Lessons from Google and Red Hat for Facebook and Open Source

November 19, 2007 Leave a comment

Matt has articulated something I’ve tried to explain to people many times. From his post:

Twentieth century software business models focus on scarcity because they’re founded upon 20th century conceptions of property (actually, their origin is a few centuries’ older than that, but never mind).

Scarcity is the absolute wrong way to build a software business in the 21st century, with the rise of digitization. It is pointless and fruitless to insist that the digital world act like the physical or analog world, and build business models that conform to this false view. To thrive in the new software world, we need to embrace its changes rather than fight them.

The old model was to assume that the value was in the software itself and to therefore lock it up. It turns out, however, as Tim O’Reilly notes, that data is the real value, not bits and bytes. You don’t discover or, rather, uncover, that value until you have abundance.

There will be a few companies that continue to effectively monetize the old world. Oracle, Microsoft, IBM, and SAP, to name the four primary ones. But these outdated models will eventually fall against the rising tide of abundance-based business models, because the new models recognize that digital bits really do want to be free, and that all attempts to artificially lock them are doomed to fail.

It fascinates me how many companies still operate under the guise of artificial scarcity. The landscape is rapidly changing and the shift is so large and fundamental that it seems some are completely missing it. That’s not all that surprising in the end, but as with most industry paradigm shifts, at least one seemingly invincible player will almost invariably not survive.

–jeremy

Filed under Google, Open Source, Red Hat

OLPC Give One Get One program is now live

November 12, 2007 Leave a comment

The OLPC Give One Get One program is now officially open.

One learning child. One connected child. One laptop at a time.

The mission of One Laptop Per Child (OLPC) is to empower the children of developing countries to learn by providing one connected laptop to every school-age child. In order to accomplish our goal, we need people who believe in what we’re doing and want to help make education for the world’s children a priority, not a privilege. Between November 12 and November 26, OLPC is offering a Give One Get One program in the United States and Canada. During this time, you can donate the revolutionary XO laptop to a child in a developing nation, and also receive one for the child in your life in recognition of your contribution.

For a donation of $399, one XO laptop will be sent to empower a child in a developing nation and one will be sent to the child in your life in recognition of your contribution. $200 of your donation is tax-deductible (your $399 donation minus the fair market value of the XO laptop you will be receiving). For all U.S. donors who participate in the Give One Get One program, T-Mobile is offering one year of complimentary HotSpot access. If you purchase one now, you should have it for the holidays. This is a great way to get a child you know a fantastic gift while also helping someone less fortunate than yourself. The XO is a Linux based laptop. I’m proud to say that LQ has purchased one:

Confirmation

Thank you for participating in Give One Get One. Your donation will bring education and enlightenment to children of the developing world, and, in recognition of your gift, you will be receiving an XO laptop for the child in your life as well. If you have any questions or problems, please contact One Laptop Per Child at service@laptopgiving.org. Should your employer wish to match your donation, we are a 501(c)(3) organization and our EIN# is 20-5471780. Thanks again, and welcome to the One Laptop Per Child community!

–jeremy

Filed under Linux, Open Source

Oracle mulls vote on open source at upcoming shareholders meeting II

November 10, 2007 Leave a comment

As a follow up to this post, Jonas just informed me that 127,717,018 shares voted in favor of the proposal. From his email:

In the end, we received the support of 127,717,018 shares. That is 3.58% of the vote and approximately $2.8 billion in assets. As I’ve mentioned before, taking into account Larry Ellison’s ownership that percentage can be viewed a bit differently. That number goes up to 4.99% if you remove his approximately 1 billion shares from the calculation.

While 3.58% may not seem like a significant number, it is a very respectable vote for a first year issue that nobody ever heard of and with no advocacy budget. These are the kinds of votes climate change proposals received years ago and now they pull in up to 30%. That is not to say that FLOSS is the same kind of issue as global warming, but it gives you an idea of how support often starts off in the single digits. It is also significant that we broke the 3% threshold, because under SEC rules we are now entitled to bring the proposal forward to the 2008 AGM if we wish.

3.58% may not sound like a big number, but as Jonas points out, since Ellison and institutional investors voted against the proposal it is a significant number. One hundred million votes is more than I’d have anticipated (although being new to this kind of thing my expectations don’t mean much), but it’s great to see the issue getting this kind of support right out of the gate.

