Re:C++ as a teaching language/programming obscure?
on
Who's Afraid Of C++?
·
· Score: 1
Based on 11223's argument, THE most efficient programming language would be assembler. Having spent the last 5 years programming in assembler, all I can say is, "Yecchhhh!!!" I am now working with C, Java, and hope to use C++ before I'm done. So, I cast my vote in favor of "The right tool for the job" crowd.
However, if I may be so presumptuous, I think that the point he is trying to make is that if you don't understand the platform thoroughly, you won't program as efficiently as possible. The platform includes available libraries, the OS, the CPU, peripherals, networking, etc.
While C is pretty close to the platform (which implies that its programmers are well-informed), it is not immune from inefficiency. A simple example of this is programs that have lots of itty-bitty functions. Those programmers obviously don't realize how much overhead goes into the call/return sequence. At least half, if not all, register contents have to be stored and several registers initialized before the function code can begin to be executed. Then, the process is reversed on return. If the function is as simple as a swap, then more code is executed in the call/return than in the function itself.
Which does lead me to a question for you C++ gurus. Is calling a method the same as calling a C funtion in terms of the call/return processing? If so, then 11223 may have a point regarding efficiency (read bloat). A lot of the class methods I have seen so far are itty-bitty.
To be sure, I have seen lots (and lots...) of examples of programming without a clue of how computers work in other languages, but let's go get a beer before starting that rant.
My experience with patents was replying to one of those "Penny for Your Thoughts" ads. They sent me a packet which required me to provide a written and drawn description of my idea. The idea had to do with anchoring swingset legs, so I didn't care at all if it got ripped off. I just wanted to know about the process.
For this, they did a quick review of the patents that already existed. I was surprised to find out that there were almost a dozen patents that had a little something to do with it. They were obviously found by keyword search, but all of them were a little bit like what I was suggesting.
So far, it was free. But the next stage was going to cost. They offered me 3 choices. 1) For 20% ownership, I had to pay $4,000; 2) for 50% ownership, I had to pay $9,000; and 3) for 100% owenership, I had to pay $15,000.
What they offered me for my money was the services of a patent lawyer, prototyping, and display at trade shows. Altho' I bailed at this point, it wasn't because I thought they were trying to rip me off. It was because I didn't believe that I would even make $3,000 from swingset anchors.
What I suggest to you is that you document your invention and have those documents notarized. At that point, you have dated proof that you invented it. Then, start talking to reputible venture capitalists. If you can demo your invention in blackbox mode, fine. If they are impressed and want to write a contract, then I wouldn't worry about getting ripped off.
The next thing you have to worry about is that your invention gets promoted at each stage. You have to deal with manufacturing, marketing, sales, accounts receivable, taxes, legal protection, and a ton of other stuff. This is where the venture capitalist should be even more help than just providing money. They should have trusted contacts to take care of any and all of this, which beats your trying to do it yourself or find professionals on your own.
So, the bottom line is get everything in writing. If it's really important, insist on a contract.
In light of the bad feelings caused by the Corel Linux beta test license, it seems possible that the GPL in particular, and OS in general being weakened by commercial ventures trying to cash in on the good press. What is your position on the Corel licensing and the potential larger issues coming out of it?
Cringely made at least one good point even tho' it has been made before (ESR comes to mind). That is that M$ is afraid of open standards.
So, instead of getting on the StarOffice/Applixware/AbiSource/whatever soapbox, the Linux/Open Source community should be screaming and shouting about open standards. TCP/IP and HTML have proven what benefit there is to open standards. So, why not open standards for all software?
Word processors have been around for a long time, so why isn't there a WP standard? You basically need formatting (indentation, justification, margins, etc.), fonts, chaptering, indexing, etc. If that were standardized, then all WPs could import and export docs that all others could read.
The same can be said about spreadsheets and presentation packages. DBs already have SQL which is somewhat standardized. And there are probably applications that I never use that could be included.
