Miguel de Icaza's answer to "Does Windows cost Microsoft opportunities?" | Hacker News Reader

Miguel de Icaza's answer to "Does Windows cost Microsoft opportunities?"

tirania.org

31 pointsDrJokepu36 comments

I wish this had come earlier in the article:

"Microsoft has new management, new employees that know open source, fresh new ideas, is becoming more open and is working actively on interoperability with third parties. They even launched the CodePlex Foundation"

It would have spared me reading it.

The only reason Microsoft wants to know open-source is to more effectively kill it. They have realized that just bad-mouthing it, calling it a virus, wouldn't do the trick. Now they use a far more sophisticated approach, precise wording (using "non-commercial" instead of "free" or "open-source") and passing a confusing message regarding patents (the "community promise" while announcing patent deals - whose terms are secret - with companies using Linux in their devices).

No. They haven't changed. Not a bit.

Exactly. I still can't understand why somebody as smart as Miguel doesn't get this. Any free software strategy that depends on the goodwill and good conduct of MS is doomed from the start.

Miguel never ceases to amaze me on that front. Given he's not a stupid person I have reduced it down to a psychological explanation regarding his early days - getting denied after an MS interview and it turning into an obsession to prove his worth to them or something. But even that feels weak to me even if it is plausible. But weak is better than nothing, because there seems to be no other reason to be so credulous about their good intentions.

How about this then: MS does create some nice technologies (yes - .NET's design is quite nice compared to Java). Why not simply use them even if MS spreads some patent FUD. Not everyone believes in those claims and not everyone lives in US.

So someone smart does something you don't understand, and you decide it must be a psychological problem?

I said he wasn't stupid, I didn't say he was smart. I do think he is smart in the technical domain. I am not so sure about elsewhere. And "psychological problem" isn't what I said either, I just implied the explanation perhaps lay in his inability to get to Microsoft and then desiring their approval. Tons of people have long lost dreams that they romanticize. For example a lot of programmers, despite the horrific stories that come out of the games industry, still regret not going in that direction. You can see how someone who built up MS in their youth and didn't get there might also have such a longing and ignore the bad stories that come out. I hope I have clarified the not stupid/smart and the psychological explanation/problem differences between the wording of my post and yours for you.

Miguel de Icaza is the foremost expert in the CLR that doesn't work for Microsoft. If he wanted to work for Microsoft in any capacity, I would bet anything that they would roll out the red carpet for him. You're either projecting or you don't know what you're talking about.

He wanted to work for Microsoft, but they didn't employ him (to be fair, he was not US citizen and he didn't fulfil the H1-B requirements).

After that, he started Gnome (complete with the CORBA-fiasco) and later Mono.

> Given he's not a stupid person

Your explanations assume good intentions. I wouldn't go as far.

The most intelligent persons are often the best at fooling and deluding themselves.

Once a path has been taken, it is very hard for an intelligent person to admit it might have been a mistake, it is too easy to rationalize and justify why you are right than to admit you might have been wrong.

So how do you explain Jim Jagielski joining the board of the CodePlex foundation? Does he also not get it?

There comes a point, when you see many smart people doing something that you think is a mistake, that it is time to ask if maybe they know something you don't.

> There comes a point, when you see many smart people doing something that you think is a mistake ...

This is when you ask : How much are they being paid ?

Are you saying that we should be doing the same thing as the people we think are "smart" without knowing why they are doing these things?

Is that what a smart person does? Or are you asking us to accept that we're the dumb masses?

What is it then? If Miguel is so prescient why is he now disappointed at the course MS has taken with .NET over the last 10 years, particularly when so many supposedly less intelligent people predicted this at the very beginning of the Mono project?

I'm not honestly sure they have the wherewithal to do so anymore.

They're not exactly in their acendency at the moment. More and more people are buying macs, linux servers are often just the assumed technology (they had to release a cloud competitor, Azure, to even stand a chance), I mean, they may be doing this OSS push because they think they have to.

Is that really the case? I'm not saying they're not trying to kill some popular open-source projects, but I've also seen a lot of cases where they did publish some good research cases out of goodwill.

Notable examples include (just stuff I already knew about - I'm sure there's lots more in areas I'm not interested in):

- all work by http://research.microsoft.com/en-us/people/simonpj/

- Haskell ThreadScope: http://research.microsoft.com/en-us/projects/threadscope/

- Singularity, process isolation and language research: http://research.microsoft.com/en-us/projects/singularity/

- Hyper-V patches for the kernel (ok - they were ugly, etc. - but they were submitted in a working state, so they tried at least)

I really wouldn't be surprised if a lot of bad moves coming from MS were just caused by the legal department not understanding the FOSS issues and writing a new license that is "good enough" for them, then forcing the developers to use it (why would a developer be allowed to question their licensing decision?). There are loads of other projects and papers published by MS which are of a great value to the FOSS community - sometimes they just carry a stupid license. That can be solved by reading the documentation and writing a clean-room implementation where needed.

I definitely disagree that any free software strategy connected to MS is doomed.

It's one thing to recognize that MS may have made a few genuine contributions to free software. It's QUITE another to build the entire free software desktop and server stack on a technology that MS sees as the core of its business, as Miguel has been advocating for years.

