Category Archives: Articles

FRAND Is Always Discriminatory

I participated in a study asking about the fairness, reasonableness and non-discriminatory nature of FRAND licensing in the context of licensing of patents in standards. I was surprised to find people there asserting there was no conflict between FRAND licensing and open source software. Here’s a simple explanation why that’s wrong.

Since patent licensing is by definition bilateral, and since open source communities that aren’t run by a single vendor are by definition multi-lateral, any standard which includes patents that require licensing discriminates against true multi-participant open source implementation. By definition, patent licensing as a precondition of implementation of a standard cannot ever be non-discriminatory. Even zero-fee licensing is discriminatory as it still requires implementers to seek permission, the antithesis of open source.

Software Freedom, Utility and Maintenance Time

Whilst many may long for a truly open source OS that meets all of their needs, the reality has always been that compromise has a role to play whenever it comes to picking your operating system. Despite the availability and increasing ease of installation of purer open source systems, there remains a trade-off to be made. Systems with a high level of software freedom and an intuitively usable interface seem to require high levels of maintenance to keep them alive. Where a system with high software freedom’s been designed to require less maintenance, the usability seems to suffer. Of course, this triangle has a third point to it too: where a system is both easy to use and maintaining it doesn’t consume too much of your time, it’s software freedom that takes the hit.

What sort of system you choose should depend on which of those three factors you prioritise. Read the details about this theory, along with some pointers for recognising systems that value software freedom in Simon’s InfoWorld Article.

How should technologists respond to terrorism?

The attack on the Charlie Hebdo office in Paris was a horrific crime. It has shocked the world and roused a great deal of public upset, outcry and anger.

Whilst it’s too late to prevent the tragic loss of life caused by the murderers, the office assault was the seed of another, ongoing attack, in which each of us is struggling, though we may be unaware of it.

Acts of terrorism provoke society into attacking itself. Justice and law making systems designed to protect and uphold our freedoms and rights are tricked into restricting and removing those self same rights and freedoms for everybody, in attempts to prevent future attacks and to placate the fierce public desire for action.

An understanding of the openness of the internet and the ways in which we benefit from it gives technologists a unique insight into the value that society gains from remaining open. That’s why Simon’s used his InfoWorld response to the tragedy as an opportunity to call on readers from the technology industry to respond to terrorism by defending openness. Check out his full article on InfoWorld.

Twitter’s open source emoji success

Twitter’s approach to the problem of emoji provides a classic case study in open source best practice:

Many different applications have looked at the success of emoji and the various rights problems associated with the use of existing images and decided that the best work around was either to ignore the rights issues and reuse existing materials anyway or to create whole new image libraries which they would then have control over. Twitter on the other hand, has taken a different, more open approach.

After commissioning a new set of emoji graphics, the company implemented a library for parsing emoji tags and replacing them with read-to-render code strings for various platforms, including HTML. Now there are hundreds of emoji available to add to your tweets. Internally, the project was treated as open source from the off, with GitHub being used as the development repository. Now, both the code and the graphics of the project have been open sourced, using the MIT license for the code and CC-BY for the graphics.

Twitter has successfully managed to implement a feature essential for its global market, get it maintained in conjunction with others, and win broad credit. In order to learn from and repeat their success, its worth looking in a little more detail at how they achieved this: check out Simon’s InfoWorld article “Twitter emoji: 5 lessons for effective open source.”

Big progress in Microsoft’s open source journey

Microsoft’s recent announcement that much of .Net will become open source and that it will support both Linux and Mac OS X is fantastic news. Along with the additional, full-featured, no-cost versions of its developer tools the company is introducing (though they remain proprietary), this represents a large, positive step in Microsoft’s open source journey.

Simon’s spoken and written a number of times about the seven stages of a corporation’s journey into open source and used Microsoft as an illustration of his ideas. This new development puts the company very clearly at the fifth stage of Simon’s scale, which is impressive, but begs the questions “what’s next?”, “how does Microsoft’s open source journey continue to develop from here?”

The answer lies in a holistic view, in which respect for open source extends to every business unit of this famously divided company. While those business units that don’t yet respect open source continue to use tactics such as patent attacks on Linux community members and covert political moves to undermine the Open Document Format, further progress will be slow in coming. Microsoft’s gradual acceptance of the inevitability of open source however, seems to be in full swing. For more detail, check out Simon’s InfoWorld article.

Google’s Supreme Court Petition

Last week Google petitioned the Supreme Court of the United States for permission to appeal the Federal Circuit decision to overturn their victory against Oracle. You may remember that Oracle sued for patent and copyright infringement over Google’s use of a small subset of the huge Java class API in Android. Oracle lost in front of a jury in the district court in San Francisco, but then won on appeal.