Matt is also blogging about this here.

–jeremy

Filed under Open Source, Oracle

Linux wins Nigerian school desktops back from Microsoft

November 9, 2007 Leave a comment

It looks like the open letter to Steve Ballmer from Mandriva may have had an impact. From the article:

Mandriva had closed a deal in mid-August to provide a customised Linux operating system and support for 17,000 Intel Classmate PCs intended for Nigerian schools, but found out last week that the company deploying the computers for the government, Technology Support Center (TSC), planned to wipe the computers’ disks and install Windows XP instead.

Now, however, a government agency funding 11,000 of the PCs has overruled the supplier. Nigeria’s Universal Service Provision Fund (USPF) wants to keep Mandriva Linux on the Classmate PCs, said an official who identified himself as the programme manager for USPF’s Classmate PCs project.

“We are sticking with that platform,” said the official, who would not give his name.

Mba-Uzoukwu wrote that Microsoft is still negotiating an agreement that would give TSC US$400,000 (£190,323) for marketing activities around the Classmate PCs when those computers are converted to Windows.

“Microsoft is able to offer a comprehensive education solution – including software, training and support – on the 17,000 Classmate PCs for 200 schools across Nigeria,” the statement said.

After public statements from Mandriva officials implied the marketing deal is legally questionable, Microsoft said last week that it complies with international law and the law of the countries in which it operates.

It’s not clear how much TSC would pay for each Windows XP licence. Efforts to reach senior managers at TSC, which is a subsidiary of Alteq.ict, an IT consulting business in Nigeria, were unsuccessful.

However, details on Mandriva’s deal with TSC have emerged. Mandriva is providing a customised OS for Nigeria for under $10 (£4.7) per licence, including support, according to its local partner.

It’s clear that Microsoft is willing to go to pretty great lengths to ensure that it’s able to put some stakes down in the growing third world economies. I’m not a lawyer, so don’t have any insight into the legality of this $400,000 “marketing” deal, but it seems that for now the machines are going to be delivered with Mandriva and stay with Mandriva. I have a feeling we haven’t heard the end of this story though.

–jeremy

Filed under Linux, Mandriva, Microsoft, MSFT, Open Source

Constructing the Bazaar: Taking advantage of the open-source development model in your project

November 6, 2007 Leave a comment

A great post by Juergen about the Open Source development model. If you truly want external community participation, it’s not enough to just throw code over the wall – you have to make participation as easy as possible.

Think distributed and think part-time to enable the Bazaar

And this then provides us with the answer to our initial question: Why do so many open-source projects not have the active community of external contributors they are hoping for? Because they have been largely developed by co-located teams of hired software engineers, 100% dedicated to the project, managed and organized like any traditional software development effort. This seems to be especially true for the new crop of ‘custom build’ open-source companies, which would like to take advantage of the open-source business model. They might hope to also enjoy the advantages of the open-source development model one day, but achieving that requires a conscious effort.

Here are a few things to keep in mind then:

* Do not be seduced by the simplicity and ease of communication that you have in your co-located team.
* Always consider what it would take for any new, external developer to jump in quickly and become productive, even if this developer can only use much less direct communication channels.
* Consider that external developers can only work part-time on your project, and do not have the time to follow long discussions or become truly familiar with your system down to the last module.
* Always remind yourself to focus on the distinct and simple (!) modules in your design and reflect them in your code accordingly. External developers may only have time to dive deep into one or two of the modules, not the entire system. Minimize extensive system understanding as a prerequisite.
* Always make sure the modules and their APIs are defined and documented clearly.
* Use code reviews to prevent any more direct usage of a module’s code that would circumvent the APIs in any way.
* Enforcing encapsulation via the programming language is not sufficient in this situation: Particularly insidious and hard to spot is the code that relies on knowledge about the internals of another module, even if it does not take short-cuts around the APIs. This problem can be caused by APIs that are designed with the assumption that the user of the API knows a few internals of the module. In co-located teams with effortless and direct communication, this is a trap one can fall into easily. Therefore, we need to strive to build our code with true knowledge isolation in mind.

Conclusion

If these points are not considered on a daily basis, the code does not have the level of modularity and simple APIs that are needed to facility bazaar-style development. Even the best intentions of new, external developers can be frustrated by unnecessary prerequisites. And as anyone trying to build a community knows: The most important ingredient is to make it easy to join and get started. In our case here this means: Make it easy to be productive and achieve the rewarding feeling of success.

