyway!
- the branching and merging features are completely overrated. They're hard to use. They don't give you anything Subversion doesn't already have. They are a hell of a lot better and easier to use though. Well, at least over CVS. (don't know about SVN though)
I personally like perforce a heck of a lot more then clearcase or VSS or CVS. The p4v tool is the major selling point for me. It's alot better designed then the competing SVN tools.
Actually my experience says that it's better to pick a cross compiler toolchain and stick with it. We have to support arm9 and x86 embedded platforms and we do on the same machines using debian testing with two separate toolchains. CXX=..../configure... is the shit!!
Works like a charm with some messing around required.
Hah, we install all of our servers with Debian testing and don't give a rats ass if it's "officially" 64bit or not. 99% of the linux applications are not written to take advantage of 64bit (unless the compiler is a hell of a lot more powerful then it really is) so whatever small performance increase 64bit could provide in terms of server application (mostly code compiling) doesn't really matter...
I hate to break it to you, but the linux community was fragmented A LONGGG time before Microsoft and/or IBM started paying attention to it. The different distros, each with their own little tweaks and changes ensure that Linux is always be fragmented. This is healthy (IMHO) because it reduces monoculture and encourages innovation. If you are not a power user and want to be able to use a stable distrobution, then go with Redhat or SuSe. As long as you keep your home partition separate from the root partition, then you can transition easily when/if your distro bellyflops and/or goes bankrupt.
Ultimately, this won't happen because of the power of opensource. Anyone can fork a distro and maintain it. Anyone can take diffs/code from another distro and apply it to their own. So your favorite distro will probably never die but live on in another name. This is an impossibility in a M$-centered world though. Innovation (except for theirs, and not even so much then...) is discouraged.
I don't understand how anyone would want to outsource all of their manufactoring to China instead of keeping it in-country.
It seems 100% inevenuable that if the product takes off and becomes popular, then it will be ripped off by people on the manufactoring floor and stolen. I guess this balances out the costs of outsourcing even more.
I don't feel a drop of pity for Samsung even while I enjoy using a cheap ripoff copy of thier top-of-the-line DVD player. (Or Nike, or GAP, etc.) hehe
Sigh, I guess that is why a business driving opensource is usually a more likely candidate for success then a group of amarture programmers. Their modivation for continuing to perfect the product is money, not peer-recongition or geek-worship.
I'm not saying FF2.0 is bad, it's just that they need better managers who can prioritize properly managing their great programmers. Otherwise, their programmers will get lost adding new features instead of fixing bugs. That'll probably result in a better product overall, as well.
I guess the real lesson is that Linden Labs needs to figure out how to make instance instansation in SL expensive so it would be worthless to try these types of attacks. Like forcing a confirmation whenever you wanted to create a new object. (or giving each new creation a price, like 1 Linden dollar or something)
Then second life could be exactly like real life...
Ouch, I think my eyes are bleeding from all of the CAPITALIZATION, braces and other symatically terrible language features. I can't even understand what the S: and E: at the beginning of some of the lines are for. C: is for comments and M: is code... I think...
Whoever designed this language should have been forced to write a compiler with it. (Considering hindsight is 20/20.;)
Just for an example, say they contribute a cluster file system. Lots of people start using it and depending on it. Then they fork their code, so that future cluster filesystem modules are proprietary. Sure, someone else could continue the free branch, but if Oracle owns all the talent, the free fork could have a hard time getting started. Now the people who depend on this filesystem are stuck on Oracle if they want patches, updates, etc.
This is not going to happen because of the variety of other FSs out there. If Oracle did stop developing their FS, eventually it would be dropped from the mainline kernel and people would migrate away from it. Just look at what is happening with RasierFSv3. It's no longer being developed and v4 isn't out (and may never get out) so people are slowly moving back to ext3. Even if it was a video driver, if the maintainer decides not to maintain it anymore, then users will migrate to another solution. It's darwinism at it's best. What's important is to allow the migration to happen as opposed to locking in the users and then increasingly charge them more and more for development. Ala M$
From TFA: President Bush has said recently that terror groups were trying to influence public opinion in the US, describing their efforts as the "war of ideas".
Obviously, if Bush and Rummie need to create a counter-idea department in the DOD, then they are losing the "war of ideas".
