Striding on open and level playing field

Striding on 'open' and level playing field

Open source veteran Brian Behlendorf speaks his heart out

Saturday, February 16, 2008

Email This

Print This

Comments

RSS

Brian Behlendorf founded CollabNet, a provider of tools and services based on open source methods, with O'Reilly & Associates in July 1999.

Before launching CollabNet, Behlendorf was co-founder and CTO of Organic Online, a Web design and engineering consultancy located in San Francisco.

During his five years at Organic, Behlendorf helped create Internet strategies for dozens of Fortune 500 companies. During that time, he co-founded and contributed heavily to the Apache Web Server Project, co-founded and supported the VRML (Virtual Reality Modeling Language) effort, and assisted several IETF working groups, particularly the HTTP standardization effort.

Before starting Organic, Behlendorf was the first chief engineer at Wired Magazine and later HotWired, one of the first large-scale publishing websites. Behlendorf is currently a director of the Mozilla Foundation and a retired director and president of the Apache Software Foundation.

In a chat with Srinivas R of CyberMedia News, Behlendorf shared his views on Apache foundation, Open Source community and related topics.

After an extended exchange of pleasantries, Behlendorf turned the clock back, giving a whiff of the earlier part of his career.

"I think many of us in the open source industry are kind of accidental leaders. We didn't start our careers thinking that we would be a leader or anything like that. We simply started writing software."

Then, the 34-year-old Behlendorf went on to elaborate. "Often what people know me best for is what I did when I was 20 years old. Being in the right place at the right time with regards to Apache. I think people get to know the kind of themes that I have been pushing for open source, for a very long time. Open source is almost boring now because it is so well accepted."

That sparked the interaction.

Correct, but open source is accepted in the Web world. What about in the enterprises? How is it happening?

To say the web world as if it was some separate thing from the rest of IT is starting to disappear. I am sure you still have public websites that are the homepage for a company.

But everything that an IT-driven company does today is being done through web-based interfaces; it's almost everything. It used to be the client-server world. People are starting to realize that the best hop doesn't have to matter anymore.

The operating system you run locally in your client… you can run Linux and be just as capable in a web-based enterprise as anybody running Windows because you can use your web-based interfaces for mailing, for calendaring you can use open office when you need to edit office documents, etc. and when you need to access the enterprise applications, they are all just web pages.

Web services means that the internal enterprises can be web-oriented and open source software is in the first and in many cases implements those web-based interfaces better than the commercial adds to.

When it comes to specific packages like Oracle Financials, there isn't an open source program I can tell you that will drop in there and a drop-in replacement for something like that or for SAP.

But there are these little, what I would call mammals in the age of dinosaurs, right, that is scurrying between the footsteps of …kind of as SAP and Oracle do battle each other, and there are projects like open ERP.

There are some companies now that are starting to form around these ERP systems or closed systems; analytic software, starting to go heavily on the open source direction, and we are talking about companies with venture funding, with real revenue, with not just a couple of customers but hundreds of thousands of support-fee paying customers, and that's significant, that's something that will grow.

Linux did not take off in the desktop space. So, what is the problem there?

One of them is that Microsoft is not fierce enough about intellectual property. Believe it or not, you know there is lots of Linux boxes sold and it is kind of a two-minute Linux where you know machine is brought in and before it is even booted up, you are installing an illegal copy of Windows on that.

And Microsoft could do more today to stop the illegal use of Windows, but in many countries it is simply rampant and Microsoft, I believe, actually depends upon illegal use of Windows software.

So, one of the problems is that Microsoft tolerates too much piracy. The second is that there is inertia. Now, every time Microsoft forces a change like from XP to Vista, that's an opportunity and Vista has been one of the best things for desktop Linux, because if you take a standard PC and upgrade, you will notice a significant slowdown. Unless you sit there and spend hours tuning it right, you will notice a slowdown.

Well, enterprises are hesitating to move to it. It still sells fewer copies than XP does. So, that has been a big opportunity and it's a good time right now because the Mandriva and Red Hat and the tier 1 Linux distributions are finally ready for enterprises.

The tools are there and there are companies buying those tools now to support them as well. So, there is really no excuse except laziness and inertia, but companies can earn tremendous amount of money if they are going to Linux desktop and many of them are.

Do you think Linux needs some kind of a support from the large companies?

I think that certainly helps. I think 10 years ago, it was reasonable to say, "Well, I am not going to use Linux because there is lack of support. I am not going to use Apache because there is a lack of that."

But today, there is no excuse and you have these big organizations that have been saying, 'we use it' and many saying, "Hey, I will sell your support contract now; so your needs are met."

So, I don't think that's a missing piece right now because I think we have that. I don't think we will ever see Microsoft officially support us but even if they do, I think it will move a lot of people over.

I think it is when people are ready for change, when they see something like Vista and go, "Wow, that takes away a lot of the power that I just spend a lot of money acquiring to buy a high-power laptop."

So, I think we need more killer apps. That might be the missing thing. There hasn't been a company that has come out with killer apps since Netscape.

And what do you say about browser war IE and FireFox?

I am at the board of the Mozilla and we believe FireFox is so much better than IE that it is one of those kinds of killer apps.

You can use FireFox on Windows. There is no doubt about that. You can use it on demand. We are very happy that people are using it. We have no interest in turning your eyeballs into cash or anything like that.

We believe in the sovereignty of the individual and as a non-profit organization, we are chartered with making sure what we create is a public asset and we think that's going to be pretty important to people as well.

I think people are realizing that even as consumers there is a role they can play in improving their environment, and improving the software they use.

So, what's your strategy for the Mozilla Foundation?

Well, the foundation's primary strategy is to make sure we are still in business. Sustainability is a big topic out there. We are non-profit but we have always been concerned about depending upon charity.

