The topic was Moblin - or more specifically, the Moblin.org open source project and our goal to get a larger and larger open source developer community together that will drive the direction of the open source software stack used on internet focused devices. Lots of "open source" in that last sentence. And that's intentional. There are many projects that try to create open source or Linux based stacks for mobile devices. But none of them have really convinced me that they are truly open source. Openmoko may be an exception - but as I wrote a couple of months ago progress there has been slow.

So why is this focus on open source important? Because only if we allow the community to take the stack and push it in any direction they want to go, contribute back, take the ideas and start new projects and overall make this project theirs, only then will the type of innovation happen on Moblin that I think will be the key to the success of mobile Linux devices. And while I love my Mac and I know that my wife likes her iPhone, I'd much rather see a free and open stack be successful and competitive and ensure that innovation will continue.

So go take a look at Moblin.org and especially at the playground.

The event reminded me once again just how important "community" is - not just in the context of the open source community in general but also in the sense of a community inside of an organization that is working on similar goals and needs to get together every once in a while to become stronger. I certainly feel that the conference created many new connections and many ideas and will help us be more successful and more aligned in our actions.

We are thinking of ways to bring in people from the outside (so far we just had a couple of outside presenters), but we also want to retain the opportunity to openly talk about open source in the context of (potentially confidential) future projects. Maybe we'll find a way to do something similar together with IDF or another suitable event. Feedback is always welcome!

---

This is (obviously) my blog at work - if you want to know what I am interested in outside of Intel, take a look at my personal blog.

In the interest of full disclosure, of course I am one of the people who have talked about this topic many times in public and who actually talked to Jon about this at linux.conf.au. But I don't think that should stop me from commenting on it here :-)

Fundamentally, some vendors release decent documentation to their components (note, Intel recently released the Programmers Reference Manual for our latest graphics chips), while others don't. Yet right now more volunteers and community members are focusing on and supporting the chips from vendors that don't release specs. That's backwards, right? It reduces the incentive for vendors to release decent specs. In order to help our goals ("our" being "the open source community" in this context), we should support those vendors who release specs. And those are many more than just Intel, but I hope you'll forgive me for not linking to competitors' sites from an Intel blog...

Releasing specs was what got Linux started almost 27 years ago (Linus used the Crawford/Gelsinger book on Programming the 80386 when he began working on the Linux kernel). And it's what enables innovation. Smart developers shouldn't be wasting their time to figure out how to program chips. They should focus on creating new exciting things using those chips. And it should be in the best interest of the manufacturer to help them do so.

PS: in case you are wondering what I have been doing (since I haven't posted here in almost a year) - I simply spent more time on my personal blog and (oops) forgot to update my blog at work... 

If you are interested in Linux, why don't you give them some feedback?

Fascinating idea, but of course it misses the economic realities in a few crucial issues. For example, they are stating

Don't you owe it to your shareholders to work with others to ensure their intellectual property isn't being violated?

Actually, the executives at Microsoft - like at every other corporation - owe it to their shareholders to extract the maximum long term value from that IP. And apparently they think that the maximum value for their shareholders can be obtained by creating FUD (fear, uncertainty and doubt) about the question whether Linux violates Microsoft IP and by using the implicit and explicit threat of lawsuits to slow down Linux adoption in the industry. If they followed the request in this initiative and enabled the Linux community to work around any Microsoft IP that Linux may or may not be infringing, then they might be doing the honorable thing, but it will take some very interesting arguments to show how that would be in the best interest of Microsoft's shareholders...

Still, I think this is a good initiative as it tries to shine some light at a rather murky issue - and from a larger industry point of view I am convinced that more clarity in this area would be very beneficial for all of us.

The question to me is a different one. Is this the way to go? Do I really want my data to live in Google's data center. What do I do on the off chance I am off line? Like, on an airplane. Or (gasp!) at a place with no network connectivity. What do I do if the network is down but I need a document. What happens if they lose data? What...?

Lots of questions. In general I prefer to be in control of my data. There's a reason why my private blogs aren't hosted but are running on a Mac Mini at home. If I don't back up my data, that's my own fault. But if something goes wrong at Google and I can't access my data... that's something I'd rather avoid.

So good luck with Google Office, I mean, Google Apps. Don't hold your breath for me to show up there...

