Google shifts software value to operations, away from IP
Google CEO Eric Schmidt argues that competition is just a "click away." By opening up Google's access to personal data as well as the software that collects it, Google is adding substance to that claim.
Google's secret sauce? Operations.
It won't work to write better software than Google, because Google is not a software company. The only way to beat Google is through superior execution; through better operations.
The way to lose to Google? That's easy: try to sell software that Google is shoveling out the door as free (open source) software.
This thought hit me last week as Google announced the open-sourcing of its Closure tools. These are JavaScript compression tools that Google uses to build its own Web services like Gmail. Basically, Google was saying, "Here's how we build our software services. You can, too."
This follows on the heels of Google open-sourcing other software like Android ("Here's a free mobile operating system to help more people connect to our services"), Gears ("Here's how to run Google-like services offline"), and more, each opening up windows into the software that runs Google's services.
In many ways, Google is giving away the recipe to those that would like to build a Google clone.
The problem? Google is so much more than software.
In fact, one of the primary reasons that Google can write and open-source so much software is that it isn't a software company. Not even remotely. I could have every line of Google's software, both open source and proprietary, and I couldn't hope to compete with Google.
Google is what Google does with the software, and not the software itself.
Ditto for Red Hat. The company used to retain significant chunks of proprietary code in its Red Hat Network offering, but it has released that software under and open-source license in the last year. Doing so hasn't had any effect--positive or negative--on Red Hat's sales, because Red Hat isn't in the business of selling software anymore.
Red Hat, like Google, is in the business of providing services to customers, services enabled by software but which are much more dependent on IT operations and overall efficiency of execution.
And while both companies rely on open-source software to fuel those operations, Google, more than Red Hat, realizes that the conversation has moved on from open-source licenses to higher-order value, a theme that is going mainstream. Cloudera CEO Mike Olson captures this theme well:
The license terms attached to products have become secondary to the value it offers. People now are much more rational about how they adopt technology across the board. Open source is a detail, not a defining characteristic. At Sleepy Cat, we were proud to be an open source company. At Cloudera, I think of us as an enterprise software company that happens to be built on open source software.
Such sentiment would resonate well within the walls of Google's Mountain View headquarters, I suspect.
Even within their respective open-source communities, neither Google nor Red Hat company is particularly saintly. Red Hat has never waited on the Linux Standards Base or industry efforts to coordinate Linux distributions (like United Linux), but instead forged ahead with its own efforts...until the LSB simply adopted Red Hat's distribution as the standard.
Google, for its part, takes flak for dominating its open-source project communities like Android. Google's contributions (or lack thereof) to outside projects, like Linux, don't always mesh well with others in the industry.
To open-source community critics of Red Hat and Google, some advice: get over it.
The companies that spend the most time chumming around, talking up interoperability and the need for everyone to work together are usually the ones losing the race, a race whose rules may be written in software but whose victory depends upon execution.
Google and Red Hat have moved beyond software. Software enables their operations, but software doesn't define such operations. Google, for its part, is open sourcing Microsoft, one line of code at a time, and Microsoft hasn't a clue as to how to respond, because it only knows the old world: competition through better IP.
That world is gone. Open source has killed it. The new world is built on execution and superior IT operations. Google gets this. Red Hat gets this. Microsoft? Not so much.
Matt Asay brings a decade of in-the-trenches open-source business and legal experience to The Open Road, with an emphasis on emerging open-source business strategies and opportunities. Matt is vice president of business development at Alfresco, a company that develops open-source software for content management. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. You can follow Matt on Twitter @mjasay. 
RedHat and Google are _both_ building products - but their secret to success is, they don't stop there, and the software isn't some magical but vulnerable linchpin holding their companies together.
Most of the software RedHat provides wasn't written by RedHat employees. Hell, a huge chunk of it was written by RedHat's direct competitors (Novell, IBM, Oracle, et al).
==
"They {Google} stand on one pillar. If that pillar should ever weaken, down comes all the eggs they have built on open source."
Ah, but you miss something - Google is using that open source software to build multiple pillars, on multiple foundations. I could just as easily say that Microsoft stands on one pillar - selling Windows. But, that would be equally false (OTOH, Microsoft does have all of its pillars built on just one foundation - the hoarding, licensing, and keeping of intellectual property).
