quote:

---

It turns out that I'm a really naive person. I thought I was arguing in favor of innovation when I was writing columns attacking software patents. It seemed obvious to me that software patents served mainly to stop innovation by real developers, who have come to fear teams of lawyers arguing that the coders' projects had violated some broad patent.

But, according to several e-mails I've received—all from executives at software companies, mind you—I've got the whole thing completely backward. Software patents protect and encourage innovation by letting developers (or at least the vendors they work for) know that they can get every penny coming to them.

What was I thinking? I'd have to be some kind of pinko Commie to think that anyone would ever do anything if there wasn't the potential for a big payday at the end.

---

But I'm skeptical about his assertion that lack of innovation is due to companies being afraid of risk. Innovation becomes harder as a field matures (the easy ideas have already been implemented), and its beginning to take rare creativity and discipline to come up with truly innovative software. There really aren't more geniuses in software than in other fields, despite the image of hacker geniuses.

If you're constantly worried about being sued then why bother? So in that sense, software patents can in fact stifle innovation.

There has been a major movement in the EU against software patents and there were demonstrations outside the European parliament when they dealt with this issue some months ago.

This is an issue that really isn't on the radar screen amongst activists in North America.

quote:

---

This is an issue that really isn't on the radar screen amongst activists in North America.

---

I wouldn't say that. It's all over Slashdot. The EFF and Stallman et al. are seriously engaged. Given the differenence in the state of the law and in political culture the fights are being waged differently (with likely differences in effectiveness as well), but I would not say that it is not on the radar screen. It certainly is amongst prgrammers.

quote:

---

Originally posted by retread:
An interesting column; there is definitely a difference between the need for copywrite (ie so people can't just copy your program - same need as writers and artists have), and patents (which are just an attempt to stop someone from using a concept ... they'd patent the number 2 if they were allowed).

---

quote:

---

In fact, most of the great software innovations seem to have come from individuals or small groups of developers who were creating something solely to solve a problem, to help a group they were part of, or simply out of curiosity and because they could.

---

quote:

---

I wouldn't say that. It's all over Slashdot. The EFF and Stallman et al. are seriously engaged. Given the differenence in the state of the law and in political culture the fights are being waged differently (with likely differences in effectiveness as well),

---

Oh definitely the EFF and Stallman are doing some great work. Its just that at this stage of the game I couldn't imagine a demo being organized in Washington or Ottawa over software issues. So I guess when I was thinking about it "not being on the radar screen", it was from the point of view of there not being that level of a "public" campaign.

quote:

---

His comment about lack of innovation by big software shops is less about patents and more about market risk. Figuring out if a derivative product will sell is a lot easier than figuring out if a innovative product will sell.

---

True.

I recall reading one of John Ralston Saul's books a few years ago (forget which one) where he was commenting on the myth of corporate leaders being great "risk takers"...and therefore deserving of the lifestyle and power that they have in our society.

As I recall he was commenting that in fact these folks are a very conservative bunch and take very few risks at all (they're really just managers). They spend their time swallowing up smaller companies that have developed products and services, corporate mergers, chasing patents on things that are ever so slightly different than existing products ...suing each other etc. ... all very "low risk" activities.

[ 07 October 2005: Message edited by: radiorahim ]

Quick summary for the non geeks amongst us:
The General Public License (GPL) requires that if you deliver GPL software to someone you must provide the source code and you may not prevent them from modifying it and redistributing it.

quote:

---

Originally posted by jrootham:
The General Public License (GPL) requires that if you deliver GPL software to someone you must provide the source code and you may not prevent them from modifying it and redistributing it.

---

quote:

---

The point that I was trying to emphasize is that it is copyright that makes this possible. Without it anyone could just take the code and keep any changes to themselves.

---

That's how I understand the Free BSD license works and its what's underneath the Mac OSX operating system.

They're able to take from Free BSD without putting back.