Let's climb on the standards bandwagon, and ask the other OS backers to join us. If all the BSDers, BEOSers, UN*Xers, Hurders, OS/2ers, and any otherers could define standards and write all Open Source software to them, then a lot of the commercial companies that have come to Linux in hopes of saving themselves from the M$ juggernaut would pretty much have to play along. And thanks to the success of the Internet, open standards carries a lot of weight with business AND end users.
As posted above, have your mgmt/HR dept. come up with a definition of what they mean by porn, and make the scanning policy public--including what will happen to people caught violating it. Whether the company is right or not, anyone with the moxy to take them to court will have an easier time doing it if they don't do this up front.
Then ask mgmt just how they expect you to scan for porn. Are they looking for curse words in the subject or text? If so, is s*ck one? Are they looking for picture attachments--how do you scan for that? Are they looking for URLs to porn sites? (Do you have a list of all porn sites on the web--you really should publish that --> so everyone knows NOT to go there;-)
When mgmt comes up with these plans to "increase productivity", I "respectfully recommend" that they work their own derrieres (is this porny?) off figuring out how to implement them.
I also wonder how the people that come up with these plans have so little imagination that they can't figure out that as much, if not more, time will be wasted trying to come up with ways to get around the ban, as was spent by a few folks who just needed a little stress relief. I mean, what proof do they have that the activity exists to the extent that it warrants spending a sysadmin's time playing junior vice squad?
... then networks could be defined to have different masks than are currently used.
When a company reserves a Class B license (and only companies do), they get 64k host addresses. In the 3 companies I have worked for with Class B licenses, over half of the host addresses were wasted. So, by allocating addresses using a subnet between Class C (255.255.255.0 -> 16m nets) and Class B (255.255.0.0 -> 64k nets), such as 255.255.240.0 (x'ff.ff.f0.00' -> 1m networks with 2k hosts each), a lot of waste could be eliminated.
However, at this stage of the game it would mean taking back all of the addresses and reassigning them, PLUS rewriting every IP stack in the world.
IPv6 is also about more than just addressing. So, the idea of staying with IPv4 means giving up those improvements.
I just hope that the DEA doesn't read this paper, because if they aren't already monitoring this guy, they will be;-), and he was only trying to get certified, when he shorted out a couple of brain cells and thought he had stumbled onto a magical fix for the infamous address shortage:-o
This is similar to the issue of harassment. And the way that has been handled is for companies to have clear statements of policy and respond appropriately to complaints.
In this case, the Linux community's statement of policy is the "Linux Advocacy HOWTO".
The consensus of the responses to articles such as this one, seems to be that we wish everyone could refrain from using gutter language. However, in many cases, the original article/posting seems to be taunting the community which guarantees at least some flames.
So, when flames are published (or even referred to), it would be really slick if one of the Linux portals either gathered comments that are more representative of the whole community or invited a representative of the community to post a response. That would hopefully make it to some of the mainstream press and demonstrate that the community does have its act together, in spite of a relative few who can't control their tempers/tongues.
This is similar to the issue of harassment. And the way that has been handled is for companies to have clear statements of policy and respond appropriately to complaints.
In this case, the Linux community's statement of policy is the "Linux Advocacy HOWTO".
The response is probably not what Mindcraft would like, but then how many objects of harassment are happy with the result? The consensus seems to be that we wish everyone could refrain from using gutter language, but to a certain extent, Mindcraft brought this on themselves.
It would be really slick if one of the Linux portals either gathered comments that are more representative of the whole community or invited a well known representative of the community to post an article. That would hopefully make it to some of the mainstream press and demonstrate that the community does have its act together, in spite of a relative few who can't control their tongues.
The reason companies are created is to make money. The reason that companies go public is to grow and eventually make more money. If you don't like it, go join a tribe in South America and live off the land.
The companies that are making money from open source software may be run by people who believe that the open source movement is a good thing, but they are not in business simply to promote it. Those who do care more about the movement than the companies should try to gain influence over the companies.
I wouldn't worry as much about M$ or IBM muscling their way into Redhat's or VA's business (the SEC has rules about this) nearly as much as the toll of being a public company would take on them. From SEC red tape, to a PR dept. making sure press releases don't cause the stock to drop, to having to meet quarterly quotas, to dozens of other worries, going public is serious business.