Long story short: If Google's entire algorithm collection and AdWords customer base list were given away open-source, it would still survive quite nicely. If Microsoft's entire codebase were given away as open-source, Microsoft would die in very short order.
Google will die very quickly if the algorithms behind the search engine and also AdWords is given away. After all, just like Microsoft, Google is a one-trick-pony (all their profits comes from search advertising). Then again, Microsoft still has Office, so they are a two-tricks-pony then."
Well, tell me how you can compete with Google's servers that can handle terabytes of data daily? Google is more than software, they have the infrastructure too. Even if you have Google's source code of all their products, it would take forever to implement them. Actually, it would take also a lot of work to build a user base.
I could be wrong but I would now rate a company by the quality of its manpower and infrastructures which leads to the quality and scalability of their services, not by looking at how many million lines of good code/algorithms they have written.
The thing, however, is that Google is a about search, and search is a very big tick or, if you want, a very long pony. Microsoft, the same thing.
Those who talk about another couple of guys in a garage beating Google, simply do not get it. It is true, you do need some guys in a garage, but you also need a huge technology shift, something on the order of going from mainframes to PC-s (Apple, Microsoft), or moving from the desktop model to cloud-based services (Google).
For internet-based services, the battle long ago shifted from products to services, and, as a result, from IP to operations. Lately, there is a further shift occurring from operations to data - what matters now is how much and how comprehensive is the data you are able to collect about your customers / trends etc. and how good you are at monetizing it. Since Google is in this business, it makes sense for them to not worry about their IP (but keep their data under lock and key).
There are other areas, where products are more important than services or customer analytics, and IP is still quite valuable in those.
They may also be governments that want to know the docs they're creating today will be accessible tomorrow.
We've seen the opposite of interoperability for years. Big companies that keep changing their file formats simply to force unnecessary upgrades and the like.
I think it's a mistake to associate seeking interoperability as a goal with weakness.
I agree. Google is open sourcing Microsoft one line of code at a time. It is a brilliant strategy.
Microsoft cannot respond competitively and Google make billions from free software, so they are in the better position in an open and free software world.
This is how you beat an incumbent, you do what they do, but do it for free and in the process you change the paradigm toward your strengths.
The ironic thing here is that this strategy is how Microsoft beat out Netscape. They gave away their browser for free and bundled it. I think I speak for Google when I say, "thanks for the idea Microsoft, it really works".
Open-sourcing code accelerates adoption and helps to build a community around the technology. It is a good way to fight an entrenched competitor indirectly, by lowering the barriers for entry for many smaller players (Chrome vs IE, Android vs WinMo). It is a good way to provide interoperability with your services already heavily invested in the technology being open-sourced (Closure, Protocol buffers, GWT).
Open-sourcing all your code-base is not a solution. It can be a security risk, a maintenance nightmare, or an IP flop. It is true that a company like Google is not just about code, but it would be wrong to dismiss the code and all the IP behind it as unnecessary for the company to stay ahead of its competitors. Companies like Yahoo and Microsoft have a lot of resources, yet they are behind Google in terms of their search-engine and ad-platform implementations. Why discount these advantages?
Open-source is just a tool. You are not guaranteed to win by using it, but it makes sense to use it when it helps to get to your objective quicker.
I agree. Google is open sourcing Microsoft one line of code at a time. It is a brilliant strategy.
Microsoft cannot respond competitively and Google make billions from free software, so they are in the better position in an open and free software world.
This is how you beat an incumbent, you do what they do, but do it for free and in the process you change the paradigm toward your strengths.
The ironic thing here is that this strategy is how Microsoft beat out Netscape. They gave away their browser for free and bundled it. I think I speak for Google when I say, "thanks for the idea Microsoft, it really works".
- by vmihaylov November 10, 2009 3:23 AM PST
- Totally disagree. Google is a software company more than any other I know. Take for example closure javascript tools. But look at it from a different angle. There are plenty javascript tools and frameworks around. YUI, JQuery and so on. But none of these attacked the problem so deeply and thoroughly as closure. Others look like toys compared to closure. And by the way they are open sourcing it years after they used it internally. As others mentioned search algorithms, I would ask why don't they opensource this or the software that manages their infrastructure, gfs and big table. If they do it eventually you'll see that hadoop will also look like a toy.
- Like this Reply to this comment
-
(18 Comments)