Linked by Thom Holwerda on Mon 16th May 2011 16:15 UTC, submitted by john
Legal This is certainly worth a meagre +1 in my book: patent troll Lodsys has actually taken the time to answer some of the concerns on the web regarding its legal threats to several small-time iOS developers. There's some interesting stuff in there.
Thread beginning with comment 473323
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[3]: Still fsckd up
by pantheraleo on Mon 16th May 2011 23:35 UTC in reply to "RE[2]: Still fsckd up"
pantheraleo
Member since:
2007-03-07

The issue here is software patents, which are generally a bad idea, but this one is obnoxious being so obvious as it is. Or do you seriously believe that without the hard work of Dan Abelow we wouldn't have in-app software update buttons?


You obviously haven't been reading my comments completely because I have mentioned multiple times that I don't think he has a legitimate patent claim. I only said that upgrade systems are often far more complex than "a button with a few lines of code" behind it.

Software patents in general are not a bad idea. Generic software patents are.

For example, if I spend hundreds of thousands of dollars on R&D to develop a super-efficient database algorithm that is blazingly faster than anything that currently exists at retrieving results, I should be able to patent that specific algorithm. But I should not be able to patent "Database retrieval algorithms that go faster". Nor should I be able to patent a generalized idea such as "If you make the algorithm simpler from a machine instruction standpoint, it will go faster" (which is usually true). In the first case, you are perfectly free to come up with your own fast database algorithm. It might even be faster than mine. But you just can't use my specific algorithm without paying me royalties, which will help cover my R&D expense.

The problem with not allowing any software patents, is that it will remove the incentive to spend money on R&D. If I spend hundreds of thousands, or millions of dollars to come up with a super-efficient algorithm, and then all of the other companies can just rip off that algorithm, I can't compete. They will be able to undercut me in price because they didn't have the R&D expenses. They just let me do all the work, spend all the money, and then took the results of my work and used it in their own product. It's only fair that they pay me to help cover my R&D expenses.

Edited 2011-05-16 23:39 UTC

Reply Parent Score: 2

RE[4]: Still fsckd up
by JAlexoid on Tue 17th May 2011 05:14 in reply to "RE[3]: Still fsckd up"
JAlexoid Member since:
2009-05-19

For example, if I spend hundreds of thousands of dollars on R&D to develop a super-efficient database algorithm that is blazingly faster than anything that currently exists at retrieving results, I should be able to patent that specific algorithm.

But you don't need to patent it. Copyright covers that algorithm perfectly well. Because without patenting you don't need to disclose the actual algorithm and reverse engineering it has the same restrictions as implementing a patented algorithm. Seriously, didn't your CS education come with a healthy dose of IP law? Mine came with 2 years of IP law.

Seriously about the DB algorithms. Why hasn't Oracle copied all of IBM DB2 algorithms? Most of the parts that are highly efficient are actually not patented.

The problem with not allowing any software patents, is that it will remove the incentive to spend money on R&D.


That is a not true. That is just used to scare some people in believing. There are a lot of countries that sink a lot of $$$ into R&D, that don't have software patents. The only part that that is true, is the part you don't believe software patents should exist - generic patents.

Reply Parent Score: 2

RE[4]: Still fsckd up
by molnarcs on Tue 17th May 2011 05:37 in reply to "RE[3]: Still fsckd up"
molnarcs Member since:
2005-09-10


Software patents in general are not a bad idea. Generic software patents are.

And who will decide what is generic and what is not?

For example, if I spend hundreds of thousands of dollars on R&D to develop a super-efficient database algorithm that is blazingly faster than anything that currently exists at retrieving results, I should be able to patent that specific algorithm. But I should not be able to patent "Database retrieval algorithms that go faster".


You contradict yourself right there, for software patents, by definition, will always be too generic. You also seem to have problems distinguishing copyright and patent protections. For your specific implementation of a faster database algorithm is already protected by copyright laws. When you decide to patent it as well, you have no choice but to describe in generic terms what you do specifically in your code. From then on, you effectively can prevent alternative implementations of a faster algorithm, ie different code that does essentially the same thing. This stifles innovation, not encourages it, which means that this is also incorrect:


The problem with not allowing any software patents, is that it will remove the incentive to spend money on R&D. If I spend hundreds of thousands, or millions of dollars to come up with a super-efficient algorithm, and then all of the other companies can just rip off that algorithm, I can't compete.


There is not a shred of evidence to support this, however, we have tons of evidence supporting the opposite (including the very patent troll we are discussing here). Software patents are a US only thing. By your logic, the rest of the world should have never invested in software R&D, which is obviously not the case. In fact, all the major innovations in software happened way before patent litigation became a favourite passtime of companies that 1) either don't innovate at all (ie have no products) 2) or companies that are afraid of a little competition. The most important innovations in software happened in spite of software patents, not because of it (example: the basic infrastructure of the internet - the protocols, DNS, mail servers, etc.)

They just let me do all the work, spend all the money, and then took the results of my work and used it in their own product. It's only fair that they pay me to help cover my R&D expenses.


Again, you confuse copyright with patents - I'm against the latter, not the former. Linux, for example, is probably covered by hundreds if not thousands of patents. The situation is so bad, that Linus encourages programmers not to look at patents at all. By now, you can't basically do anything more complex than "hello world" without bumping into software patents. So we ended up in a situation where people who follow the normal evolutionary path of development (linux is a good example, because you can see the whole damn thing evolving in front of your eyes)will inevitably violate patents, without ever looking at your "hard work" - simply because, as I said, software patents will always be, by definition, generic.

And lastly, the product cycle in software development is how long? A year, maybe two. You can't sell software older than that. Eliminate software patents completely (or simply don't implement them) - and you will end up exactly where you are right now as far as technology goes. Why? Because being first to market is far more important in software than in any other field. Being better or just staying ahead of the competition has its own rewards and provides incentive enough to invest in R&D. The only way software patents could foster innovation is if they were severely limited, like 6 months or a year at most (but again, as I said at the beginning of my reply, there isn't a single shred of evidence that supports this theory).

Edited 2011-05-17 05:38 UTC

Reply Parent Score: 2