1) When a company goes public, they have to keep an eye on the stock price because they are legally committed to making a profit for their investors. 2) A public company is exposed to hostile (or at least not-so-nice) takeovers that can result in major dislocations of employees and product support--even the possibility of dropping a product line. 3) Usually, investors don't try to micro-manage the company, but investors with a significant % of shares can create havoc by voting for/against members of the board of directors and other issues.
I like the posting suggesting that an "Open Source" mutual fund be created. This is an excellent idea because if there were an "Open Source" mutual fund, it could be chartered to help alleviate the above pressures put on companies by investors. This would also provide a source of capital with "pure" motives.
The purpose of the fund could be designed to be supportive, such as requiring that as much as 25% of the investment money be put into start-ups that meet certain requirements. These requirements could be using open source licenses for most of their software, or supporting open source software in most of their customer contracts.
The fund could be modeled on the "Social Conscience" funds (for instance, investing in companies that are pro-environment, etc.). I don't know if it would have to invest in public companies only, but I suspect that if the charter was clear, it would not have to limit itself to that.
The obvious question is, "Who would invest?" Well, since IBM, Intel, etc. have already started, I would say that once the fund was set up, it would probably have no trouble attracting investors. And since the charter would plainly state that it's intention is to invest in companies for the purpose of encouraging the open source movement, no one could easily corrupt it. I would like to see that the minimum investment be no more than $1,000, and preferably $200.
Finally, this could be a way to reward the individual contributors of open source code. The fund could set aside some shares that would be given to programmers using open source licenses. Or the fund could provide something like the grants that are given to universities for specific projects.
Do we have any financial movers and shakers that know enough about this to get it going?
I started in this business working for Amdahl Corp. 15 yrs. ago. For those who don't know, Gene Amdahl was the lead engineer for the System/360 in the 1960s. When he designed the System/370, he wanted to use state-of-the-art VLSI technology, but the bean-counters at IBM said that it would cost too much to retool the manufacturing plants.
So, Gene went out and found financing and built the Amdahl 460/V6. If you've ever seen a chart of the cost of hardware vs. the cost of software, the point at which hardware costs going down cross the line of software costs going up is the year that Amdahl starting selling mainframes.
I later became an IBM customer and was underwhelmed with the service. (Of course, this was about the time that M$ was spanking them, and Louis Gertner(?) said, "LET THE LAYOFFS BEGIN!" Now, I am again working for a competitor. I can assure everyone that IBM plays the same games that M$ does, even today.
I have also had several experiences with IBM's lack of coherence to standards, including their own proprietary ones. They come close, but when they miss, they might or might not correct the problem. If they don't, then they pretend that they are right and often win by sheer weight. (ie. An AS/400 TCP/IP app was receiving too much data. The IP header length field was correct, so the stack was obviously not reading it, but just taking whatever landed in the input buffer. Yet, they spent a month insisting that the sending host's driver was broken! It was only after the customer collected traces at about a dozen points along the path that they finally admitted their mistake and fixed it. I've got dozens of these examples.)
This is why I am not pleased that IBM has noticed Linux. They talk the talk, but they don't walk the walk! How much Open Source have they contributed. Some public domain sample code is all I've ever seen.
At least with M$, you know what you're dealing with. Now that IBM is "on our side", we'd better watch our back.
RMS should read the GPL
on
GNU Inside?
·
· Score: 4
RMS is making a mockery of the GPL by insisting that Linux distributions put GNU in the name. From the GPL:
0. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language.
Linus has met and championed these conditions. The Program, Linux, is licensed in the letter and spirit of the GPL. Now, if FSF wants to package and distribute Linux and call it GNU/Linux, that is their right, because Linus followed the rules. But if Red Hat or Suse or anyone else wants to package and distribute Boomshakalakalaka/Linux, that is their right also.
I want to know from RMS, "What will it take to satisfy you other than this silly name thing? Is it money you are after? Fame? Honoraria? What is it you want?" If he can/will answer that, then everyone who has benefitted from the fanatastic software produced by FSF should make an attempt to satisfy him and then we can get on with the real business of building an environment where source code is freely distributed.