[ 08 October 2005: Message edited by: radiorahim ]

quote:

---

Originally posted by radiorahim:
I think the article nicely points out that most of the truly "inovative" software developments have come from either the public sector or from the academic community. Very little has come from proprietary software.

---

quote:

---

To live, everyone needs to get paid. Even Linux programmers. So who do you suggest picks up the tab?

---

Some open source programmers are indeed getting paid to work on projects and some aren't. Not everyone out there is working for free.

Sun Microsystems for example is a major contributor to the Open Office project.

The model with some of the more "commercial" versions of Linux is that the software itself is free (or cheap) and they make money providing support and services.

quote:

---

Originally posted by Gir Draxon:
To live, everyone needs to get paid. Even Linux programmers. So who do you suggest picks up the tab?

---

Quibble, there are multiple BSD's (Free, Open, and Net are the big ones). I am not sure which one was used by Apple.

quote:

---

Originally posted by Ron Webb:
Who do you think pays them now? Their employers, of course, and gladly. If your company used Linux as its O/S, wouldn't you love to have a Linux developer on staff?

---

Yes, but that is corporate software development. It's not "public sector" or "academic". There we have the company paying the bills. I was wondering where radiorahim was going with his plan for public funding of software development. That is, people who don't run Linux or even have computers will still be paying development costs.

quote:

---

Originally posted by retread:
The reason copywrite is necessary is because otherwise there's no point in writing most software - its too easy to copy.

---

quote:

---

Originally posted by retread:
But why spend millions developing a better processor if your competitor can directly copy and manufacture it a couple of months after you release it?

---

However, I probably diverge from you when it comes to actual works - for instance, how long should the copywrite on a novel last? I'd argue that the concepts in the novel (ie the fantasy gendre started - perhaps unintentionally - by Tolkien) should be free from day one, but the actual novel itself should be protected for at least a few decades. The same thing with software - no one should be able to patent something like windows, or object oriented programming etc, but a given piece of software should be protected (decades sounds silly in this context because its hard to imagine software lasting that long, but its the same principle).

The problem I see is that there's a tendency to want to patent concepts, rather than implementations. The former is almost never necessary or good for society, the latter is required for workers in the field to make a living. Moreover, patenting implementations never slows innovations - you're always free to make a better implementation. Patenting concepts is a barrier; imagine if someone had patented the idea of communications protocols (it could almost happen under today's laws).

I'm interested in hearing why you think the code should be open for copying ... isn't that like allowing me to publish say Atwood's latest novel in my name after a few years? I'm not sure how it would help creativity to let people copy - I suspect in fact it'd lessen creativity because people wouldn't be forced to come up with their own (and potentially better) implementations.

[ 10 October 2005: Message edited by: retread ]

quote:

---

There we have the company paying the bills. I was wondering where radiorahim was going with his plan for public funding of software development. That is, people who don't run Linux or even have computers will still be paying development costs.

---

The whole open source software movement is working quite well right now.

As RW mentions, some programmers are doing work as a "labour of love", some programmers are working for non-profit foundations like the Mozilla Foundation, others work for companies that are making heavy use of open source software, and in other cases governments ... also heavy users of open source software are contributing as well.

quote:

---

Originally posted by retread:
The problem I see is that there's a tendency to want to patent concepts, rather than implementations. The former is almost never necessary or good for society, the latter is required for workers in the field to make a living. Moreover, patenting implementations never slows innovations - you're always free to make a better implementation. Patenting concepts is a barrier; imagine if someone had patented the idea of communications protocols (it could almost happen under today's laws).

---

I think it depends on a number of issues, like whether the programmer is an independent contractor, or is she/he employed by a company or corporation. A number of scenarios are possible. If you're an independent contractor doing a job for MS, then access to certain layers of the their proprietary code are likely necessary to complete the job.

