No one has ever explained to me just why a company wants to Tivoize Linux?

The explanation B2B free software vendors gave to the FSF was apparently good enough that the FSF restricted the anti-tivoization provisions of the GPLv3, in its final form, to what amounts to consumer products.

B2C vendors weren't as concerned, as long as the license explicitly allowed two things (that it does): termination of any support responsibility and disconnection of modified-software devices from networks when the network owner (even when it is also the device supplier) doesn't like what the modified software does.

How does keeping me from changing the code on a device I've purchased help the manufacturer's bottom line?

Some parents might prefer products that they would use in their homes to not have published mechanisms available that would be accessible to parties other than the original vendor for replacing the manufacturer-supplied software, but might still prefer that the vendor software updates could be applied. The GPLv3 either requires the provision of software installation information for consumer products, or that the product not be updateable at all, which means that, to the extent that this preference exists, it can only be met with non-GPLv3 software.

I've never read about the FSF making any distinctions b/w business and consumer users

See the GPLv3, Sec. 6. Note, there, that the "anti-tivoization" provisions only apply to what the GPLv3 calls "User Products", which are, essentially, what would in normal parlance be consumer products as opposed to business products.

All they care about is the liberation of software.

That's what the advertisements say, but that's not what the license says. Which is the problem with the license.

But yeah, "background music", either via speaker or headphone, is otherwise usually about as conducive to productivity as leaving a television on within visual range. I think the reasons that we keep answering those surveys otherwise, is really that on at least a subconscious level we want to be paid for listening to music we like.

I think the reason is that, in fact, they do improve productivity in many real world work environments. Not because "listening to music" improves productivity (which, as TFA notes, it doesn't), but because it decreases productivity less than the office chatter that it often serves to mask.

Effective active noise-cancelling headphones without music would be even better, but active noise cancelling headphones that work to shut out distractions when they aren't being used to play music (etc.) are more expensive.

I really do not see what Linus's problem is with the GPL3.

I don't know about Linus, but my problems include (but are not limited to) the fact that it restricts available software functionality and imposes restrictions on the use (without distribution) of software, contrary to the whole point of free software, and that it creates difference classes of users, with different rules that apply when targetting them.

The first is "Tivoization". Vendors should not hardwire checks to prevent "unapproved" software from being run.

I disagree, and so (in a different way) does the FSF, apparently. I disagree in that I think this rule is entirely unnecessary, the FSF appears to think that vendors of business hardware should may legitimately need to do this, but vendors of consumer hardware should not. Personally, I think that software freedom is enough, and hardware restrictions of the type imposed by the GPLv3 anti-tivoization provisions on "consumer" devices are not only unnecessary, but actually antithetical to software freedom. This is particularly the case because a mechanism for addressing the same concerns that is consistent with software freedom is suggested by the text of the GPL -- just require that hardware with which GPL software is distributed have specifications that are available under a particular set of license terms, and which is available on the terms as the software source.

Such a check really is software no matter that it's been hardwired in.

So? Since when does software freedom involve limiting the available functionality of software?

With a scheme like that in place, you can't fix so much as a typo let alone a simple bug.

Sure you can. You may not be able to deploy it on the original hardware, but that's a different issue.

Linus is apparently okay with Tivoization.

So is the FSF, so long as the user who wants the freedom to select a device with that feature is part of the privileged class of "business" users. Consumers who might prefer tamper-proof devices, and the manufacturers who want to sell to them, are locked out by the GPLv2.

It's simple: Add rules to any position of public office restricting the revolving door of private industry. Make those in power commit to avoid working for those they are regulating or accepting "contributions" from them.

If the best paying positions by a wide margin for people with knowledge in the field are in private industry, then -- while this might have some effect on regulatory-capture-through-career-planning -- it would also assure that few competent people would ever enter the regulatory agency in the first place. The imbalance of resources and skill between private industry actors with lots of profit on the line is itself a source of effective regulatory capture, as regulators are dependent on information.

