You're right, let's forget about 'arbitrary ethics'. Let's have the money, the technology, whatever. We have plenty of evidence showing that children can't grow smart, educated and open to this world without a computer and an internet connection. Wait... But how the hell people did it before 1981, without MS-DOS ? Wew, I'm so glad we came out of the dark ages just when I started school.
Seriously, how this post could be mod'ed 'insightful' in the first place is beyond me. Please go to school kiddies, to the ones that bring you cultural content, not 'clicking know-how'.
I have been tickling computers since the age of 7. The only things I learnt at school was that copying software was illegal but you had to do it since even the school could not afford all the licences, that products were opaque, badly documented and only upgraded every few years, creativity and introspection was near to impossible, and generally speaking choice was non-existent. I don't see where 'MS technology in school' will do better.
We could have a debate on 'benefits of computer technology at school', but there can't be one on private interests independence. Most US people believe that giving bucks to MS necessarily participate to a wealthy and highly profitable commercial circuit for their economy. They only fail to notice who get the profits. When MS funds schools, it enlarges the digital divide : you get to use the software, they keep the rest (knowledge for instance).
Anyway, you'll have to explain to me with very clear wording why 700 children won't have to choose between a free solution that gives freedom (to use, share and modify the software) and a costly solution that restricts freedom.
And also more experimental... An OpenGL based toolkit (portable, fast, effects, etc) with native anti-aliased fonts, SVG and 3DS widgets ? XML serialization ? Plenty of doc and examples ? And Free Software ?
Eye candy first : http://ngl.sourceforge.net/shots.php
Ryzom is a game (due to summer 2003) based on the NeL library ( http://www.nevrax.org ). This is free software, not open source : my distinction is merely technical rather than ideological, the business plan of the supporting company being based on this very important detail.
The author of the article as well as many of/. posters fail to notice that running a business and choosing a software licence are two different and orthogonal things. People always mix the two notions because they think you can't run a business based on free software.
Well, I'm writing free software, and being paid for it. The Nevrax team (which writes NeL) has about 30 coders paid to write free software.
The key point : our companies don't sell software, they sell service. MMORPG developers sell server access, of course, but you could run any other support (see RedHat and Mandrake for more details).
Now back to the article, the author asserts that free software is necessarily developed in a Linux way, ie. completely ditributed and out of control from any corporation. Endorsing free software does mean endorsing the 'Bazaar'. You can keep your perfect corporate 'Cathedral' and produce free software, this a matter of license.
The question is not to see if 'free software' developers do it better than non-free one, the question is about choosing a license and understanding its impact.
Now I won't explain here why software should be free, but I can tell you it makes *a lot* of sense for MMORPGs companies. And that free software is not necessarily software developed by long-beared geeks in their spare time, IBM produce a lot of GPL material and they were suits:)
first thanks for the nice first article. I personnaly believe that software should be free, but claiming it to be open (I'd prefer 'disclosed') is a necessary step forward, at least from a practical point of view. I would like to answer to your follow up : my opinion is that you made some step backwards and basically invalidated some of your stronger points in the original article. So here are my own answers.
Intellectual property
The claim is "Key parts of complex software systems are true intellectual property of the creators, and the owners would never open up that code". I'm wondering what is 'true' IP, and what are the others. You clearly explained that you don't protect your IP by keeping it in your safe, but by relying on copyrights and applicable laws.
There is no such thing as 'corporate secret'. Coca-Cola is well known for keeping is recipe ultra secret, and only a few blessed people are supposed to have access to it. What for ? We can reverse-engineer Coca-Cola's recipe, and have plenty of means to find an industrial process which gives the same drink. It did not stop concurrents either. I wouldn't be surprised if Pepsi actually needed more efforts to find an alternative to Coca-Cola rather than plainly copying it. Coca-Cola currently tries to hide the fact that nothing can't prevent a smart company to produce a clone and sell it. Nothing but Coca-Cola's brute marketing and intimidating force. Oh wait, yes, if Coca-Cola recipe was patented, they could prevent that. But it would be disclosed.
Similarly, there is no such thing as 'crucial algorithm'. There are only smart implementations. Reverse engineering software is extremely easy today. Clever software writers know that, and may elect to patent and thus disclose their algorithm. It's only a smart move if you are cynical : most patents are obscure descriptions vaguely desribing their purpose, and we had plenty of cases and time to realize that algorithms should not be patented (thanks Mr. Jeff Bezos for its spectacularly stupid and demonstrative patents). But I'm disgressing, I only wanted to show that you can't protect your IP until you disclose it.
Now if I get a video encoding tool, and if the source code only shows the UI part because the vendor removed the 'crucial algorithms', I'd be rather disappointed. Be sure that all vendor will consider the core of their software as 'crucial algorithm'. Be sure that the only interesting evaluation of quality can be made on this core part.
I ran in two recent exemplary situations : both Nvidia and 3DLabs disclosed some sources related to their 'shader language compilers', boasting about their transparency and such. It turns out that their code only contain front-ends. A piece of code that you can almost automatically generate from the language specs. The back-end is the core part I would like to see to make my opinion (and fix bugs, and improve, etc). But these corporations don't want that. You end up trusting them with curious metrics (for a developper) : "I'm used to Nvidia products, they usually work well".
