What's behind Microsoft's fall from dominance?

[Fats]

Android has really left a bad taste in my mouth, something I've never experienced with Microsoft. Microsoft force the operators to accept updates (I think apple do too).

Me personally, I don't have problem with my cheap Samsung still being on Android 2.3 release. It runs the apps I need.

Androids Linux origins really show through and it's not something that works commercially.

Yeah, that's why Android is on 75% of new smart phones.

[Fats]

My God!
I really don't like this guy, but I wholly agree with him here.

Microsoft is here to stay and our OS' are relegated to hobbyist pursuits.

I also think Microsoft will be here for a long time but it's also clear iOS and Android are currently in the driver's seat and Windows (Phone) needs to need run hard to catch up.

[Fats]

Linux users tend to think that everything is owed to them, IMO they are worse than people who pirate windows.

I, as with probably a lot of other open source proponents, find it an abomination of the patent system that if you own a legal DVD drive and a legally bought DVD it is still illegal to program an open source DVD player.
Important difference.

[Fats]

What you're suggesting is that no patents should be valid because if you buy the parts to make something legally then anything you make shouldn't be able to violate a patent.

No, what I am saying is that the right to play back a DVD should be included when buy the DVD and playback should not be restrictred by software patents. Software patents are an abomination of the patent system.

[Fats]

Software feels different because once you've made it once the cost to reproduce it are lower. But someone fitting algorithms together and designing structures has a right to patent their work, whether that is created in software or hardware.

Software is different and that's why you have copyright for it. Software patents are an abomination of the patent system.

[Fats]

The issue there is that you can have a patented piece of hardware that you can then duplicate using software without having to pay a patent license fee.

Could you give an example of such a hardware patent that could be replaced fully with pure software ?

[Fats]

A very valid point and an important distinction, Staf.
Patents just stifle progress.

I do think certain industries like pharmaceutics needs protection from cloning for a certain amount of time to earn back the research. I don't count software as one of those industries though. There patents are more used as protection of the monopolies/duopolies etc. against competition than it is to protect real innovative inventors from the big guys.

[Fats]

Forgive me for not diving into all claims in detail.

Will this do?
 
http://www.google.com/patents/US5093831

No, the fast hardware implementation part of the patent seems to an essential part of the patent so it should not apply to pure software solutions.

This is part hardware and part software
 
https://www.google.com/patents/US5926786

If I understand correctly they did patent here a design of an ASIC. Which I think should never have been patented as not passing the "non-triviality for somebody skilled enough" requirement. Although hard to judge so long after the fact, I don't see anything in the patent where another design team with the same specs couldn't have come with the same solution.

Here is a bad circuit patent.
 
http://www.faqs.org/patents/imgfull/20110025518_01

Sure, but I don't see how a bad circuit patent would make software patents any better.

[Fats]

So if I design a CPU that has a couple of op-codes that are carefully designed to implement that functionality then because it uses software then it works round the patent?

No the hardware in the CPU falls under the patent not the software using the hardware; the people making the CPU have to license the patent. Or would you claim that one can't write software for this CPU without taking a license on the patent ?

With all patents it's possible for multiple people to come up with the solution, the patent system rewards the person who files first. It's assumed they are the ones that put the most effort and money into it.

Effort and money is not related to patentability; see this for one of the requirements. A patent has to be more than some (tough) engineering and being first. And yes, I do think this rule is not applied strict enough. Problem is that there is not enough incentive for an examiner to be very strict, agreeing with a patent proposal is easier than back-and-forth bickering.

Let me tell you I went through this mumbo-jumbo a few times and am listed as inventor on a few patent (the rights are with my employer though). But this only enforced my belief something is terribly wrong in the current patent system.

Because people talk as if software patents are the only bad ones out there, just because you can find a bad software patent doesn't mean that they are all bad.

As said before, for software copyright is enough to drive innovation and lack of patents won't stop people writing software as they can't earn back their investment. In software, patents are hampering innovations and favoring the behemoths. Even very smart programmers don't need or deserve a 20 year monopoly on their ideas in the fast-paced software world.