If you're an IC and say, developing bar code scanning s/w with Windows interface to work with say, OLC's lottery ticket end of things, then you may, probably, more than likely get away with murder as far as borrowing code goes. We might be surprised with the liberties that some professional programmers take in this regard. As long as the end result isn't being integrated into, or becomes a part of a liscenced product and becomes patent protected itself. Designers hate it when someone else attempts to take full legal credit for their stuff. Copy or scab the code and obtaining permission to doso if required tends to flatter the hell out of the designer. Engineers and s/w developers scab code all-the-time. They'd be hamstrung if they couldn't, and things would take a lot longer to do.

Using other people's code is fine (in fact, the use of libraries is vital to both efficiency and good code). But needing more than the interface is a shortcut that often leads to maintainance problems down the road. Again, there's no need for open source. Writing a scanner for windows, all you should need is their API. Okay, I work on a UNIX platform, mainly do hardware design, the coding I've done has mostly been on various realtime operating systems, and don't have experience programming on MS operating systems. With that caveat out of the way, I doubt that using little tricks out of Windows is a good way to go, as opposed to using their published interface.

The desperation with which software companies are pursuing potentially counterproductive actions regarding "intellectual property" is a measure of their realization that any brilliant kid in front of a computer can beat them to the next "killer app", if the killer app isn't too large (there's a physical limitation to how much bug-free code one can write in a year).

quote:

---

Here's an example of the stupidity of patent. Microsoft convinced the USPTO they invented the electromechanical mouse in 1985 when Xerox had been using it since 1978.

---

The question is: Does the world owe it to software companies to keep them alive and give them a fat profit? Is the ability of software companies to siphon a certain share of corporate money in return for providing some of the building blocks for keeping the computing infrastructure working and growing, a moral imperative? Would there be something ethically wrong with a world where most software was effectively owned and enhanced in common by the users (individual, corporate and governmental) rather than bought from software companies?

Well, the argument basically seems to be that fewer programmers would be able to work for a living at software companies, and this would be a hardship. True enough. On the other hand, the reason would be that software was cheaper, and probably more responsive and adapted to local circumstances. So we're talking reduced costs. That probably means that companies would be in a position to either simply be more productive, or to use technology in more extensive and interesting ways, or both. I would suggest that as cheap oil has enabled economic growth, in a high-tech environment so does cheap software. The upshot being that those people would be able to work somewhere, just not necessarily as full-time software programmers.

Personally, I think it's ethically a good thing on balance. I'm not really happy with people like Bill Gates or the Oracle guy sucking huge amounts of money out of the world in return for stuff that ought to be a commodity. Plus: It's the Freedom. Plus: All this stuff about copyrights. OK, so say a programmer works for Adobe and writes big chunks of a program. Who owns the copyright? Not the programmer. So, what's so sacred about legal fictions owning copyrights? The legal fiction didn't have any ideas or use any creativity.

Novels are, by the way, quite a different case. Anyone interested in the ramifications of copyright and so on in different sorts of intellectual creations would be well advised to go to gnu.org and look at some of the very thoughtful analyses of the subject. Richard Stallman opines that novels, poems and so forth, as well as most sorts of scholarly research article, should not be GPLed because alterations to such things are very problematic--they can change and hence misrepresent the author's intent or even scientific findings. He considers most programs to be more akin to recipes, which few people have any qualms about sharing, altering, improving, adjusting to be usable by vegans, et cetera.

I'm not too directly involved in the discussion, because most of what I do is hardware design. However, its analogy would be to remove all patents (again a fairly common argument), which would directly effect me. I'm left unmoved by references to Bill Gates' billions, simply because an overwhelming majority of programmers are middle class wage owners. People like Gates won't be effected, he's rich enough that he can absorb anything. People like my co-workers will be. Its like saying there's no reason to pay to see local theatre because Tom Cruise makes ten million dollars a film.

quote:

---

Originally posted by jrootham:
Not necessarily. The mouse was not invented by Xerox but by Doug Englebart. Xerox may simply have respected the idea that it was an unpatented invention.

I haven't seen the patent in question. Is it for mice in general or a specific type of mouse?

---

As I said, the electromechanical mouse. One that's still in use today and was in use in the 1970s. I used such a mouse on my Apple //c in 1985, incidentally.