As Juergen mentions, this is something that’s taught in the most entry level of classes. It really is something we all know. It’s all too easy to take shortcuts though, especially when deadlines are looming and a quick chat with a person in another team lets you use undocumented internals to save time. Being cognizant of this trap will serve you well, and save time, in the long run.

–jeremy

Filed under Open Source

Where's my Gphone?

November 5, 2007 Leave a comment

Google finally made the highly anticipated Gphone related announcement today:

Despite all of the very interesting speculation over the last few months, we’re not announcing a Gphone. However, we think what we are announcing — the Open Handset Alliance and Android — is more significant and ambitious than a single phone. In fact, through the joint efforts of the members of the Open Handset Alliance, we hope Android will be the foundation for many new phones and will create an entirely new mobile experience for users, with new applications and new capabilities we can’t imagine today.

Android is the first truly open and comprehensive platform for mobile devices. It includes an operating system, user-interface and applications — all of the software to run a mobile phone, but without the proprietary obstacles that have hindered mobile innovation. We have developed Android in cooperation with the Open Handset Alliance, which consists of more than 30 technology and mobile leaders including Motorola, Qualcomm, HTC and T-Mobile. Through deep partnerships with carriers, device manufacturers, developers, and others, we hope to enable an open ecosystem for the mobile world by creating a standard, open mobile software platform. We think the result will ultimately be a better and faster pace for innovation that will give mobile customers unforeseen applications and capabilities.

It’s important to recognize that the Open Handset Alliance and Android have the potential to be major changes from the status quo — one which will take patience and much investment by the various players before you’ll see the first benefits. But we feel the potential gains for mobile customers around the world are worth the effort. If you’re a developer and this approach sounds exciting, give us a week or so and we’ll have an SDK available. If you’re a mobile user, you’ll have to wait a little longer, but some of our partners are targeting the second half of 2008 to ship phones based on the Android platform. And if you already have a phone you know and love, check out mobile.google.com and make sure you have Google Maps for mobile, Gmail and our other great applications on your phone. We’ll continue to make these services better and add plenty of exciting new features, applications and services, too.

This is fairly inline with what I was expecting. While some were anticipating a hardware device from Google, a platform plus stack release makes much more sense. They don’t have to get into a very low margin high capital business and they can keep existing partnerships in place without the added stress of direct competition. This move should have fairly large repercussion for the entire industry. With the availability of a full SDK for this platform, Apple is really going to get hurt if they are too closed with their SDK, which will be released soon. Looking at the Open Handset Alliance members, you’ll notice both Nokia and FIC are missing. You have to wonder how this announcement will impact Maemo and OpenMoko, respectively. I’d guess we’ll see many more partners and stepped up competition as a result of this announcement, so I’ll keep an eye out and post an update when the dust has settled. One thing is clear, the Linux mobile landscape is heating up.

Additional Reading:
TechCrunch
Edgadget
Mashable
Techdirt
Linux Foundation (which has pointers to many of the Linux mobile initiatives and players, including: ACCESS, A La Mobile, Celunite, FST, Mizi Research, OpenMoko/FIC, Purple Labs, Trolltech, LiMO, LiPS, Moblin and more)

–jeremy

Filed under Apple, Google, Linux, maemo, Nokia, Open Source, OpenMoko

How badly is CentOS hurting Red Hat?

November 5, 2007 1 Comment

That’s the question asked in this article:

Why does Red Hat tolerate CentOS? The Community ENTerprise Operating System is an identical binary clone of Red Hat Enterprise Linux (minus the trademarks), compiled from the source code RPMs that Red Hat conveniently provides on its FTP site. It is also completely free, as in beer. CentOS provides no paid support, but it does track Red Hat updates and patches closely, and usually makes them available within a few hours or at most a few days of the upstream provider, which it refers to for legal reasons as “a prominent North American Enterprise Linux vendor.” Free support for CentOS can be found in numerous places around the web, and a few third parties offer modestly priced paid support for those who want it.

It’s easy to understand what CentOS is. The question is, how much business is it really taking away from Red Hat? The answer: probably more than you think. For a hint about what’s going on, check out this amazing comparative chart on Google Trends.