Here's an idea: Be a better army and work with one side or another in the Iraqi civil war instead of letting both sides slaughter your troops!
I was under the impression that Linus doesn't let any code into his tree without the contributor signing over the copyright to him. He owns the copyright to Linux and that would cover all of the code within, right?
I thought that copyrights don't really have any say with GPLed code....
If Oracle wants to offer support Linux-based database solutions, it ought to come up with it's own distro. NOT kill RedHat.. the no. 1 distro.
I supposed Oracle is going to fork RHEL, something that is clearly allowed in the GPL... which most of RHEL is based on.
I think though that whoever winds up with the majority of the marketplace for Enterprise Linux will be the one providing the better support and bug fixes, something that Oracle doesn't do currently. I think that Oracle is in the much weaker position here and it'll be up to them to prove themselves. I think that Ellision will find that being a vulture in the OSS world won't would as well as it did in the DB world. At least, I hope not.
You never answered the bit about Scalia limiting a state's right to legalize marijuana. Scalia is an activist when he has to rule on an issue.
If he's for it (right to assualt rifles at home) he'll rule against the government. If he's against it (state legalization of marijuana), he'll rule with the government.
If anything, Scalia (and his 2nd, Thomas) use the Supreme Court to preach their morality (conservitism) then any other judge there. Ben
Slashdot Mirror
Is it me or is Firefox looking more like a OS then "normal" application?
Maybe they should be drawing on concepts from the OS community in deciding how to allocate resources and give permissions to processes^H^H^Hscripts?
Maybe having a firefox "top" and "ps" utilities would be very helpful.
Cheers
Ben
If a program as unique and popular as PacMan or VisiCalc was produced by an open source organization, what would happen? Would it be immediately pounced upon by imitators and perhaps even more interesting versions? I'm asking because I'm not sure. I tend to think it would be watered down somehow and I doubt anyone would get much credit or money from their inspiration.
Look at gcc or glibc or linux kernel or libstdc++. These are all hugely popular and used programs with no equally popular forks. It's not worth anyone;s time or effort to maintain a extremely divergant fork. Of course people add new features and code to these projects all of the time but by and large, most of the submissions are folded back into the main branch. Everyone benefits because noone gets a monopoly position with any of these tools.
The problem is a matter of perception IMHO. Games are easy and profitable to produce as closed source because they are one-time programs. kernels and word processors are much easier to produce as open source because they will be used over and over and over again and have more and more features added. You can make money from either. Just look at Redhat, Novell, etc...
Cheers
Ben
This technology would make for awesome screens if they can scale it up to full resolution and make the glass larger.
of course, they're probably hitting some kind of physics barrier which prevents more pixels or large size glasses to be made. Otherwise, this would be very big screen news.
I doubt anyone's going to wear that attachment piece on their eye though, it just too big.
Unless it's for work or something.
Ben
Is funny!!
Design, Design, Design is by far the most important aspect of a good computer scientist/engineer.
Why oh why don't universities teach this as the INTRODUCTORY CLASS??
If you cannot design, then you are a hacker/coder, not a engineer.
This applies if you want to do games or build the next big thing on the internet.
It is even fundemental if you want to build your NES emulator.
Design is the art of taking a full level project specification and breaking it down into smaller, easier to handle parts. And then you break down those smaller parts until you have pieces which are easy to test and develop.
It's the ability to look at available packages and choose one based on the functionality it gives you.
Having good design skills is a lot like having good puzzle skills. You have to think about the problem you want to solve from different angles until
a pattern jumps out at you and you figure out the basis needed to solve the puzzle. Everything after that is mechanics.
Ideally, you'll be able to design from scratch on every project.
However, realistically, you'll have to design within the confines of previously built and running code.
Learn how to design/think abstractly and everything else is just details,
Ben
PS. Algorithms and data structures are the tools you'll have when you design so it's good to learn the important ones in parallel.
Operating systems and computer architectures are of secondly importance, mostly useful in optimization situations. (I'm coming from an app. developer's prespective)
PSPS. Reading lots of other code and understanding how it works helps a lot.
Duh. M$ is leveraging their desktop operating system monopoly to gain an advantage (cross-subsidizing from their monopoly) in console gaming. That may be illegal.