International copyright laws
I'm an European (French to be more precise). I'm not sure how I should handle your paragraph. It basically states that countries outside the US have no laws. As a spoker from 'dark warehouses in faraway places', I can testify that we do have laws. Some even prevent people from stealing.
I'm being sarcastic because first I think you deserve it, and second because this kind of ignorance is what is playing against your 'open your software' argument. All countries (including non-democratic ones) have plenty of laws regarding copyrighted materials and IP protection.
I'm disclosing all the software I write. It only takes two steps:
- I'm applying a copyright, which says who authored the code and when. This let me claim my author rights anywhere on the globe.
- I'm choosing a license which says how my material should be copied, distributed, and so on. I can do that because I'm the copyright holder. Now I juste have to find a good license which fullfills my need and is known to work in all courts (eg. http://www.gnu.org/licenses/licenses.html), or get a good lawyer write one for me.
This reminds me of the NRA fanatics behaviour : they pretend they need guns to protect themselves before even knowing - or wanting to know - how their country is protecting them and if there's any danger in the first place. This harsh comparison emphasizes two key points : there are laws that protect you (or your IP), and there are not thieves at every corner of the streets.
National Security
I would insist more heavily on this point. National Security is the last domain I would ever like to see closed software. Thinking that my national secret services rely on software which are secret to them is either a joke or a nightmare. And eminent cryptologists have made their point clear since a long time : there's no security through obscurity, only insecurity.
Already inspected
Microsoft claims to closely inspect its software, and even launched a noisy campaign along the lines last year. Are you trusting Microsoft software ? I expect to get a full panel of anwsers from "no way" to "sure".
Objective quality can only be achived via disclosure, and only if the disclosure is permanent. You can't pretend you gave the source code to a 3rd party QA company and they were OK. You must keep the inspection possible for anybody at anytime, else you can't build any trust. ISO QA certicates have provision for this : they are only valid if the contents are open and kept open.
Now subjective quality is another point : many companies have their own QA certification process and will want to do the inspection by themselves. I have seen a bunch of them.
There's no such thing as 'already inspected' software.
And what about making fun with those log entries ? My Apache was used to be hit more often by Nimda and CodeRed rather than regular users, so I decided to make an effort for this unexpected public.
Slashdot Mirror
You're right, let's forget about 'arbitrary ethics'. Let's have the money, the technology, whatever. We have plenty of evidence showing that children can't grow smart, educated and open to this world without a computer and an internet connection. Wait... But how the hell people did it before 1981, without MS-DOS ? Wew, I'm so glad we came out of the dark ages just when I started school.
Seriously, how this post could be mod'ed 'insightful' in the first place is beyond me. Please go to school kiddies, to the ones that bring you cultural content, not 'clicking know-how'.
I have been tickling computers since the age of 7. The only things I learnt at school was that copying software was illegal but you had to do it since even the school could not afford all the licences, that products were opaque, badly documented and only upgraded every few years, creativity and introspection was near to impossible, and generally speaking choice was non-existent. I don't see where 'MS technology in school' will do better.
We could have a debate on 'benefits of computer technology at school', but there can't be one on private interests independence. Most US people believe that giving bucks to MS necessarily participate to a wealthy and highly profitable commercial circuit for their economy. They only fail to notice who get the profits. When MS funds schools, it enlarges the digital divide : you get to use the software, they keep the rest (knowledge for instance).
Anyway, you'll have to explain to me with very clear wording why 700 children won't have to choose between a free solution that gives freedom (to use, share and modify the software) and a costly solution that restricts freedom.
Quickly sed'ed for correctness : http://mksp.zerodeux.net/contributors.html.
OK, let's try with an hyperlink.
And also more experimental ... An OpenGL based toolkit (portable, fast, effects, etc) with native anti-aliased fonts, SVG and 3DS widgets ? XML serialization ? Plenty of doc and examples ? And Free Software ?
Eye candy first : http://ngl.sourceforge.net/shots.php
Then clic, clic, yada yada.
Ryzom is a game (due to summer 2003) based on the NeL library ( http://www.nevrax.org ). This is free software, not open source : my distinction is merely technical rather than ideological, the business plan of the supporting company being based on this very important detail.
/. posters fail to notice that running a business and choosing a software licence are two different and orthogonal things. People always mix the two notions because they think you can't run a business based on free software.
:)
The author of the article as well as many of
Well, I'm writing free software, and being paid for it. The Nevrax team (which writes NeL) has about 30 coders paid to write free software.
The key point : our companies don't sell software, they sell service. MMORPG developers sell server access, of course, but you could run any other support (see RedHat and Mandrake for more details).
Now back to the article, the author asserts that free software is necessarily developed in a Linux way, ie. completely ditributed and out of control from any corporation. Endorsing free software does mean endorsing the 'Bazaar'. You can keep your perfect corporate 'Cathedral' and produce free software, this a matter of license.
The question is not to see if 'free software' developers do it better than non-free one, the question is about choosing a license and understanding its impact.