I used to think most CentOS users were either just fooling around and not really running production servers, or else were using it in small IT shops populated by Linux geeks with a do-it-yourself culture. But a recent conversation with a friend set me straight. (I published part of this conversation here.) My friend runs the web site of a big city daily newspaper. Although it’s not in the same league as the Wall Street Journal or the New York Times, it serves tens of millions of page views per month and pumps a big time advertising revenue stream into the coffers of the media conglomerate that owns it. The site is managed by a small in-house team and runs on several dozen dual and quad Intel servers with a classic LAMP stack that includes Apache, Perl, MySQL and of course Linux.

Until fairly recently they ran this web site on an old version of Red Hat with essentially no outside support. But they found that the up-to-date versions of the applications in their stack didn’t run so well on Red Hat 7.3, so they decided they needed to upgrade to something more recent. Naturally the first thing they looked at was RHEL 4 (this started a while back), and then RHEL 5. But they freaked out when they saw Red Hat’s prices. $1,299 per year for 24×7 support on the two socket version of RHEL 5, or $2,499 for the unlimited socket version. True, if you cut back to 12×5 support those prices come down, dropping to $799 and $1499 respectively. But even if they run RHEL on a mix of two and four socket machines, they’re still looking at $50K per year minimum for the privilege of sticking the little red logo on their servers.
Bottom line? They decided to go with CentOS 5, which they are now rolling out to their production servers.

So, is CentOS hurting Red Hat? I don’t think so, in fact I’d say it’s actually helping Red Hat. You see, the company mentioned didn’t balk on Red Hat because of anything but price. That means if they didn’t move over to CentOS, they would have moved to Debian, or Ubuntu, or Suse… or insert some gratis distro or direct Red Hat competitor here. Now, with a CentOS base let’s say the company grows or needs to run an app that is only supported on RHEL/SLES. What distro do you think they are likely to pay for in that scenario.

It goes further than that, though. Many members of the CentOS community are also valued members of the Fedora community. Red Hat gets Open Source at a very fundamental level and sees the benefit of this. Also, with no free version of their enterprise product, CentOS serves as a way to get broader exposure and therefore enables additional gratis testing for many RHEL bits. In the end I see CentOS and RHEL as more complimentary then competitive. I know it doesn’t appear that way at first glance, but when you really look at the market each is serving, it becomes a little more clear.

–jeremy

Filed under Fedora, Linux, Open Source, Red Hat

An open letter to Steve Ballmer from Mandriva

November 1, 2007 1 Comment

The CEO of Mandriva just posted an open letter to Steve Ballmer:

I’m sure we’re way too small for you to know me. You know, we’re one of these tiny Linux company working hard for our place on the market. We produce a Linux Distro, Mandriva Linux. The last edition, Mandriva 2008 was seen as a pretty good version and we’re proud of it. You should give it a spin, I’m sure you’d like it. We also happen to be one of the Linux companies that did not sign an agreement with your company (nobody’s perfect).

We recently closed a deal with the Nigerian Government. Maybe you heard about it, Steve. They were looking for an affordable hardware+software solution for their schools. The initial batch was 17,000 machines. We had a good answer to their need: the Classmate PC from Intel, with a customized Mandriva Linux solution. We presented the solution to the local government, they liked the machine, they liked our system, they liked what we offered them, the fact that it was open, that we could customize it for their country and so on.

Then your people entered the game and the deal got more competitive. I would not say it got dirty, but someone could have said that. They fought and fought the deal, but still the customer was happy to get CMPC and Mandriva.

So we closed the deal, we got the order, we qualified the software, we got the machine shipped. In other word, we did our job. I understand the machine are being delivered right now.

And then, today, we hear from the customer a totally different story: “we shall pay for the Mandriva Software as agreed, but we shall replace it by Windows afterward.”

How is it, you wonder, that a country like Nigeria has the money to not only buy the Mandriva+Classmate solution but also then throw Mandriva away and pay for all those Windows licenses? The answer, I’m sure, is that they don’t. Microsoft is probably marking the license cost to near $0 and may even be tossing in some kicker money (speculation on my part). The thing is, Microsoft has the money to do these kinds of things… and will for some time. Is that really fair competition though? I’d say no, but I’d say it’s also an admission that they can’t win with their old way of doing things. That has to be scary realization for a company that is so used to winning. Kudos to Mandriva for keeping up the fight and for winning deals like this one. I’d guess we won’t be seeing them sign one of the patent covenant deals any time soon.

