Is Open Source a Growing Market or a Temporary Trend? Are successful open source vendors simply becoming proprietary software vendors with open source community products? No. Open source means more than open source code base and the business models for the companies and communities surrounding various projects have become very sophisticated and responsive to customer needs. The future is open source development for all software vendors. One way to understand this future is to review open source as a marketplace:
A market can be defined at a minimum by a definable Need, Customer's with money, Vendors with products, and an acceptable Business Model. Let's review each of these for open source:
1.) Open Source Code Base
- Allowing customers access to the source allows them to more easily maintain and customize their software. Any major enterprise that is buying a software product - will want access to the full open source, so they can understand the inner workings when customizing or developing extensions, so they can support themselves long term or in emergencies, etc. In the past companies (especially governments) would require that the source of the product they purchased would be put into escrow to ensure they would have access to the source if the source company ever went out of business. However, that only let you have the source if you needed it - not to allow introspection earlier.
- BIG ONE: Most of the major (very expensive) vendors all have the same features, thus they are spending hundreds of millions of development dollars on rebuilding the same commodity aspects. Given there is only a set number of dollars for development in the industry, commodity/core features, once complete, should be made open source, and shared by all, which forces software vendors to innovate on top of those core features. Instead major vendors have looked to add features that either lock cusotmers into their platforms or push product-chaining strategies. In 2010, why are we paying $300 for a desktop operating system, $500 for office software, $Millions for an App Server? It may be that we are 20 years behind where we would have been in software technology had it not been for the majority of investment dollars and strategy being pent up by major proprietary vendors.
2.) Lower True Cost of Ownership
- Open source is lower cost. I have spoken to CIO's who have read the IBM article proposing that open source while having lower licensing costs is more expensive for long term cost of support. However, the opposite is true. Open source vendors have a business model which allows software developed in the open, to be supported the same as proprietary software, i.e., there is no need to have on staff, open source developers who know how to find patches, compile them, etc., instead customers with support from open source vendors get product bulletins/alerts, phone/email/web support, customer portals, hot patches, service packs, etc. Lower cost to start - lower cost to support. Add this to the fact that customers can often evaluate the community versions of the product before the project is funded or even during the initial development phase even lowers the need for long term procurement steps. Why work 6 months on an RFP - when you could download the software and try it out?
3.) Lighter Weight Solutions
- "Lighter Weight" means many things. One of the big demand enterprises want for software is to simplify it's installation, configuration, scaling, management, developer experience, etc. Open source software often leads in each of these areas simply because, if they did not, they would not have been successful in open source against competing open solutions. Open source development forces products to be developed in a way that they can spread virally - meeting multiple the demands of the enterprise. In open source - it's not the "strong" who survive (those with vendor lock in, massive sales power, or monopolisitc position), it's the innovative, the adaptive ones who become successful. Open source is forced to be responsive to the consumer and one of the main directions for open source is lighter weight.
4.) Full Industry and Defacto Standards Support
- Open source must also support industry, defacto standards and simplify integration/interoperability or again they will not grow in popularity. This allows customers a broader capability to fight vendor lock in. Open source often must fit within a heterogenous environment - to work with many other products already leveraged within the enteprise. Major vendors instead work to drive an exclusive platform stack. While the sales pitch is that a homogeneous stack from one vendor simplifies learning cycles, manageability, interoperability and lowers cost - they instead are comprised of usually multiple different assets acquired by the single vendor, may have a single management stack but require integration for each, and are exponentially expensive as the customers have less ability to negotiate lower prices for the additional components. That is, software suites are often more expensive than the individual products becuase the customer loses the ability to negotiate on each component. Vendors also design aggressively to lock customers onto their platform to ensure long term support revenue and to ensure a foothold for other product sales. I still remember a major vendor who purposely slowed down the JSR 168 spec so that once the other industry partners couldn't wait any longer for the spec and decided to release this first iteration, this vendor immediately released their support for the spec but added extensions allowing them to claim they had better support for demand than others who only supported the spec. They purposely manipulated the spec to ensure their market position. While standards are a valuable aspect, they are often slowed due to major vendors desire to leverage their platforms.
- Defacto standards, those that form in open source and become standards by wide adoption allow better development for innovations. Many vendors can build standards and those that suite customer needs the best will result as leaders - and then may additionally become industry standards. Thus successful open source vendors, who are driven to support industry standards and defacto standards, protect customers as well as provide them access to the leading innovations..
5.) Stop Product Chaining
- As described above, open source must follow market demand and is less able to focus on vendor lock in for product chaining. Product chaining is a common product development tactic where a vendor builds a new product which requires use of one of their other products, or which supports industry open standards but allows more features when using their closed standard. One easy example is a popular Email program that supports IMAP but provides many more capabilities if used with their proprietary email server. Open source by its nature, cannot force product chaining - or it would never have grown to success. Whatever excuse you use to justify the use of the products in the chain ("oh but their backend email system is really good") it still leaves a bad taste in the buyers mouth and ultimately lowers desire for future products from the same vendor.
Vendors and Customers with Money
Multiple open source projects are seeing double digit growth and expanded market penetration. I do not believe this trend is simply because companies are looking to reduce costs due to the troubled economies worldwide. Enterprises have continued to need to "do more with less". As systems are implemented, they must be maintained and often grown, while at the same time there is continual new demand for new services. While the major software vendors have worked at customer lock-in their product functionality has been fully replicated in open source stacks or 100% of the base functionality needed are available and customers are opting for just the basics. Enterprises will increasingly look to move commodity functions to open source platforms to reduce long term costs which allows incremental funding to be leveraged for new projects.
There are many business models being leveraged in the monetization of open source and facilitating their growth in popularity. These include viral growth through community editions, multiple revenue streams (EE versions, Add on Products, Services, Solution Sets, OEM, etc.), product support models (forums / email / phone support, customer portals, product bulletins, hot patches, service packs, etc.), services departments (installation/confiuration, custom development, product enhancement requests, etc.), and marketing/sales departments (community advocacy, inbound marketing, paid search, etc.). The models are very sophisticated in 2010 and are simply an adaptation of standard (tried and proven) business practices.
While the statements above detail how proprietary vendor solutions have been replicated by lower cost open source, that enterprises are moving these commodity functions over to open source vendors, the interesting twist that has occurred in open source is the development of innovations within open source. Even though open source is often focused on commodity features - the leading open source vendors are additionally outpacing proprietary vendors with new innovations (SpringFramework, Liferay AlloyUI, MySQL Query Analyzer, Firefox Add ons, etc.). The reasons are many but notably due to the nature of open source, that instead of 400 engineers from a major vendor, these successful open source projects are collaboratively built (or defined) by tens of thousands of free-for-now users, enterprise cusomers, partners who all have full access to the soure code. The innovations of thousands will always outpace the [often self concerned] development of hundreds.
Given that demand is growing exponentially, that the business models are very sophisticated, it is natural for a broad marketplace to be developed. Some quick examples of open source marketplaces are: sourceforge, http://bitnami.org/stacks, apachefriends.org. These are each marketplaces for the collaboration, development, download, etc. for a broad range of technologies. Additionally, however, major vendors are developing their own marketplaces, like http://www.liferay.com. Individuals can find technical resources, download code, collaborate with others, while partners can advertise themselves, and community members can contribute. Many open source vendors will develop and expand fully encompassing eco-systems. Open source, is a broad eco-system of marketplaces - not one - and definitely not none.