Equally importantly, ethics rules governing regulators don't address the problem of industry influence on the chief executives and legislators who define the boundaries of regulatory authority and the funding of regulators and control the composition of the courts to which regulatory decisions are appealed.

But class action lawsuits never deliver anything of real value to the people who actually suffered from whatever prompted the class action suit.

Sure they do, they provide notification of a potential cause of action that they are quite likely to have overlooked on their own. Potential class members are provided notice and an opportunity to opt-out and preserve the right to file individual, direct-action lawsuits.

Class actions are far from perfect, but they exist to deal with an economy of justice problem with diffuse harms (cases where they are available overlap, to an extent, with cases where regulatory agencies and states attorneys-general have the right to file claims on behalf of the public or affected specific citizens, but regulatory capture and other effects of corporate influence on government make them necessary.)

The reason corporations want to foreclose them is because they are often the only effective method of addressing harms, especially where the harm to each individual is fairly small. Mass direct actions have a lot more overhead to coordinate than class actions, and individual direct actions require essentially each plaintiff to bear the full cost of an action to prove the same points. This makes direct action, particularly individual direct action, impractical where the individual harm isn't very large but lots of people are affected.

Attempts to use an EULA to foreclose class action lawsuits ought to be void as contrary to public policy, since class actions are provided as a mechanism for economy of justice.

And even .Net is being deprecated by Microsoft in favour of the Tablet Formfactor Metro apps.

.NET isn't being deprecated in favor of Metro. It may be the case that, for client UI for non-web applications, Microsoft is favoring Metro-style apps over WPF or Windows Forms. But .NET isn't a client UI library (it has WPF, WinFomrs, and even Metro libraries), and preference for one UI-style over another is pretty much orthogonal to the use of .NET.

"This is contrary to law and sound economic policy" means ( "This is contrary to law" ) AND ( "This is sound economic policy" )

No, it doesn't. That would be written, "This is contrary to law and is sound economic policy." (More likely, it would be written, "This is contrary to law but is sound economic policy.")

When written correctly, with the Oxford Comma in place, it would have the intended meaning

There is no place for the serial (also known as "Oxford" or "Harvard") comma in that sentence, since where it is used at all, it is used in separating the final item in a series of 3 or more.

Nor would commas separating the elements do anything to tell you whether the common portion of the list ended with "is" or "to" and whether the first element started with "contrary" or was just "law".

If one was especially concerned about avoiding potential confusion of the meaning, one could rewrite the sentence as "This is contrary to both law and sound economic policy", but while that would be more explicit, it is unnecessary, but in any case throwing superfluous commas into the sentence doesn't help anything.

It is one thing to pedantic. It is something worse to be pedantic and wrong.

This sounds like going back to a variation on self-signed certificates.

Its not. Its a verification scheme orthogonal to certificate chains which can be used either alongside traditional certificate chain verification or without traditional certificate chain verification. It is compatible with self-signed certs, but equally compatible with CA-signed certs. Ideally, you'd use it with CA-signed certs, since CA-signed certs -- though they have known problems -- are better than nothing (unlike self-signed certs) on a first connection with no prior information, but after that TACK pins are useful to detect later CA-assisted shenanigans.

If you want a PKI based on signing certificates by CAs, the CAs need to be entities whose primary income does not derive from signing certificates.

No, what you need is an effective mechanism to detect and revoke trust in nefarious CAs. If CAs aren't trusted, they are of no value to potential clients, and thus the stream of income from signing certificates dries up. The problem isn't that CAs derive income from signing certificates, the problem is that there is no effective accountability mechanism that imposes sufficient consequences to make it so that improperly signing certificates reduces the marketability of that CA's signing services.

Or, the judge could consider 9 lines out of many thousands of lines in the original work as de minimis copying and admonish Oracle for bothering the court with trifles.

First, that's completely unrelated to the API issue [the 9 lines are implementation that was copied, not API], and, second, the judge already ruled (order dated May 10) against Google's motion asking for a judgement as a matter of law that rangeCheck was non-infringing that asserted a wide variety of arguments, including the de minimis one.