Oh, and if you are looking for an alternative to Microsoft Office? Take a look at OpenOffice and add a nice email / calendar solution. I personally like text based clients, but to make it comparable to Outlook, how about Thunderbird. Free (actually, open source) and runs locally on your system. Whatever your OS might be.

Interesting. Not necessarily representative (we all know how easy it is to cheat on polls on the web), but I have to assume that Dell has reasonably safeguarded the integrity of their site. So I'd mark this down as very interesting.

I thought this was a fascinating topic - open source has so fundamentally changed the software world and many (most?) software startups these days are at least partly involved in (or using) open source. And finally, for software startups, M&A seems to be the typical exit these days (the pre-announced IPO of MySQL not withstanding). So I was surprised to see only about two dozen attendees in the room...

The panelists (from Black Duck, Palamida, IBM Ventures and DLA Piper) covered the different sides of the issue - two consulting companies dealing with open source risks, a VC and a law firm focused on M&A.

And the issues that were raised were really quite interesting. Many licenses used for software today are not assignable, so if you buy a software company that based their business on such software you may have to "re-aquire" that software. Sounds easy: download the software again. But what if the original version isn't avaialable anymore? What about internal changes that haven't been tracked since the download?

Software companies today use a large array of software components that they download from the internet - under a variety of licenses. And the degree of legal analysis done while assembling these components is typically minimal. With interesting consequences once you do due-dilligence. From products that are built from components that have incompatible licenses to the most-feared "open source contamination" of the comany's IP.

And the value of companies is suddenly less in the hard assets and more in the people that work there and the viability of the community around them. That sounds like the reality in any acquisition. But think about it: with proprietary software, even if the key developers leave, you still own the IP and can start ramping a new team. In open source, the developers can turn around and use the exact same software to start a new business that competes with you. Which means that employee retention suddenly becomes the biggest challenge in an M&A deal.

And this is the piece that really was the biggest "a-ha" for me, as it means that working as a developer in an open source startup (and becoming a key contributor) gives you a better chance to really participate in the success of the business than in a proprietary software company. I don't think this was the main message that the panel wanted to leave us with - but to me it was a very interesting point to bring home.

But reading the original announcememt by Ubuntu's CTO Matt Zimmerman things sound a lot less black and white. Which is disappointing in some ways. All the announcement seems to say that Ubuntu will not make composite mode the default in its next release (which would require the use of proprietary video drivers with the cards of several vendors). But it unfortunately still say this:

The exception granted to the Ubuntu Licence Policy at the project's
inception, which allows the use of closed-source components where
necessary to provide sufficient hardware support, remains valid and
pragmatic. For example, this exception has been applied in the past with
regard to wireless network drivers. We consider making full use of video
hardware to be an important component of that hardware support.

And goes on to say that the decision not to make composite mode the default is more based on the lack of maturity of that technology than on the requirement to use proprietary drivers for it.

Not the signal I would have been hoping for.

He has started his "we write open drivers for free" campaign and even posted a nice little FAQ. I love this one:

Q: You forgot to mention the most obvious benefit to the company, "They might sell more of their devices!" How can they turn down an offer like that?

A: Good point

Needless to say - I am all in favor of more companies figuring out just how important open source drivers are. And how easy it is to write them. Or even have them written. Let's hope that Intel's example (and Greg's initiative) help to spread the word.

I really like WordPress, it's easy to use, powerful and best of all, it's open source. It has an active developer community that continues to make it better.

This switch alone will make sure that I'll be more active posting here at Intel as well (I have been a bit... shall we say... tardy lately...)

Oops

Good thing nothing important happened during that time. Or did it? Let's see.

• The FCC forced at&t to guarantee net neurality - at least for a few years. That's good news, I think.
• Also, Zune was launched and no one noticed (it actually looks according to preliminary data like Apple's iPod even increased market share last quarter!).
• The expected fight between PS3 and Xbox360 had an unexpected winner - the oddly named Wii!
• ...

Ok, maybe stuff did happen. And much that I should have written about. Oh well - with CES and MacWorld coming up, I'm sure there'll be more to talk about next week.