–jeremy

Filed under Linux, Mandriva, Microsoft, Open Source

BusyBox Developers and Monsoon Multimedia Agree to Dismiss GPL Lawsuit

October 30, 2007 Leave a comment

It’s great to see that the first U.S. GPL lawsuit filed has been settled with fairly little fanfare. From the press release:

The Software Freedom Law Center (SFLC) and Monsoon Multimedia today jointly announced that an agreement has been reached to dismiss the GPL enforcement lawsuit filed by SFLC on behalf of two principal developers of BusyBox.

BusyBox is a lightweight set of standard Unix utilities commonly used in embedded systems and is open source software licensed under the GNU General Public License (GPL) version 2. One of the conditions of the GPL is that re-distributors of BusyBox are required to ensure that each downstream recipient is provided access to the source code of the program. Monsoon Multimedia uses BusyBox in its HAVA TV place-shifting devices.

As a result of the plaintiffs agreeing to dismiss the lawsuit and reinstate Monsoon Multimedia’s rights to distribute BusyBox under the GPL, Monsoon Multimedia has agreed to appoint an Open Source Compliance Officer within its organization to monitor and ensure GPL compliance, to publish the source code for the version of BusyBox it previously distributed on its Web site, and to undertake substantial efforts to notify previous recipients of BusyBox from Monsoon Multimedia of their rights to the software under the GPL. The settlement also includes an undisclosed amount of financial consideration paid by Monsoon Multimedia to the plaintiffs.

“Although we really hated having to ask our attorneys to file a lawsuit to get Monsoon Multimedia to abide by the GPL, we are extremely pleased that they worked so hard and so fast to come into compliance,” said Rob Landley, a developer of BusyBox and a named plaintiff in the lawsuit.

The settlement did include a monetary piece, as the initial speculation indicated. As a whole, the outcome should serve to deter other companies from violating the GPL for fear of real damages being brought against them. Kudos to Monsoon for doing the right thing, but I’d still like to see the product they produce support Linux from a client perspective.

–jeremy

Filed under GPL, Open Source, SFLC

New York Times opens up code

October 29, 2007 Leave a comment

It’s great to see the outcome when someone inside a company truly gets Open Source. From the article:

The New York Times likes open source — so much so that, as it gradually moves more of its print operations online, it is nurturing a Web development team that has released two of its own open source projects.

XSL Cache is a PHP extension the Times is using to cache stylesheets on its Web site. DBSlayer is a tool the team developed to overcome LAMP scaling limitations that caused database replication processes to overwhelm the DB connection limits.

New York Times senior software architects Jacob Harris and Derek Gottfrid say they’ve received a mixed reception from the community, because some people just can’t understand why a print media company would jump feet first into the open source philosophy. But open source software use isn’t new to the Times, says Gottfrid. “I’ve been here a number of years, and open source has always played an integral part in everything we do.”

Gottfrid says the most frequent comment he hears about the New York Times’ foray into open source Web development is “‘That’s surprising.’ They general think of us just as a newspaper, so they don’t understand why we would be doing this. Is there a secret angle? But as soon as they read some posts, they start to break it down and understand. Now there’s a bunch of people interested in this.”

Harris agrees. “We’re no longer just a print company, we’re a technology company. We need to express ourselves in technological terms. The best way to do that is to give the developers a voice.”

While it’s a bit surprising to see a mainstream media company ahead of the curve like this, it’s a clear indication that a couple motivated people can make a huge difference even at a large company. Open Source enables this kind of thing. Kudos to Jacob and Derek for making it happen. I’d guess it was more work internally then they lead on in the article. I’d also guess it will pay large dividends in the end. I found another part of the article interesting:

Harris says he’s received some pushback from members of the open source community. “We went to OSCON, and some people were very excited that we were there, and some were like, what the hell are you doing here. A lot of people feel like we’re pretty much irrelevant, and that they know what our business model should be. When you talk to geeks, they say ‘just stop printing the newspaper,’ but that would pretty much bankrupt us overnight.”

I’ve been meaning to post about the current advertising bubble for a while. It’s outside the normal scope of this blog, but is something I follow fairly closely. This is a public reminder to make that post, but suffice it to say that I do not think the “free but ad supported” model is the panacea some people think it is (although it is indeed the right fit in many situations).

–jeremy

Filed under Open Source

← Older posts

Newer posts →