That environment is far from guaranteed. With the suits smelling money, the GPL is sure to come under attack. And if there is a pile of petty bickering going on within the community, there won't be enough resistance to prevent them from corrupting the intentions of open source and/or free software.
You could also buy something from FSF. They make a little profit off the CDs and books they distribute. And they are happy to take donations. And, if you're really feeling guilty, they have a list of projects that need help, including documentation--if you're not a coder.
OK, I stand corrected. I thought we had broken 0.5G at some point. But I looked it up, and we're at about 280M. The point is that there is a tiny minority of teenagers that are going beserk, while the rest (around 30M) are basically trying to get a handle on this crazy world. Some of them look and act strange by mainstream standards, but hey, it's America. Deal with it!
I read GEB when I was an undergrad in CS. (That wasn't really 20 years ago, was it?!?) I was just barely able to follow the math and AI theory. But thanks to the Escher and Bach interludes, I got a chance to stop climbing mental mountains and enjoy the scenery.
In the new preface, DH says, "GEB is a very personal attempt to say how it is that animate beings can come out of inanimate matter." (I will be rereading it with that in mind.) This is more than just AI. For some additional insight into this theme, check out "Emergence: From Chaos to Order" by John H. Holland. I think he even references GEB there.
First, there are about 600M people in the US, and if about 10% are teenagers, then that makes 60M. To profile 60M people based on the extreme psychotic behavior of a very few is stupid.
Second, what's coming out about the guys in Colorado is that they were heavily into Nazism and only occasionally hung out with the TCM. They also built bombs and managed to get hold of illegal firearms. The profile to watch for is hate for racist/religious/political reasons, and disrespect for life. Kids playing video games and wearing unusual clothing, etc. don't match that profile.
To the kids that sent Jon the EMails: don't think that the experiences you have had suddenly started a few years ago. I am in my 40s, and these EMails brought back memories of my high school experience that I had swept under the rug. However, let me give you some hope and some alternatives.
In the first place, developing friendships in high school is like shopping at 7-Eleven: there isn't much choice, and you don't get much value for what you put into it. If you have 1 or 2 friends there, you're lucky. So try to expand your territory. Find organizations outside of school that accept you as you are (my Linux user's group would probably welcome you with open arms). Also, keep your eyes on the future. College is totally different from high school, and you will very likely meet several people there that you can relate with.
For those who survived high school without any major emotional bruises, let's try to give these kids an outlet. Sponsor a geek club of some sort (write some new video games and make a bunch of money--then you'll have the mainstream singing your praises!)
But for the time being, you might as well accept the fact that the mainstream is going to flail around trying to DO something and after this passes from the nightly news, we can get on with our weird lives.
The reason Caldera is doing this is to make money. They don't seem to care about the "movement", but why should they? There's nothing in the GPL that says you have to wear your heart on your sleeve.
A lot of the above postings seem to take the position of, "What does this mean to me, Al Franken?" Hopefully, it means that there will soon be enough growth in the Linux user market that hardware vendors will provide M$ and Linux drivers and more software vendors will port their applications. If that happens, then even if you use Slackware, you will benefit.
So, let's just hope that this brings another chunk of users over, and stop worrying about whether it passes the litmus test of "good Linux".
BTW, I have been using Calera's distro (currently OpenLinux 1.2) for the past 3 years as my primary home OS. This is because at some time during the week, I also use M$ Win 3x, Win 9x, NT, Sun/OS, AIX, IBM OS/400, IBM VM/ESA, and IBM MVS/ESA, and I didn't feel like spending the time up front learning everything I needed to know to build a usable system. However, in that 3 years, I have finally learned enough about X to edit.Xresources and.Xdefaults, I have downloaded enough software that needed to be tweaked that I am getting comfortable with recompiling source, and I have built my own IDE with vim, ddd, and xterm for testing and running tkinfo.