What a fascinating thought. Not exactly new, but it's rare that an widely listened to leader in the open source world states it that bluntly. The current success of Apple should have made this obvious to most people, but it clearly isn't. Certainly most open source hackers couldn't care less about the look, feel, and usability of their software, as long as it is good code.

Ah.

Maybe Mark is on to something even more fundamental. Beauty means different things to different people. And for the true software developer, maybe beauty means well architected and easy to maintain code. And therefore that's the beauty they aim for. So a lot of people in the open source community are already aiming for beautiful software. Only the audience that understands that form of beauty is rather small (would that be a good description of the early community of Linux kernel hackers, for example?)

So what's the conclusion? To me it's that we need both, and that we need to get the people together that can provide both. Beautiful code (as in "software that works") and beautiful look and feel (as in "Steve Jobs will want to copy it in the next version of Mac OS").

Ubuntu certainly trys to go in the right direction of making Linux beautiful. But as a person who uses Mac OS Tiger as his main desktop OS for day to day work I have to say that even Ubuntu has some ways to go.

No, I am not going to start a blog entry about all the things that I think are wrong with the RIAA's and MPAA's approach to this topic. An Intel blog might be the wrong place for that, anyway. Instead, I want to talk about all the good things you can do with torrents that are within the current legal framework and that aren't being challenged by interest groups today.

Look for example at the efficient distributions of new releases of open source software. Most of the major distributions (and a number of the bigger projects) offer torrents in addition to traditional download sites. The advantage? They avoid the major bottleneck that always happens around new releases of their software; the moment a user was able to get a part of the software downloaded to his machine, he himself becomes a re-distributor and reduces the load of the central server. And the more people participate in the swarm and keep seeding, the better is the overall available bandwidth for additional people who want to download.

This works so well that there are examples of rather traditional outlets that don't offer ftp or http download anymore. ChinesePod, an online Chinese language program, offers their podcast for bulk download - one hundred episodes at a time. But only as torrents.

Peer to peer networking is a highly useful technology that can help overcome the bottlenecks of today's Internet. And open source efforts like Azureus continue to drive this technology and make it more powerful and easier to use for the end user.

Check it out!

Dirk

Despite the bombastic title, this is actually an event that brings together a number of interesting players. Political and technical leaders in the open source community in China. And a number of open source people from Europe and the US (including Jim Zemlin from the Free Standards Group, Brian Behlendorf from CollabNet, David Axmark, one of the founders of MySQL and of course myself).

The goal of COPU and of this event is to help the Chinese Linux ecosystem to convert from an open source consumer to a community member. And that is harder than one might think. The language and cultural barriers are huge. And the understanding of the advantage of a global community for open source projects is not very widely shared.

Especially on the political side we see a number of people who only see open source as a way in which China can create a national software industry that is independent from the rest of the world. These people of course miss the point of open source - the fact that strength is in numbers. And that while forking is certainly one of the fundamental rights in open source, it always needs to be an ultima ratio, a last resort if a project truly is going the wrong way.

Creating local "copies" of existing projects just to avoid the cultural barriers is a certain way to miss out on the biggest opportunity that open source offers to you. The combined innovation and quality control that a large diverse project will give you. It may seem simple when forking to just "bring over" the important changes in the main project into your local project. But as time goes on you either lose contact to the main branch of the project (and therefore miss out on the innovation), or you spend the majority of your time on back-porting the changes made elsewhere (and therefore miss out on the ability to adapt and enhance the project for your local market).

If instead the local communities join the global open source projects (like the Linux kernel, X.Org, OpenOffice or Apache) they can bring their wisdom and insights, their bug reports, feature requests and of course their code contributions into the mainstream and at the same time advance the project and benefit from the progress that is made by developer teams elsewhere.

Unfortunately English is the lingua franca of the open source world. But frankly, written English that is at least marginally comprehensible is not hard to learn - and the tolerance for non-perfect English used to present good feedback (or good code) is very high. And while the tone on many mailing lists might seem harsh, once you participate for a while you will find out that the majority of the developers is genuinely happy and excited to welcome new contributors.

Yes, there's a learning curve and there may be frustrations involved. But the advantage of participating in the global open source community are so big that this seems to be a small price to pay.

It all sounds painfully obvious. But listening closely here, it seems that it's not a choice that is popular here in China. There will be a lot of work ahead for COPU to be successful.