Perhapse. Mostly, I see it as a ploy to get kids to beg their parents "Oh, I can build games for the 360. Buy me a 360, please!!!!"
If Sony was smart, they would go to a major games developer (EA?) and try the same thing for PS3.
Or just wait for Mono to be ported to PS3 and then write drivers for it.
I too think this is an innovative thing because hopefully, M$ will develop a decent UI for managing small-scaled 3D games ala flash.
It would be interesting to see their interface.
On the other hand, I can (and will) build games for my ipod on linux because
1) I can (gcc toolchain for arm is free and the tool C++ is standard)
2) no one is blocking me from doing so.
Cheers
Ben
I think I've figured out the cruical difference between Creationists and Evolutionists.
Creationists support a static view of the human body and believe it has always retained the same form and structure.
They admire the current form of the human body with all of its prefections and don't want to imagine it in a "lesser" state.
Evolutionists support a dynamic view of the human body, where it has changed over time.
They admire the adaptions that allowed it to adapt to changing environments. They note the deficencies of the human body and work to discover how it can be improved and how/why it came to be.
I don't know how/why we are intelligent enough to concieve of a God and/or evolution but I'm more inclined to believe the arguments of the evolutionists because:
a) they have more evidence
b) they have the power to change my environment in a real physical manner (think drugs that defeat natural germs, etc...) where the creationists can only offer faith and comfort.
c) the mind's imagination is an extremely powerful force for making the imaginary seem real.
I admire static beauty but admit that all things change. We grow up, we grow older, eventually we will die. Buildings are worn down, they are destroyed/fall down, and new ones are built. Same for trees, plants, seashores, forests, etc.
Hopefully, our children will learn from our mistakes and change themselves/their environments so their children can live happier lives.
Cheers
Ben
- the Perforce tools suck so badly. p4v is OK, but the command line tools are unacceptable.
They're not too terrible. Of course, I don't care because p4v is amazing anyway!
- the branching and merging features are completely overrated. They're hard to use. They don't give you anything Subversion doesn't already have.
They are a hell of a lot better and easier to use though. Well, at least over CVS. (don't know about SVN though)
I personally like perforce a heck of a lot more then clearcase or VSS or CVS.
The p4v tool is the major selling point for me. It's alot better designed then the competing SVN tools.
Cheers
Ben
Funny, this was on an extremely hard interview I had recently.
I couldn't answer the question...
Ben
Actually my experience says that it's better to pick a cross compiler toolchain and stick with it. ./configure ... is the shit!!
We have to support arm9 and x86 embedded platforms and we do on the same machines using debian testing with two separate toolchains. CXX=...
Works like a charm with some messing around required.
Cheers
Ben
Hah, we install all of our servers with Debian testing and don't give a rats ass if it's "officially" 64bit or not. 99% of the linux applications are not written to take advantage of 64bit (unless the compiler is a hell of a lot more powerful then it really is) so whatever small performance increase 64bit could provide in terms of server application (mostly code compiling) doesn't really matter...
Cheers
Ben
I hate to break it to you, but the linux community was fragmented A LONGGG time before Microsoft and/or IBM started paying attention to it. The different distros, each with their own little tweaks and changes ensure that Linux is always be fragmented. This is healthy (IMHO) because it reduces monoculture and encourages innovation. If you are not a power user and want to be able to use a stable distrobution, then go with Redhat or SuSe. As long as you keep your home partition separate from the root partition, then you can transition easily when/if your distro bellyflops and/or goes bankrupt.
Ultimately, this won't happen because of the power of opensource. Anyone can fork a distro and maintain it. Anyone can take diffs/code from another distro and apply it to their own. So your favorite distro will probably never die but live on in another name. This is an impossibility in a M$-centered world though. Innovation (except for theirs, and not even so much then...) is discouraged.
Cheers
Ben
I don't understand how anyone would want to outsource all of their manufactoring to China instead of keeping it in-country.
It seems 100% inevenuable that if the product takes off and becomes popular, then it will be ripped off by people on the manufactoring floor and stolen. I guess this balances out the costs of outsourcing even more.