Slashdot Mirror
Based on 11223's argument, THE most efficient programming language would be assembler. Having spent the last 5 years programming in assembler, all I can say is, "Yecchhhh!!!" I am now working with C, Java, and hope to use C++ before I'm done. So, I cast my vote in favor of "The right tool for the job" crowd.
...) of examples of programming without a clue of how computers work in other languages, but let's go get a beer before starting that rant.
However, if I may be so presumptuous, I think that the point he is trying to make is that if you don't understand the platform thoroughly, you won't program as efficiently as possible. The platform includes available libraries, the OS, the CPU, peripherals, networking, etc.
While C is pretty close to the platform (which implies that its programmers are well-informed), it is not immune from inefficiency. A simple example of this is programs that have lots of itty-bitty functions. Those programmers obviously don't realize how much overhead goes into the call/return sequence. At least half, if not all, register contents have to be stored and several registers initialized before the function code can begin to be executed. Then, the process is reversed on return. If the function is as simple as a swap, then more code is executed in the call/return than in the function itself.
Which does lead me to a question for you C++ gurus. Is calling a method the same as calling a C funtion in terms of the call/return processing? If so, then 11223 may have a point regarding efficiency (read bloat). A lot of the class methods I have seen so far are itty-bitty.
To be sure, I have seen lots (and lots
... "The king has no clothes!"
My experience with patents was replying to one of those "Penny for Your Thoughts" ads. They sent me a packet which required me to provide a written and drawn description of my idea. The idea had to do with anchoring swingset legs, so I didn't care at all if it got ripped off. I just wanted to know about the process.
For this, they did a quick review of the patents that already existed. I was surprised to find out that there were almost a dozen patents that had a little something to do with it. They were obviously found by keyword search, but all of them were a little bit like what I was suggesting.
So far, it was free. But the next stage was going to cost. They offered me 3 choices. 1) For 20% ownership, I had to pay $4,000; 2) for 50% ownership, I had to pay $9,000; and 3) for 100% owenership, I had to pay $15,000.
What they offered me for my money was the services of a patent lawyer, prototyping, and display at trade shows. Altho' I bailed at this point, it wasn't because I thought they were trying to rip me off. It was because I didn't believe that I would even make $3,000 from swingset anchors.
What I suggest to you is that you document your invention and have those documents notarized. At that point, you have dated proof that you invented it. Then, start talking to reputible venture capitalists. If you can demo your invention in blackbox mode, fine. If they are impressed and want to write a contract, then I wouldn't worry about getting ripped off.
The next thing you have to worry about is that your invention gets promoted at each stage. You have to deal with manufacturing, marketing, sales, accounts receivable, taxes, legal protection, and a ton of other stuff. This is where the venture capitalist should be even more help than just providing money. They should have trusted contacts to take care of any and all of this, which beats your trying to do it yourself or find professionals on your own.
So, the bottom line is get everything in writing. If it's really important, insist on a contract.
In light of the bad feelings caused by the Corel Linux beta test license, it seems possible that the GPL in particular, and OS in general being weakened by commercial ventures trying to cash in on the good press. What is your position on the Corel licensing and the potential larger issues coming out of it?
I was referring specifically to standards for word processing, spreadsheets, presentation software, etc. I'm already 100% behind existing standards.
Cringely made at least one good point even tho' it has been made before (ESR comes to mind). That is that M$ is afraid of open standards.
So, instead of getting on the StarOffice/Applixware/AbiSource/whatever soapbox, the Linux/Open Source community should be screaming and shouting about open standards. TCP/IP and HTML have proven what benefit there is to open standards. So, why not open standards for all software?
Word processors have been around for a long time, so why isn't there a WP standard? You basically need formatting (indentation, justification, margins, etc.), fonts, chaptering, indexing, etc. If that were standardized, then all WPs could import and export docs that all others could read.
The same can be said about spreadsheets and presentation packages. DBs already have SQL which is somewhat standardized. And there are probably applications that I never use that could be included.