The point is, US patent law is allegedly first invention priority so that Xerox or Apple should have been able to bring suit and claim Microsoft stole their patent.

quote:

---

Whoever patents first, licences first and gets a free ticket to print money.

---

Under normal circumstances, wouldn't the inventor patent first?

quote:

---

I've no issue with open source programming in itself, just with removing the option of copywriting work.

---

Hmm. It's not completely clear that you have a complete grasp of the subtleties here. I know of no serious players in the Open Source arena who are arguing for the removal of copyright (note spelling, it has some semantic content). As I pointed out above all the open source licences depend on copyright for enforcement.

There is also the confusion between free as in beer and free as in speech. It is not entirely trivial to deny the first in the presence of the latter, but it is possible.

Stallman is neither an academic nor independently wealthy. He may make his living writing and speaking now, but he certainly did make his living by charging people substantial fees to write (and fix) software under the GNU licence.

quote:

---

The mouse was not invented by Xerox but by Doug Englebart

---

What has Microsoft done with this patent? If they are not charging license fees to Xerox or Apple there is no point in taking them to court.

quote:

---

Magoo: what part of:

------------------------
The mouse was not invented by Xerox but by Doug Englebart
------------------------

did you not understand?

---

quote:

---

Xerox just didn't bother. Duh to Xerox on that one.

---

Patents are not necessarily useful to individual inventors. To enforce a patent involves going to court. You need deep pockets and/or considerable determination to spend the time and money.

Anyways I just Googled and:

quote:

---

1968 Computer mouse makes its public debut

The computer mouse makes its public debut during a demonstration at a computer conference in San Francisco. Its inventor, Douglas Engelbart of the Stanford Research Institute, also demonstrates other user-friendly technologies such as hypermedia with object linking and addressing. Engelbart receives a patent for the mouse 2 years later.

---

And I couldn't find a Microsoft patent for the mouse in 1985.

Now, it could well have been thrown out in court later on, but I suspect that Microsoft's rip-off of Apple exceeded what they had to give back if there was a lawsuit.

quote:

---

Microsoft got the jump on Apple and forced Apple to pay licencing fees.

---

Probably true, just not for the mouse.

It's copyright not copywrite and the issue was trademarking "windows" not copyrighting it.

quote:

---

Originally posted by retread:
I've no issue with open source programming in itself, just with removing the option of copywriting work.

---

Flat removal is not a workable option. But there is quite a bit of question just what copyright should be like, and whether it should be a "one size fits all" or whether different kinds of works should be treated differently, and how long it should last, and what rights should and should not be involved, and for whom.
Incidentally, I think that while it's trivially true that the GPL depends on copyright, at the same time if there were no copyright law for computer programs (or more precisely, if there were a law regarding programs which granted relevant rights to the public, thus avoiding EULAs and similar contract-based end-runs), then the GPL would not be necessary. Whether that's desirable or not, it's not really the case that the spirit of the GPL would be crushed by a no-copyright legal regime.

quote:

---

The reference to a recipe trivializes it; something that takes tens of people two years to do is more of an effort than a recipe. The sustained effort is closer to a novel or a textbook than a recipe (but I'll read the site to get their argument).

---

It's more a matter of function than length. The point is that a recipe is something designed to produce a particular outcome. It's not a work of art, nor is it a factual claim or a structured argument. So it's useful to be able to make changes if you find that either you have a way of making it give a better version of the desired outcome, or you have a slightly different outcome in mind. So, say you have a recipe for chocolate pate with a mango coulis. If it calls for chocolate chips, you might well want to adjust it and instead put in a good bittersweet chocolate, making it a better chocolate pate. Or, if you hate mangoes, you might want to make it a raspberry coulis instead. Programs are similar; if there are bits that are buggy, or not user friendly, or if you think it would be useful if it had a couple of extra functions that you need, it's handy to be able to make changes. In both cases, it's somewhat beside the point that you might be interfering with the artistic vision of the original author.
Design specifications, architectural blueprints and so on and so forth have similar characteristics.