When we initially started out, we asked for donations from large companies. And, we were very nervous about that because one year maybe you are the media darling and everyone loves you and the next year you suddenly become very unstable.

What we realized is that the user base was so broad. There are tens and millions of people using it today, probably over a hundred million actually.

It's about 15-20 per cent in the US and about 40 per cent in Europe, and that traffic can mean something. You can build affiliate relationships with organizations to fulfill that, and so when you select one of those drop-down engines and you search on Amazon and that search leads to a transaction, then Amazon gives us a little bit of credit.

When you do a search on Google, there is a little bit of credit that comes to us. So, all of these drive revenue to the Mozilla foundation.

We actually created a for-profit city area because of the way that tax law works in a way that foundations exist to have a charitable purpose and draw their income from donations.

In the long run, we would like to find other ways to bring in revenue. May be it is through subscription program or through raising more donations, may be getting grants; but the reality is that to write a high-quality web browser and compete costs real money.

Like browser, are you looking at bringing any other applications? Well, we did launch a mail client Thunderbird and we created a for-profit organization just like we did with the browser around the mail in an organization.

We will be launching that a bit more formally in the next couple of weeks, but this is a technology that today has 15 million users. So, we will see where the mail client goes, probably goes into related areas like calandering and task management and personal information management mechanism, which also suggests that this is kind of a mobile strategy that is needed as well, but there is so much good stuff happening out there in the open source and you are familiar with the Android, the Google mobile effort.

We don't want to duplicate effort when we don't have to. We want to build on top of what is built. We are not anti-company. We are not anti-capitalist. We are very hard-core capitalists actually. We just believe in competing on a level playing field. And we are here to help keep the playing field level.

So, that means you are looking at the mobile field as well.

Yeah, and we haven't really articulated a big vision for mobile. We have FireFox 3.0 running now in a kind of mobile environment, stripped down, lightweight, and we are very enthusiastic about that.

But we see efforts like Android and we see efforts like other Linux in the phone kind of effort. We want to build on top of that. We don't want to replace those efforts.

Coming to CollabNet, could you just explain what exactly is it?

What we build is a set of collaboration tools designed after how open source communities work, but scaled up to meet the needs of enterprises.

So, we go to companies like Barclay's, Sun Microsystems, Oracle, and Intel, and we create these collaboration infrastructures for them where they pay a per-month per-user fee.

We even do some public projects like Open Office project. So, we are the plumbers of collaboration, and we have 400 major companies now as our customers about a million total users.

Now, we have another - something called Subversion. Subversion, we give away for free when we sell this support services and customization and those kinds of things; and when companies deploy it for 200 people and they need someone to help them tune their servers or to come in and teach them.

So we do a lot of training as well. So, Subversion is deployed across 200,000 public servers out there. We can go and there is a survey that is done once a month to count the number of Subversion-based servers out there and repositories.

And for each server, you have between 5 and 10 users. So, that suggests somewhere two and four million users out there of Subversion, the open source client.

You came to India through the acquisition. So, are you looking at more acquisitions?

Not at this time. We feel we have a pretty full product plate. In addition to Subversion and in addition to SoruceForge, which is our Oprah kind of suite with different tools that work together in concert, we also have a new tool called Qbit, which is a virtualization platform for deploying bills and test systems.

We are a 200-person company, about 100 in Chennai and then another 100-110 around the world.

Does consultancy still interest you?

One of the things that I learned at Organic is that running a shop based on timing materials is a tough business. We certainly do consulting in support of our products.

When a company comes and says, "We need somebody come in to teach us how to use these tools, to teach us how to change our development processes,"

Sometimes there is no other option but to go in and say, "Okay, I will stand three weeks on site with you and your team and whatever we get accomplished in those three weeks will do."

So, we do a very healthy amount of consulting along those lines, but we would be very happy to partner on that kind of work too and we have partnered with our work.

We have also partnered with organizations where there were improvements to our product or improvements to the open source Subversion product.

And most importantly, we know that there is a very large industry out there built around consulting.

You are involved in Apache from the beginning. So, what does the future hold for Apache?

We were always built on the idea of communities writing code rather than one single person kind of doing everything. It's about 350 people now.

I think that's large. I am still a member. I still follow some of the activities. I still attend some of the conferences, but I have largely left the direction to others and as an outsider now.

Apache seems very content to focus on building useful modules of technology. One of the more interesting projects recently (we have about 50 projects total) has been Hadoop, which is a technology that actually has come out of Yahoo for doing search in a new way.

So, I think Apache will continue in the vein of, let's build tightly defined project with a community that is large enough that they know each other by first name. It doesn't have to be something as large as FireFox.

Apache does not pay any developers. It requires volunteer effort and that does act as a set of constraints on what's possible, and so it's likely that that will remain small interesting modules that become the best of class in their field.

Hopefully, it can become the standard for doing whatever it is that they do, just the way that the web server became the standard web server, right? But, I think that's the direction.

(srinivasr@cybermedia.co.in)

©CyberMedia News

Email This

Print This

Comments

RSS

Interact, Collaborate & Share

With Lotus Quickr your everyday business content can be easily shared, enabling operational efficiency. Download the Lotus Quickr demo to learn how.

know more..

	Save 40% energy on data solutions
	Try visual studio 2008. It’s Free
	Complexicity or Simplicity - Choose

	Motorola : Good Mobile Messaging
	Smart Enterprises = Smart CIOs
	TAM : Freedom from password management

	Mahindra Connect benefits from SharePoint Server 2007
	Solving the "lost data" problem
	Wipro meets its future demands through E-Cube

	HP Business Availability Center
	Transforming Business
	Aligning IT and Business