The fact that he's disappointed that MS hasn't allowed the free implementations of .NET to compete on an even footing just shows how naive he is.

I'm not sure what options you leave him. When he is disappointed, he's wrong. If he wasn't disappointed, it would be even worse. Why not accept what's happening and hope for a change in the future?

If MS starts executing their patent rights, apparently the whole Linux is illegal in US (they said so!). Operating systems are also the core of MS business. According to what you say, we're also naïve to use Linux. Is that right?

Obviously it would be a lot easier for MS to pursue patent litigation on technology invented entirely in-house at MS (like .NET) than it would be for them to try to argue that Linux itself infringes on their patents. Surely you can see the difference.

TBH no, I don't. For any patent lawsuit you'd have to prove that the technology was invented entirely in-house or in some other way that gives them the right to sue. It doesn't matter whether it's been published before, or not. Didn't stop them from suing TomTom because of FAT, right? It's not Linux that's patented - it's just a number of technologies which are included in Linux. Userland (like .NET libraries) or kernel (like FAT) doesn't make a big difference here.

Not having any reason to sue doesn't stop people either - see SCO fiasco. I do believe that MS can sue the majority of companies without a reason - and win.

As the article states, and Miguel grudgingly admits, the threat of litigation has had a chilling effect on the growth of .NET. So perhaps it doesn't worry you, but obviously plenty of other smart people have weighed their chances and chosen to invest their money and energy instead in technologies like LAMP and Java, just as a lot of us predicted at the outset.

It doesn't worry these people either: http://mono-project.com/Companies_Using_Mono We didn't get the chance to compare the current world with the imaginary "patent free .net" world. So far it's just talk - who knows how many companies are really worried and how many just prefer other VMs for other reasons. All we hear right now is FUD from MS and FOSS trolls. Some people using Mono just stay quiet and enjoy the possibilities.

Sure - I obviously prefer one side of this argument and I may be completely wrong here :) As long as we're aware of all the potential problems and take time to consider options - let's use whatever meets the requirements. My view is that there are bigger threats from MS than lawsuits over Mono and all equally unlikely to affect me in any way. I guess I'll EOT here...

Wouldn't the Windows EULA protect these companies and their users from patent lawsuits? I assume they run Windows in addition to Mono.

> For any patent lawsuit you'd have to prove that the technology was invented entirely in-house

No. That's no defense against a patent lawsuit. Not knowing the patented technology existed may only prove you didn't infringe on purpose.

> Notable examples include (just stuff I already knew about - I'm sure there's lots more in areas I'm not interested in):

Has Microsoft given up on patents related to these technologies?

Because if they didn't, any free software built on them is under threat of patent litigation any time Microsoft feels like it.

There is no other reason for Microsoft to invest in basic research than to build up a patent portfolio that can be used offensively against companies and projects that threaten its market position.

If you came up with the same ideas on your own and built your product based on them, you'd end up in the same situation. You might not even know about it. If you're small enough, they won't care. If you're big enough, someone will sue you whether they have a reason or not. You cannot possibly avoid all patents and are not protected from patents which don't exist yet, but will in the future (you'll be killed by the lawsuit costs anyways).

If you have a problem to solve, there are only so many ways to solve it. If it's an interesting problem you can assume there's a patent covering it already.

What exactly do you gain by evading MS's patents? Is there a reason to worry about patents at all?

In patent infringement, they will sue your users. Since there is nothing like the EULA in free software, the publisher of the product is not liable for your use of it.

Frequently, the threat of patent lawsuits is enough to turn potential users away, specially those with deep pockets.

As far as I know, Singularity is not open-source. Which is too bad, because it looks interesting and could benefit from more community involvement.

It is open-source http://singularity.codeplex.com/ But is not properly FOSS. Currently it's available with the Microsoft Research License Agreement. So if you're interested / involved in any managed OS research, I wouldn't recommend looking at it. However, you might be interested in the bootable images.

Source available does not mean open source. The term "open-source" has a very specific definition.

Singularity is not open-source, though its source is available. So was the Rotor, for example.

I shouldn't have to spell this out but "The Singularity RDK is for academic non-commercial use" means it is not open source.

I would call it "toxic-source" instead of "open-source".

If they ever decide to sue you for patent infringement over your singularity-like kernel, the fact you had access to their source code could be used to imply you infringed on purpose.

I advise anyone who is involved in any open-source project to stay as far away as possible from any Microsoft-originated code and to warn other developers of these risks.

There are many schools that have embraced the study of Microsoft code that are generating a steady flow of "land mines" who could end up employed in FOSS projects and who could conceivably be used as weapons in IP lawsuits.

Call me paranoid, but being paranoid doesn't automatically make me wrong. ;-)

The overall position of Icazas over the years makes me even less willing to trust Mono. When someone says "I can't imagine any problems at all with Microsoft", I look for nearest exit.

C# is interesting technology but I refuse to use it.

Please show me where de Icaza has said that phrase. It's certainly not in this article, and yet you reference it as if you're quoting him.

Sorry if you read it that way. I imagined that most people would read my quotes as a paraphrase, as describing the gist of what Icaza said.

> I still believe that Microsoft lost a great opportunity of having .NET become the universal runtime of the net

Phew, I'm sure glad we dodged that bullet. Imagine that !

Reply on news.ycombinator.com