Carlo Piana :: Law is Freedom ::

This is a joint blog post written together with Matija Šuklje with the goal to disband some popular FUD that you lose your warranty if you flash or root your device

Rooting your phone

Does rooting your device (e.g. an Android phone) and replacing its operating system with something else void your statutory warranty, if you are a consumer?

In short: No. Just the fact that you modified or changed the software of your device, is not a sufficient reason to void your statutory warranty. As long as you have bought the device as a consumer in the European Union.

I am here speaking software patents. I was replying to a thread on G+, and wrote a very long reply, that indeed is a post. It's on Red Hat patent promise, which has some unfortunate language that can be interpreted as "in general software patents impede innovation", meaning that in some cases they don't. I think that my reply can be of more general interest and make an excerpt here, with a few minor edits.

The post

I hate to defend rich companies and not getting paid for it (not that it would be the first time ;-), but in this case I will make an exception, also because the people who conceived the strategy of registering software patents for defensive purposes are good friends and trustworthy Free Software friends. Red Hat is a company full of cash, which is eating chunks of market away from even richer companies. It makes software, Free Software, it contributes heavily to Linux, it has no proprietary exploitation of the software it makes.

This is it, it's over. The last remaining pending issue spawning from the 2004 Decision (the so called "Monti decision"), by which the European Union slapped Microsoft with an unprecedented antitrust remedy, has ended, barring an unpredictable appeal. A decision imposing 899 million euro fine, for non compliance with the obligation to provide complete and accurate interoperability information under Reasonable And Non Discriminatory conditions, was by and large upheld by the General Court in case T-167/08, where I represented the FSFE and the Samba Team, intervening in support of the Commission.

I have now read the decision in its all 26 printed pages. Among many details concerning procedural fine points that would bore to death most of the readers, I have found some points that are worth pointing out, since they confirmed my/our positions that we put forward since 2005. That's when the whole "implementation" phase started, after the President of the Court of First Instance (that was the General Court called back then) refused to suspend the 2004 Decision pending judgement on the merits.

Today Italy mourns the victims of the earthquake that in two different occasions struck the lands between Emilia and Lombardy.

To the victims and their family a word of sorrow from the deepest of our hearts.

In Italian "api" means "bees" (plural). API in computer science means "Application Programming Interfaces", which are bits of code in a computer program that expose functions and calls to other computer programs so that they can interact. For instance, a platform (say, Gnome or MS Windows) can expose a function to call upon a printing dialog, and all application running on it only needs to invoke that function and an interaction can happen, so the application can print. The internals of the same API can be reimplemented many times without changing the outward facing interfaces, so that the applications written against them remain workable ‒ actually this happens on a regular basis.

A Federal Judge for the Northern District of California has ruled that the source code that declare a method to invoke the same function as in the API (thus representing the outward facing part) is not subject to copyright. This follows a quasi identical ruling of the European Court of Justice in the SAS case [Case C‑406/10].

Update: see the comment on the final judgment on the copyright protection of the API.

There has been a lot of noise in some areas of the Internet around what to make of the verdict that the jury has taken in the Oracle v. Google case.

For the benefit of the readers, here the questions and what the jury has answered. For Europeans, it is very odd to see a jury to decide in matters that are strongly legal in nature, but that's how it goes up there. Bear in mind, though, that the jury is only responsible for the assessment of the facts, it's up to the judge to have a final say about the law.

Also, bear in mind that the judge has instructed the jury to decide as if the API (Application Programming Interfaces) are a copyright subject, but that is just a speculative statement, the matter will be settled by the judge in its final decision.

Note: Updated to include lock-in and tying. Some changes on moral rights to clarify that

Laws are more often than not an annoyance, despite their aim to improve the legal framework in any given field. Free Software (AKA "Open Source") has thrived despite the absence of any legal recognition by the law, if not in spite of rules that clearly are shaped around proprietary software. In many jurisdictions it has passed the enforceability test. So, no laws seem necessary to make it work. Yet, can some legal principle be put forward, and included in some laws, to help?

"If it works don't fix it", so goes the common saying. But if it works now doesn't mean it will work forever. It is nevertheless upon lawyers, and legislators alike, to foresee problems ahead of their actual happening, and brace for the potential harmful event. But any laws that would regulate Free Software would likely harm some parts of it, and change the games to favor one kind over another, or impose conditions that are not welcome or productive – something that legislator, even with the best intentions, often do – and in general could cause as many troubles as they would produce benefit. "Primum non nocere" is the paradigm for medical actions, even though drugs by definition only produce a net benefit by inflicting some limited damage. Is there a medicament that has entirely good effects without any negative ones? Arguably there is not. But with laws we can achieve something closer to this optimal benefit, which economists know as "Pareto Efficiency".

So this is a call for Pareto Efficient Laws, and Pareto optimal only laws.