The case will be a significant landmark for the technology industry regardless of the outcome, but the real impact of any decisions reached will be felt first by Android. The petition itself then suggests that the wider impact of the case is narrower than others might have feared, as its effects only come into play with substantial systems of APIs. Still, in the event of an Oracle win, open-source-licensed APIs would suddenly become much more important and the work of creating developer ecosystems around proprietary APIs would become much more challenging.

The Google petition includes three detailed and thorough arguments for the overturning of the Federal Circuit decision. For a look into what those arguments are and more on the significance of the case, try Simon’s InfoWorld article.

 

Superclean code puts LibreOffice at the head of the trend

If you’ve been around open source for a while, perhaps your understanding of the quality of open source code is based on older projects like the infamously unstable OpenOffice.org codebase. Times have changed however and evidence of this can be found in the work of code improvement vendor Coverity. They recently announced that LibreOffice, (four years old this weekend and based on the old OpenOffice code), has a defect density of just 0.08, compared with similar sized open source projects which averages at 0.65 and proprietary code which averages  at 0.71.

LibreOffice is an outlier, an extreme example of clean, defect free code, but it also fits into a larger trend. Since the publication of its 2013 Coverity Scan Open Source Report, Coverity has asserted that open source code quality now outpaces that of proprietary code. While an open source license doesn’t guarantee quality, it does allow for evaluation of quality and encourages collaborative efforts toward improvement. Which is why you can expect to see the trend noticed by Coverity continue over the coming years.

Read Simon’s full article on InfoWorld.

Bringing LibreOffice to Android

At the LibreOffice Conference, The Document Foundation issued a tender document looking for bids to develop an Android implementation of LibreOffice. Could this influx of money affect the ethics and work ethic of today’s open source community? It’s not actually the first time the community has experimented with bringing a LibreOffice editor to Android. The scale of the task however is enough to dampen volunteer enthusiasm and given the lack of commercial motivations for engagement, disappointing low volunteer turnout is not actually a great surprise.

When it comes to money, The Document Foundation is faced with the mixed blessing of having plenty available. In what way is that blessing mixed? As a charity the foundation is legally required to spend those funds over the course of the following year or so. With the clock moving swiftly on, The Document Foundation has already invested in development infrastructure for testing, the backing of community activities and the hiring of sys admin and administrative staff. There’s still a sizeable portion left over though. Knowing that spending in areas where the community is already intrinsically motivated might well reduce contributions, TDF has decided to focus remaining funds on development of the Android port, hoping to bootstrap a necessary new community in the process. Now it simply remains to be seen if anyone will bid to do the work!

Read Simon’s full coverage in his InfoWorld article.

Alice kills trolls

The effects of Alice Corp. v CLS Bank are beginning to be seen and they highlight the landmark status of the case. The Court of Appeals for the Federal Circuit, previously considered by many to be strongly pro-patent, has now used the Alice decision to resolve large numbers of patent cases by finding the software patents in question to be invalid. In fact, even lower courts are beginning to use Alice to strike down patent cases, declaring them invalid because of unpatentable subject matter.

Of course, the patent trolls are not going to be caught sleeping. Conceding that Alice invalidates many of their cases, pro-patent advocates will now begin hunting ways to get around the ruling. As software patent consultant Bob Zeidman said “every time there’s a court ruling it just means that you have to word the patent claims differently.” This might well mean adding function claims to patents, a move which also significantly limits their disruptive power. Perhaps that’s a compromise open source developers could live with, but let’s not let our guards down. In the meantime though, I propose a toast, “to Alice”.

For more on this topic, read Simon’s article on InfoWorld.

All your data are belong to us

If we store… data in a place with public access, it will eventually become public.

Simon’s conclusion might seem a little paranoid, but it seem’s there’s plenty in the news at the moment to back up his position. From the hacking of celebrity Apple iCloud accounts, to the DEA using decades old phone records (stored by AT&T) as a covert source in investigations, vulnerable data is in the headlines.

The NSA and GCHQ also come under the spotlight. Simon’s contention is that all intelligence agencies worth the name are now actively gathering all online information that doesn’t require illegal action to take. This information includes all email, instant messages, Web pages, and social media traffic just for starters. The legality of the information harvesting is justified by a whole series of explanations, definitions and broad interpretations of legal doctrines, which enable such agencies to claim the information gathered is “public”.

Most of this data never gets looked at. Instead it’s cached in gigantic data lakes. According to the NSA’s legal advisers, “wiretapping” or “hacking” starts at the point a human being actually analyses or interprets the data. Of course, given the availability of the data, as soon as legal permission is given, tools like the NSA’s XKeyscore enable them to fish the data lake for relevant information. Given that you can’t dissociate data gathering from data usage, we all need to account for all the costs of putting information online, not just the ones associated with our primary goals. Don’t just think about what your data is being used for now; think about what it could be used for when it’s no longer your job. Read Simon’s full article in InfoWorld.