Now I won't explain here why software should be free, but I can tell you it makes *a lot* of sense for MMORPGs companies. And that free software is not necessarily software developed by long-beared geeks in their spare time, IBM produce a lot of GPL material and they were suits
Disclaimer: I'm _not_ from Nevrax.
[ref: http://www.developer.com/article.php/1548611]
:
Hello,
first thanks for the nice first article. I personnaly believe that software should be free, but claiming it to be open (I'd prefer 'disclosed') is a necessary step forward, at least from a practical point of view. I would like to answer to your follow up : my opinion is that you made some step backwards and basically invalidated some of your stronger points in the original article. So here are my own answers.
Intellectual property
The claim is "Key parts of complex software systems are true intellectual property of the creators, and the owners would never open up that code". I'm wondering what is 'true' IP, and what are the others. You clearly explained that you don't protect your IP by keeping it in your safe, but by relying on copyrights and applicable laws.
There is no such thing as 'corporate secret'. Coca-Cola is well known for keeping is recipe ultra secret, and only a few blessed people are supposed to have access to it. What for ? We can reverse-engineer Coca-Cola's recipe, and have plenty of means to find an industrial process which gives the same drink. It did not stop concurrents either. I wouldn't be surprised if Pepsi actually needed more efforts to find an alternative to Coca-Cola rather than plainly copying it. Coca-Cola currently tries to hide the fact that nothing can't prevent a smart company to produce a clone and sell it. Nothing but Coca-Cola's brute marketing and intimidating force. Oh wait, yes, if Coca-Cola recipe was patented, they could prevent that. But it would be disclosed.
Similarly, there is no such thing as 'crucial algorithm'. There are only smart implementations. Reverse engineering software is extremely easy today. Clever software writers know that, and may elect to patent and thus disclose their algorithm. It's only a smart move if you are cynical : most patents are obscure descriptions vaguely desribing their purpose, and we had plenty of cases and time to realize that algorithms should not be patented (thanks Mr. Jeff Bezos for its spectacularly stupid and demonstrative patents). But I'm disgressing, I only wanted to show that you can't protect your IP until you disclose it.
Now if I get a video encoding tool, and if the source code only shows the UI part because the vendor removed the 'crucial algorithms', I'd be rather disappointed. Be sure that all vendor will consider the core of their software as 'crucial algorithm'. Be sure that the only interesting evaluation of quality can be made on this core part.
I ran in two recent exemplary situations : both Nvidia and 3DLabs disclosed some sources related to their 'shader language compilers', boasting about their transparency and such. It turns out that their code only contain front-ends. A piece of code that you can almost automatically generate from the language specs. The back-end is the core part I would like to see to make my opinion (and fix bugs, and improve, etc). But these corporations don't want that. You end up trusting them with curious metrics (for a developper) : "I'm used to Nvidia products, they usually work well".
International copyright laws
I'm an European (French to be more precise). I'm not sure how I should handle your paragraph. It basically states that countries outside the US have no laws. As a spoker from 'dark warehouses in faraway places', I can testify that we do have laws. Some even prevent people from stealing.
I'm being sarcastic because first I think you deserve it, and second because this kind of ignorance is what is playing against your 'open your software' argument. All countries (including non-democratic ones) have plenty of laws regarding copyrighted materials and IP protection.
I'm disclosing all the software I write. It only takes two steps
- I'm applying a copyright, which says who authored the code and when. This let me claim my author rights anywhere on the globe.
- I'm choosing a license which says how my material should be copied, distributed, and so on. I can do that because I'm the copyright holder. Now I juste have to find a good license which fullfills my need and is known to work in all courts (eg. http://www.gnu.org/licenses/licenses.html), or get a good lawyer write one for me.
This reminds me of the NRA fanatics behaviour : they pretend they need guns to protect themselves before even knowing - or wanting to know - how their country is protecting them and if there's any danger in the first place. This harsh comparison emphasizes two key points : there are laws that protect you (or your IP), and there are not thieves at every corner of the streets.
National Security
I would insist more heavily on this point. National Security is the last domain I would ever like to see closed software. Thinking that my national secret services rely on software which are secret to them is either a joke or a nightmare. And eminent cryptologists have made their point clear since a long time : there's no security through obscurity, only insecurity.
Already inspected
Microsoft claims to closely inspect its software, and even launched a noisy campaign along the lines last year. Are you trusting Microsoft software ? I expect to get a full panel of anwsers from "no way" to "sure".
Objective quality can only be achived via disclosure, and only if the disclosure is permanent. You can't pretend you gave the source code to a 3rd party QA company and they were OK. You must keep the inspection possible for anybody at anytime, else you can't build any trust. ISO QA certicates have provision for this : they are only valid if the contents are open and kept open.
Now subjective quality is another point : many companies have their own QA certification process and will want to do the inspection by themselves. I have seen a bunch of them.
There's no such thing as 'already inspected' software.
And what about making fun with those log entries ? My Apache was used to be hit more often by Nimda and CodeRed rather than regular users, so I decided to make an effort for this unexpected public.