Let's climb on the standards bandwagon, and ask the other OS backers to join us. If all the BSDers, BEOSers, UN*Xers, Hurders, OS/2ers, and any otherers could define standards and write all Open Source software to them, then a lot of the commercial companies that have come to Linux in hopes of saving themselves from the M$ juggernaut would pretty much have to play along. And thanks to the success of the Internet, open standards carries a lot of weight with business AND end users.
As posted above, have your mgmt/HR dept. come up with a definition of what they mean by porn, and make the scanning policy public--including what will happen to people caught violating it. Whether the company is right or not, anyone with the moxy to take them to court will have an easier time doing it if they don't do this up front.
;-)
Then ask mgmt just how they expect you to scan for porn. Are they looking for curse words in the subject or text? If so, is s*ck one? Are they looking for picture attachments--how do you scan for that? Are they looking for URLs to porn sites? (Do you have a list of all porn sites on the web--you really should publish that --> so everyone knows NOT to go there
When mgmt comes up with these plans to "increase productivity", I "respectfully recommend" that they work their own derrieres (is this porny?) off figuring out how to implement them.
I also wonder how the people that come up with these plans have so little imagination that they can't figure out that as much, if not more, time will be wasted trying to come up with ways to get around the ban, as was spent by a few folks who just needed a little stress relief. I mean, what proof do they have that the activity exists to the extent that it warrants spending a sysadmin's time playing junior vice squad?
... then networks could be defined to have different masks than are currently used.
;-), and he was only trying to get certified, when he shorted out a couple of brain cells and thought he had stumbled onto a magical fix for the infamous address shortage :-o
When a company reserves a Class B license (and only companies do), they get 64k host addresses. In the 3 companies I have worked for with Class B licenses, over half of the host addresses were wasted. So, by allocating addresses using a subnet between Class C (255.255.255.0 -> 16m nets) and Class B (255.255.0.0 -> 64k nets), such as 255.255.240.0 (x'ff.ff.f0.00' -> 1m networks with 2k hosts each), a lot of waste could be eliminated.
However, at this stage of the game it would mean taking back all of the addresses and reassigning them, PLUS rewriting every IP stack in the world.
IPv6 is also about more than just addressing. So, the idea of staying with IPv4 means giving up those improvements.
I just hope that the DEA doesn't read this paper, because if they aren't already monitoring this guy, they will be
This is similar to the issue of harassment. And the way that has been handled is for companies to have clear statements of policy and respond appropriately to complaints.
In this case, the Linux community's statement of policy is the "Linux Advocacy HOWTO".
The consensus of the responses to articles such as this one, seems to be that we wish everyone could refrain from using gutter language. However, in many cases, the original article/posting seems to be taunting the community which guarantees at least some flames.
So, when flames are published (or even referred to), it would be really slick if one of the Linux portals either gathered comments that are more representative of the whole community or invited a representative of the community to post a response. That would hopefully make it to some of the mainstream press and demonstrate that the community does have its act together, in spite of a relative few who can't control their tempers/tongues.
This is similar to the issue of harassment. And the way that has been handled is for companies to have clear statements of policy and respond appropriately to complaints.
In this case, the Linux community's statement of policy is the "Linux Advocacy HOWTO".
The response is probably not what Mindcraft would like, but then how many objects of harassment are happy with the result? The consensus seems to be that we wish everyone could refrain from using gutter language, but to a certain extent, Mindcraft brought this on themselves.
It would be really slick if one of the Linux portals either gathered comments that are more representative of the whole community or invited a well known representative of the community to post an article. That would hopefully make it to some of the mainstream press and demonstrate that the community does have its act together, in spite of a relative few who can't control their tongues.
The reason companies are created is to make money. The reason that companies go public is to grow and eventually make more money. If you don't like it, go join a tribe in South America and live off the land.
The companies that are making money from open source software may be run by people who believe that the open source movement is a good thing, but they are not in business simply to promote it. Those who do care more about the movement than the companies should try to gain influence over the companies.
I wouldn't worry as much about M$ or IBM muscling their way into Redhat's or VA's business (the SEC has rules about this) nearly as much as the toll of being a public company would take on them. From SEC red tape, to a PR dept. making sure press releases don't cause the stock to drop, to
having to meet quarterly quotas, to dozens of other worries, going public is serious business.