I don't feel a drop of pity for Samsung even while I enjoy using a cheap ripoff copy of thier top-of-the-line DVD player. (Or Nike, or GAP, etc.) hehe
Ben
Sigh, I guess that is why a business driving opensource is usually a more likely candidate for success then a group of amarture programmers.
Their modivation for continuing to perfect the product is money, not peer-recongition or geek-worship.
I'm not saying FF2.0 is bad, it's just that they need better managers who can prioritize properly managing their great programmers.
Otherwise, their programmers will get lost adding new features instead of fixing bugs.
That'll probably result in a better product overall, as well.
Cheers
Ben
This is a great funny story.
I guess the real lesson is that Linden Labs needs to figure out how to make instance instansation in SL expensive so it would be worthless to try these types of attacks.
Like forcing a confirmation whenever you wanted to create a new object. (or giving each new creation a price, like 1 Linden dollar or something)
Then second life could be exactly like real life...
Cheers
Ben
This is why I'm listening to the local University music station 90% of the time.
By some miricle, they're not into the "payola" thing and they don't have advertistng. (I know my spelling sucks)
so sometimes they pay really crazy farout music, but most of the time, it's really, really good.
Cheers
Ben
Ouch, I think my eyes are bleeding from all of the CAPITALIZATION, braces and other symatically terrible language features.
;)
I can't even understand what the S: and E: at the beginning of some of the lines are for. C: is for comments and M: is code... I think...
Whoever designed this language should have been forced to write a compiler with it.
(Considering hindsight is 20/20.
Ben
Wow, it sounds like you've been in the same project for far too long.
Isn't code maintance really boring?
Ben
Indeed, why do you need a mouse when you can have emacs instead. It is a desktop environment! ;)
Ben
I'm crazy to reply to a Anon but anyway....
Just for an example, say they contribute a cluster file system. Lots of people start using it and depending on it. Then they fork their code, so that future cluster filesystem modules are proprietary. Sure, someone else could continue the free branch, but if Oracle owns all the talent, the free fork could have a hard time getting started. Now the people who depend on this filesystem are stuck on Oracle if they want patches, updates, etc.
This is not going to happen because of the variety of other FSs out there. If Oracle did stop developing their FS, eventually it would be dropped from the mainline kernel and people would migrate away from it. Just look at what is happening with RasierFSv3. It's no longer being developed and v4 isn't out (and may never get out) so people are slowly moving back to ext3. Even if it was a video driver, if the maintainer decides not to maintain it anymore, then users will migrate to another solution. It's darwinism at it's best. What's important is to allow the migration to happen as opposed to locking in the users and then increasingly charge them more and more for development. Ala M$
Cheers
Ben
From TFA:
President Bush has said recently that terror groups were trying to influence public opinion in the US, describing their efforts as the "war of ideas".
Obviously, if Bush and Rummie need to create a counter-idea department in the DOD, then they are losing the "war of ideas".
Here's an idea: Be a better army and work with one side or another in the Iraqi civil war instead of letting both sides slaughter your troops!
Ben
I was under the impression that Linus doesn't let any code into his tree without the contributor signing over the copyright to him.
He owns the copyright to Linux and that would cover all of the code within, right?
I thought that copyrights don't really have any say with GPLed code....
Cheers,
Ben
If Oracle wants to offer support Linux-based database solutions, it ought to come up with it's own distro. NOT kill RedHat.. the no. 1 distro.
I supposed Oracle is going to fork RHEL, something that is clearly allowed in the GPL... which most of RHEL is based on.
I think though that whoever winds up with the majority of the marketplace for Enterprise Linux will be the one providing the better support and
bug fixes, something that Oracle doesn't do currently. I think that Oracle is in the much weaker position here and it'll be up to them to prove themselves. I think that Ellision will find that being a vulture in the OSS world won't would as well as it did in the DB world. At least, I hope not.
Just my two cents
Ben
Because the consultant analysing the site says so.
;)
So it must be true!!!
Ben
You never answered the bit about Scalia limiting a state's right to legalize marijuana. Scalia is an activist when he has to rule on an issue.
If he's for it (right to assualt rifles at home) he'll rule against the government.
If he's against it (state legalization of marijuana), he'll rule with the government.
If anything, Scalia (and his 2nd, Thomas) use the Supreme Court to preach their morality (conservitism) then any other judge there.
Ben