Novels aren't comparable in this way. Someone wrote a novel to be the novel they wrote. They arguably wrote it as some kind of artistic unity. Changing the ending because it depressed you is problematic. Poems, even short ones, are more like novels; so is art. For this reason among others, Stallman is actually prepared to concede that computer games are probably not workable candidates for GPL-like licenses; they contain too much original art, as a rule (although game *engines* are another story).

Political polemics and research papers are even more problematic. You can't just let people take your research paper, change the data and the conclusions, and redistribute it. There has to be some kind of rule about that. It's easy to make an argument that people should be able to freely redistribute research papers, retaining the attribution (i.e. not claiming to have written it oneself). I'm not claiming it's a conclusive argument, but it can certainly be made and is fairly robust. But it's pretty clear that even if you bought that argument, such redistribution rights should *not* include rights to make alterations.

quote:

---

I'm left unmoved by references to Bill Gates' billions, simply because an overwhelming majority of programmers are middle class wage owners.

---

quote:

---

People like Gates won't be effected, he's rich enough that he can absorb anything. People like my co-workers will be.

---

But you're back to the question, basically, of whether the world owes you a living--and not just *a* living, but the particular living you have chosen. Does the world owe sofware-house programmers a living writing sofware, even if the economy in general will be better off without those software houses? At this juncture, I'd like to ask you--do you consider the losses of jobs in the vacuum tube business caused by the rise of transistors a bad thing? How about the massive layoffs that corporations regularly indulge in for "efficiency"? How about movement of jobs to sweatshops in third world countries, allowing you to buy cheaper goods? How about Wal-Mart?
Now, if you've been a stalwart backer of labour in all its struggles against various forms of offshoring, downsizing, and unionbusting then you have a leg to stand on in making this argument. Otherwise, you don't. But even if you do, the case is more like the vacuum tube one.

quote:

---

Its like saying there's no reason to pay to see local theatre because Tom Cruise makes ten million dollars a film.

---

No, it's not remotely analogous. If Tom Cruise owned the local theatre company and was taking half the receipts off the top, perhaps. But really, Tom Cruise is a bit of a red herring. The point is that the Gates-types own the copyrights and rake off much of the cash, and they're not analogous to Tom Cruise--more like producers. The programmers at their companies don't make any Tom Cruise salaries, either.

quote:

---

In passing, I note that most advocates for removing copywrite are either independently wealthy, or working on salary in academia

---

quote:

---

Originally posted by retread:
I'm interested in hearing why you think the code should be open for copying ... isn't that like allowing me to publish say Atwood's latest novel in my name after a few years? I'm not sure how it would help creativity to let people copy - I suspect in fact it'd lessen creativity because people wouldn't be forced to come up with their own (and potentially better) implementations.

---

Believe me, clever programmers will always be on the lookout for better algorithms, not for the money, but for the sheer fun of it. (Not to mention the bragging rights.) And corporations will always be on the lookout for employees who clearly demonstrate their understanding of the algorithms too. Who better to supervise your Linux systems than the guy who wrote part of it?

Again, let me emphasize that I'm not totally opposed to limited copyright protection. I think the guy who comes up with a better TCP/IP stack (or the corporation that employs him) ought to have a chance to make a fair profit on it. But how much is fair? I don't think "the maximum possible profit" is necessarily the right answer. There are other considerations besides profit. For instance, at what point does fair return for a clever idea start looking more like monopolistic control over a de facto standard?

quote:

---

Ever since Autodesk had to pay $25,000 to ``license'' a patent which claimed the invention of XOR-draw for screen cursors (the patent was filed years after everybody in computer graphics was already using that trick), at the risk of delaying or cancelling our Initial Public Offering in 1985, I've been convinced that software patents are not only a terrible idea, but one of the principal threats to the software industry.

---

[ 26 October 2005: Message edited by: DrConway ]