1) When a company goes public, they have to keep an eye on the stock price because they are legally committed to making a profit for their investors.
2) A public company is exposed to hostile (or at least not-so-nice) takeovers that can result in major dislocations of employees and product support--even the possibility of dropping a product line.
3) Usually, investors don't try to micro-manage the company, but investors with a significant % of shares can create havoc by voting for/against members of the board of directors and other issues.
I like the posting suggesting that an "Open Source" mutual fund be created. This is an excellent idea because if there were an "Open Source" mutual fund, it could be chartered to help alleviate the above pressures put on companies by investors. This would also provide a source of capital with "pure" motives.
The purpose of the fund could be designed to be supportive, such as requiring that as much as 25% of the investment money be put into start-ups that meet certain requirements. These requirements could be using open source licenses for most of their software, or supporting open source software in most of their customer contracts.
The fund could be modeled on the "Social Conscience" funds (for instance, investing in companies that are pro-environment, etc.). I don't know if it would have to invest in public companies only, but I suspect that if the charter was clear, it would not have to limit itself to that.
The obvious question is, "Who would invest?" Well, since IBM, Intel, etc. have already started, I would say that once the fund was set up, it would probably have no trouble attracting investors. And since the charter would plainly state that it's intention is to invest in companies for the purpose of encouraging the open source movement, no one could easily corrupt it. I would like to see that the minimum investment be no more than $1,000, and preferably $200.
Finally, this could be a way to reward the individual contributors of open source code. The fund could set aside some shares that would be given to programmers using open source licenses. Or the fund could provide something like the grants that are given to universities for specific projects.
Do we have any financial movers and shakers that know enough about this to get it going?
I started in this business working for Amdahl Corp. 15 yrs. ago. For those who don't know, Gene Amdahl was the lead engineer for the System/360 in the 1960s. When he designed the System/370, he wanted to use state-of-the-art VLSI technology, but the bean-counters at IBM said that it would cost too much to retool the manufacturing plants.
So, Gene went out and found financing and built the Amdahl 460/V6. If you've ever seen a chart of the cost of hardware vs. the cost of software, the point at which hardware costs going down cross the line of software costs going up is the year that Amdahl starting selling mainframes.
I later became an IBM customer and was underwhelmed with the service. (Of course, this was about the time that M$ was spanking them, and Louis Gertner(?) said, "LET THE LAYOFFS BEGIN!" Now, I am again working for a competitor. I can assure everyone that IBM plays the same games that M$ does, even today.
I have also had several experiences with IBM's lack of coherence to standards, including their own proprietary ones. They come close, but when they miss, they might or might not correct the problem. If they don't, then they pretend that they are right and often win by sheer weight. (ie. An AS/400 TCP/IP app was receiving too much data. The IP header length field was correct, so the stack was obviously not reading it, but just taking whatever landed in the input buffer. Yet, they spent a month insisting that the sending host's driver was broken! It was only after the customer collected traces at about a dozen points along the path that they finally admitted their mistake and fixed it. I've got dozens of these examples.)
This is why I am not pleased that IBM has noticed Linux. They talk the talk, but they don't walk the walk! How much Open Source have they contributed. Some public domain sample code is all I've ever seen.
At least with M$, you know what you're dealing with. Now that IBM is "on our side", we'd better watch our back.
No.
RMS is making a mockery of the GPL by insisting that Linux distributions put GNU in the name. From the GPL:
0. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed
under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program"
means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it,
either verbatim or with modifications and/or translated into another language.
Linus has met and championed these conditions. The Program, Linux, is licensed in the letter and spirit of the GPL. Now, if FSF wants to package and distribute Linux and call it GNU/Linux, that is their right, because Linus followed the rules. But if Red Hat or Suse or anyone else wants to package and distribute Boomshakalakalaka/Linux, that is their right also.
