Lambda the Ultimate

Thanks. I won't think anymore posting such a question is harmless.

Sure, in a court of law. But in an online forum?

I don't know the specifics of what's in their contracts, but that part of my comment is probably detracting from the important part of the discussion. Perhaps people could check their contracts, and simply comment that they are not allowed to comment?

Reading the patent also exposes you and your employer to the risk of treble damages (in the U.S.) for willful violation. This is one of those situations, like not knowing the combination of the safe in a store where you work, where ignorance is strategically better than knowledge.

I would think posting on a forum that you have deliberately chosen not to read a patent that from its title covers something you are working on might also not be the best strategy for avoiding a finding of "willful violation", but IANAL.

I am not a layer either, but how can you wilfully violate something when you don't know what it is? The idea itself is not patentable, so it is only their method that is protected. If one has not read the method, how could one copy it? If it were to turn out that my independently developed method were the same it would be purely accidental. Further one could say that independent invention means it is obvious (to one skilled in the art) and therefore an invalid patent in the first place.

If I had not said I had not read it, then it might be assumed that as a member of the forum I had the opportunity to read it. In a case of the balance of evidence, contemporary evidence with a consistent timeline is more valuable than a statement at a later date. In my experience it is better to set out your position in return at the earliest opportunity.

Willingful violation incurs treble damages in the states; anyways, I won't read patents.

There are some things people only talk about over beer.

I'm curious in discussing the language design proposed. Thanks for the warnings — my own research seems different enough to be quite safe. But I won't be reading further on the patent, and refrain from similar postings in the future.

That said, I'm not sure of your technical argument (though that might be because you had to avoid looking at the details).

Explicitly specifying immutability is the same as explicitly specifying mutability. Hence an ML ref or Haskell ST monad is a valid mechanism for enforcing this with clear prior art (in the sense that is explicitly marks and safely transmits mutability).

I'm not sure on either statement. For Haskell, I'm not sure if you refer to STRef (which is the analogue of ML refs) or to the ST monad itself somehow; I'll talk about STRef.
1) A transformation from a language with explicit mutability to a language with explicit immutability is bijective, yes, but it does not keep the same annotation overhead, nor is the number of produced annotation bounded in the number of consumed annotations.
2) Mutability and immutability are also not duals for other reasons — most language designers are partial to immutability hence try hard to make designs asymmetric. For instance, ML refs/Haskell STRef require one more indirection for mutability — see below for why it matters.
3) More importantly, if you get normally mutable data through a value of immutable type, it becomes immutable as well — that's what I meant with transitive immutability.
Translating this to mutability, once you access an immutable value through a ML ref/Haskell STRef, you could in fact mutate it without restriction, as if all members were cells and not direct pointers. I think this is closer to Lisp's set! than to ML/Haskell mutation, which requires pointers to have an extra level of boxing, but I'm not expert on either Lisp or ML.
Back to immutability in this context, fetching an immutable pointer would require "inlining" STRefs in the pointed data.

As far as I can tell, this work is a direct transcription of explanations I made to people at Microsoft about BitC immutability. The particular approach to immutability that they have applied to patent has been discussed multiple times on the BitC programming language list, not as something that we had any doubts or difficulties with, but as something that we took for granted as obviously doable. We even discussed the escape analysis necessary to ensure constructor safety.

This is the second case of a patent application by someone on the Midori group since my departure that appears to be directly based on my own work and/or my discussions at Microsoft regarding my prior work and the prior work of others. I do know that Duffy has done a bunch of work on resolving some issues in C# that preclude proper implementation of readonly, but he is not the inventor of either the notion of transitive immutability in a programming language nor of immutability permitting construction. Both were discussed on the BitC list prior to my arrival at Microsoft, including a much more general form of immutable initialization relying on confinement. I discussed this work with Duffy while I was at Microsoft. Heck, I've done a Ph.D. dissertation and two formal verification efforts on this topic; it's hardly new.

Given this, I think there is a problem insofar as the patent fails to credit an inventor at Microsoft (me), but more to the point, the patent application fails to cite prior art about which they knew: BitC, EROS, and KeyKOS. Given the prior work, I would have refused to support this filing on the basis of prior art.

Many people at Microsoft hold the view that outsiders can't do anything interesting. They are, in my experience, appallingly unaware of prior art in the field, and sincerely believe that they have invented many things that have existed for some time.

read-only meaning "transitively read-only" was in KeyKOS ("sense keys" iirc) when I used it in the 80's; given how central that concept (along with transitive effective immutability via COW nodes) was to the security story, I'd imagine it had been in GNOSIS even before then.

Transitive read only was certainly in KeyKOS and GNOSIS, but the patent is dealing with a different set of issues.

To be clear: during the brief time I was at Microsoft, Duffy was investigating questions that would have led him in this direction, which is how we came to be discussing it. It's very possible that he simply forgot about our discussions and reinvented the solution. And if he has a solution that works in a future .Net framework, then more power to him.

It's the patenting of prior art that I object to.

Thanks a lot for your answer. I quoted it in the parent discussion, hope I didn't misunderstand anything:

http://patents.stackexchange.com/a/10214/6081

You are welcome to contribute there directly.

I looked at your profile, but your homepage link is 404 — the website mentions Jonathan Shapiro:

http://www.eros-os.org/~shap

Of course you are entitled to be anonymous, but I'm sure people seeing this discussion and trying to invalidate the patent would be happy to know who you are if you don't mind.

I'm also sorry to hear your comments about Microsoft — given both Meijer and Peyton Jones work there, I'd have hoped for better (not that I generally like Microsoft in any sort of way).

that is all. :-)

MSR files patents also; I have my own personal opinion about this, but I can't share it.

Note that Meijer left Microsoft a couple of years ago. I don't think he ever worked for MSR (he was associated with VB).

Microsoft is also a big company with a lot of diverse behavior like any big company (it is rumored that Google files patents also).

it is rumored that Google files patents also

Indeed, but it was the first time I heard about patenting *programming language features* -- although on further inspection it appears MS has been doing that for a few years already. Does Google also patent language features?

My understanding of the patent situation was that software isn't patentable, but lawyers find hole by masquerading software as parts of mechanical devices. Patenting programming languages feels like a whole new level of absurdity.

I find this both depressing (they're coming at our doors!) and amusing in some ways: it's official, the industry cares about programming languages.

Only sound type systems should be patentable.

Aren't unsound type systems the new hip thing?

I thought patents are there to encourage innovation...

That's a common misconception. Encouraging innovation is the excuse for patents. Their primary function is to stifle competition. One of the things to keep an eye on in an economy is the fraction of resources that go into maniplating the legal system for advantage rather than making better products. When that ratio gets too high, the economy is in deep doodoo.

If you haven't noticed the sarcasm there's nothing more I can do for you.

Who, me?

Whomever failed to appreciate the ellipsis at the end of my previous message.

I dread to think what would happen if we had gangs of unemployed lawyers wandering about.

Unfortunately, software is patentable. The status of this changed quite a long time ago.

Not in the UK or Europe. although there appear to be some ways around it. See: http://en.wikipedia.org/wiki/Software_patents_under_United_Kingdom_patent_law

Unfortunately, no relevant player can afford to ignore the American market, so that does not make much of a practical difference. Globalisation means that you always get the union of regional dysfunctionality.

PS: Has the UK already become not Europe? :)

I migrated that server a couple of weeks ago and forgot to check the web server config. It's serving again now; thanks for letting me know.