Wednesday, January 24, 2007

Firefox must cross-sell other Open Source desktop apps

Firefox is arguably the most successful Open Source application ever. At the current time, it is estimated that about 15-20% of all desktops worldwide run Firefox. That's an impressive market share. In absolute terms, that means Firefox runs on a couple of hundred million computers, beating even Apache, which has a 58% share of its own market segment.

Other Open Source desktop applications like OpenOffice.org and the GIMP haven't had that level of market penetration, even though they offer excellent quality and value. The main reason is lack of awareness. How many Windows users have heard of these products or have been tempted to try them out? Some advertising and sales promotion are urgently called for.

I would suggest that the Open Source community evolve a common strategy to popularise its products. We can't afford to be disjointed and tactical, fighting battles on a segment by segment basis.

Specifically, our most successful product, Firefox, must advertise and cross-sell OpenOffice.org and the GIMP (or GIMPShop, its Photoshop-like variant). Firefox already cross-sells its sister product Thunderbird, but as a member of the Open Source community, the Mozilla Foundation needs to be more farsighted and (may I say) less parochial.

The page that appears whenever a new version of Firefox is installed should do more than carry version information. It must educate the user with copy like this:
"Congratulations! You have installed the latest version of the Open Source Firefox browser. The features of this version are ...
Would you also like to download these other excellent Open Source applications?

Thunderbird brings you safe, fast, and easy email, with intelligent spam filters, quick message search, and customizable views. Download here

OpenOffice.org is a fully functional office suite with a word processor, spreadsheet, presentation software, drawing tool and personal database. It's designed to be compatible with Microsoft Office and is 100% free! Download here

GIMPShop is an Open Source graphics tool that provides most of the features of Adobe PhotoShop at no cost! Download here"

Winning market share on the Windows platform is a tactical victory, at best. It's as secure in the long term as resting on a crocodile's back. Ultimate victory will only come when the Windows platform is itself replaced by Linux and other Open Source operating systems. The groundwork for that is to get users familiar with what's called "transitional software" that runs on both Windows and Linux.

Imagine a world in which 25% or more of all desktops run Firefox, Thunderbird, OpenOffice.org and GIMP/GIMPShop. The obstacles to desktop Linux will be drastically reduced when users are already familiar with the major applications on that platform.

I've sent in this suggestion to the Mozilla Foundation. Now the ball's in their court.

Saturday, January 06, 2007

Free Java - I told you so :-)

I predicted this a year and a half ago. Java was going to be Open Source. However, I was both overoptimistic and not optimistic enough. I thought it would happen in 2005 itself, but we've had to wait until 2007 to see the first dribbles of Sun's code. But I never dared to dream that Sun would choose the GPL as the license for Java. I'm still speechless. Yes, someone I know suggested this as far back as 2004, but to see it actually take place is a pleasant shock.

What do I think is the likely impact of an Open Source Java? Well, what happens when a logjam is finally cleared?

Here's the technology stack we have today ("HILTS"):









Spring
Hibernate
Tomcat
Java
Ingres
Linux


Java has been the inconvenient piece in the middle preventing the whole stack from being freely distributable. What Sun has done with its latest announcement is effectively make the whole stack free, not just Java. Now we'll start to see Linux distributions that set up complete application platforms as part of their installation. This will significantly lower the cost of adoption of Open Source.

OLPC - What its greatest impact will be

You must have heard of the One Laptop Per Child project. It's an ambitious, even audacious, attempt to raise development standards of poor nations by putting inexpensive computing power into the hands of the next generation of its citizens -- its children. I guess I shouldn't be surprised at this initiative. As someone I know wrote 5 years ago, "Open Source is doing what god, government and market have failed to do. It is putting powerful technology within the reach of cash-poor but idea-rich people." The only way OLPC has let down that prophecy is the fact that governments are involved.

To date, governments of the following countries have signed up: Argentina, Brazil, Libya, Nigeria, Rwanda, Thailand and Uruguay. India, to my surprise, has not. India reportedly thinks the scheme is "pedagogically suspect".

I don't know how the project will turn out. I wish it all success, of course. As in all projects, execution will be key to its success or failure.

I would normally dismiss heavy government intervention as being ineffective and a waste of money, but for the success of India's (note the irony here) mid-day meal scheme for schoolchildren. Not only does that scheme provide much-needed nourishment to millions of poor children, it has taken the burden of feeding them off the shoulders of their parents, so the children are encouraged to go to school rather than go to work to feed themselves and their families. School dropout rates have fallen as a result, and "universal primary education" is no longer a mere slogan.

Will the OLPC similarly succeed in raising educational standards in the developing world?

I believe that whatever its track record in that area, OLPC is going to have a major impact on the world for one simple reason. It shows that laptops can be produced for roughly a hundred US dollars each ($135 today, but expected to fall as production ramps up and hardware prices continue to fall). Today, you or I cannot buy an OLPC laptop, because they are only made available to governments for distribution to schoolchildren. However, containing the market impact of such a large influx of cheap laptops in an economy will be like holding back a stream with one's bare hands. The price of laptops in general will experience severe downward pressure in all the countries that OLPC touches.

I cannot overemphasise the beneficial effects of this development. Progress is not so much about invention as about democratisation. Many modern inventions were known even in the nineteenth century (e.g., the automobile). But it was only in the twentieth and twentyfirst centuries that these inventions were mass-produced and made affordable to all. Democratisation achieved much more than mere invention.

OLPC may or may not achieve its ambitious goals of lifting developing nations out of their educational quagmire. But I think it will indirectly democratise computing power. I'm eagerly waiting for that to happen. Watch the fun!