I want to know from RMS, "What will it take to satisfy you other than this silly name thing? Is it money you are after? Fame? Honoraria? What is it you want?" If he can/will answer that, then everyone who has benefitted from the fanatastic software produced by FSF should make an attempt to satisfy him and then we can get on with the real business of building an environment where source code is freely distributed.
That environment is far from guaranteed. With the suits smelling money, the GPL is sure to come under attack. And if there is a pile of petty bickering going on within the community, there won't be enough resistance to prevent them from corrupting the intentions of open source and/or free software.
You could also buy something from FSF. They make a little profit off the CDs and books they distribute. And they are happy to take donations. And, if you're really feeling guilty, they have a list of projects that need help, including documentation--if you're not a coder.
OK, I stand corrected. I thought we had broken 0.5G at some point. But I looked it up, and we're at about 280M. The point is that there is a tiny minority of teenagers that are going beserk, while the rest (around 30M) are basically trying to get a handle on this crazy world. Some of them look and act strange by mainstream standards, but hey, it's America. Deal with it!
I read GEB when I was an undergrad in CS. (That wasn't really 20 years ago, was it?!?) I was just barely able to follow the math and AI theory. But thanks to the Escher and Bach interludes, I got a chance to stop climbing mental mountains and enjoy the scenery.
In the new preface, DH says, "GEB is a very personal attempt to say how it is that animate beings can come out of inanimate matter." (I will be rereading it with that in mind.) This is more than just AI. For some additional insight into this theme, check out "Emergence: From Chaos to Order" by John H. Holland. I think he even references GEB there.
First, there are about 600M people in the US, and if about 10% are teenagers, then that makes 60M. To profile 60M people based on the extreme psychotic behavior of a very few is stupid.
Second, what's coming out about the guys in Colorado is that they were heavily into Nazism and only occasionally hung out with the TCM. They also built bombs and managed to get hold of illegal firearms. The profile to watch for is hate for racist/religious/political reasons, and disrespect for life. Kids playing video games and wearing unusual clothing, etc. don't match that profile.
To the kids that sent Jon the EMails: don't think that the experiences you have had suddenly started a few years ago. I am in my 40s, and these EMails brought back memories of my high school experience that I had swept under the rug. However, let me give you some hope and some alternatives.
In the first place, developing friendships in high school is like shopping at 7-Eleven: there isn't much choice, and you don't get much value for what you put into it. If you have 1 or 2 friends there, you're lucky. So try to expand your territory. Find organizations outside of school that accept you as you are (my Linux user's group would probably welcome you with open arms). Also, keep your eyes on the future. College is totally different from high school, and you will very likely meet several people there that you can relate with.
For those who survived high school without any major emotional bruises, let's try to give these kids an outlet. Sponsor a geek club of some sort (write some new video games and make a bunch of money--then you'll have the mainstream singing your praises!)
But for the time being, you might as well accept the fact that the mainstream is going to flail around trying to DO something and after this passes from the nightly news, we can get on with our weird lives.
The reason Caldera is doing this is to make money. They don't seem to care about the "movement", but why should they? There's nothing in the GPL that says you have to wear your heart on your sleeve.
.Xresources and .Xdefaults, I have downloaded enough software that needed to be tweaked that I am getting comfortable with recompiling source, and I have built my own IDE with vim, ddd, and xterm for testing and running tkinfo.
A lot of the above postings seem to take the position of, "What does this mean to me, Al Franken?" Hopefully, it means that there will soon be enough growth in the Linux user market that hardware vendors will provide M$ and Linux drivers and more software vendors will port their applications. If that happens, then even if you use Slackware, you will benefit.
So, let's just hope that this brings another chunk of users over, and stop worrying about whether it passes the litmus test of "good Linux".
BTW, I have been using Calera's distro (currently OpenLinux 1.2) for the past 3 years as my primary home OS. This is because at some time during the week, I also use M$ Win 3x, Win 9x, NT, Sun/OS, AIX, IBM OS/400, IBM VM/ESA, and IBM MVS/ESA, and I didn't feel like spending the time up front learning everything I needed to know to build a usable system. However, in that 3 years, I have finally learned enough about X to edit
So don't knock making it easy to get started.