 |
 |
 |
| ARTICLES | September 2007 |
Six myths about open source
Open-source RDBMS supplier Ingres puts the case for open source by dispelling some of the myths that have built up and pointing to the increasing number of products now in mainstream use
IN EVERY poker game there comes a moment of truth, when one opponent calls another’s bluff. Everyone lays down their cards and a winner is declared based on the cards they hold in their possession.
Well, we think it’s time now to call the bluff on closed-source vendors, which have long argued that open source is a niche IT option, or its unable to support mission critical applications and any update to the code is unregulated. Some even propose open source does not come with professional support and try to spread fear, uncertainty and doubt (FUD) around who owns the intellectual property rights over the code.
With the plethora of myths purported by the closed-source camp, Ingres reveals the truth behind today’s open source software and dispels the FUD being spread by closed-source vendors.
1. Open source is a niche IT sector
Maybe this was a fair comment when Linux and Apache were in their infancy, but many years later open source has established a firm foothold in the data center and this statement could not be further from the truth. The market opportunity for open source is huge and it is no longer the case that open source is just a viable alternative to closed-source vendors’ offerings; the savvy CIO knows that in many cases, open-source software offers superior performance, flexibility, security and functionality to closed-source software.
Last November, Gartner wrote: “Open source is entering the mainstream”1, while IDC noted in July that, “open-source software represents the most significant, all-encompassing and long-term trend that the industry has seen since the invention of the fundamental data storage architectures and SQL APIs in the early 1980s.”2
There are many well-established, open-source products and services on the market: most IT directors are as familiar with names such as Red Hat Linux, MySQL and Apache as they are with Windows and Internet Explorer. Likewise, no longer is open source taken to mean the same as freeware or shareware. The risk-conscious IT decision maker knows that open source means that they can use, study, modify, enhance and redistribute the product and source code with the assurance that there is a wide array of support and services available to support mission critical deployments.
But forgetting the analysts, industry watchers and open-source enthusiasts, when Microsoft makes a very public move into a market, you know it is here to stay. The Microsoft/Novell announcement—which will see the former paying $348 million to Novell to enable its open-source Linux software to work with Windows—generated seismic waves throughout the open- and closed-source communities. Ingres has argued for some time that commercial companies will become the next constituency to enter the open-source world. Ultimately, we believe this move will accelerate the adoption and, therefore, the success of open source within the commercial sector.
2. Open source cannot support mission-critical applications
Many opponents to open source fuel the idea that open source is not reliable enough to run mission-critical applications and that the quality of open source products is poor.
Yet open-source products are subjected to the same levels of performance, stress, functional, security and regression tests that closed source products are. In addition, an advantage of the open source development model is that it enables collaboration with the end user very early in the development cycle. End users can participate in the design and development process, and can test the new features as soon as they are available in community editions of the product. As a result, bugs and design flaws are identified early in the development lifecycle and are fixed faster and users benefit from a high quality product specifically targeted at their needs.
There are numerous examples of professionally developed, managed and supported open-source projects, which have a proven track record in mission-critical deployments. One only has to look at Linux, JBoss and Ingres as providers of such solutions.
What’s more, customers appear to agree with the open-source providers, often reporting—sometimes contrary to their initial scepticism—that open source has driven performance improvements by delivering applications faster and with less downtime than closed source products.
Look no further than the public sector which is championing open source adoption with Gartner reporting on France’s Association of Developers and Users of Open Source Software in Administrations and Local Communities3, and the Malaysian Public Sector Open Source Software Master Plan4, as two such examples. Additionally, the same Gartner analyst commented in August 2006: “Government attitudes toward open-source software are evolving toward a more sensible approach, aiming to leverage the benefits that open source can offer as both a licensing and community model.”5
3. Open source companies do not own their intellectual property
There is a misconception that open-source companies do not own their own intellectual property (IP) and, therefore, cannot closely guard their IP in the same way that a closed-sourced company can. Some opponents of open source argue that because it has many different contributors on a single project, development has the potential to introduce legal snags that customers do not want to deal with.
The truth is that open-source software is subject to the same copyright laws as closed source software, but open source software suppliers choose to share their IP with others. As a result of this, open-source software cannot be monopolised, no one organisation can control the price for support and services for open source solutions and the competition to provide superlative support services at an attractive price-point makes it more cost-effective for customers.
Intellectual property has always been a key driver of innovation in the software industry. The open-source development model allows for continuous innovation in a truly collaborative environment through the sharing of IP, meaning products are brought to market faster.
4. Open source technology does not offer professional level support
This myth keeps many companies away from open source, but it is useful to distinguish between the community that surrounds and supports many open-source projects and the support that customers purchase from open-source software providers. Such providers offer professional support and make it possible for the software to run mission-critical applications for major, global companies everyday. And there has been a lot of discussion around this area recently, following Oracle’s announcement in October 2006 that it would provide the same support for its Unbreakable Linux product—derived from Red Hat’s Linux technology—as it provides for its other products.
While it is always good to hear that vendors are supporting Linux as well as other operating systems, it is our belief that Oracle will ultimately fail in its attempt to corner the Linux support market since it is unlikely that Oracle will be able to create a support solution that rivals the Red Hat Network, or build the institutional expertise that Red Hat has built since its inception. We also struggle to see the value in using the open-source ecosystem to contract the market, rather than build something new and release it to the open source community.
But what the Oracle/Red Hat example does highlight clearly is the importance of providing superlative professional support for organisations selecting open-source software. Indeed, the business model for most open-source software providers depends on customers buying support and services. Therefore, the support and services offered by open-source software vendors must equal or better those offered by closed-source vendors. If it is not, open source will not be successful and competitive in the marketplace.
5. Open source is unregulated and anyone can contribute code
Some opponents of open source propagate the myth that anyone can access and change open-source code, which makes it unsecured and unreliable.
Yet the truth is that access to open-source code is controlled, and any changes to the source must either address a problem, or enhance the product. Source code changes go through rigorous peer review as well as acceptance and regression testing and so, in the open source meritocracy, a developer must earn the right to submit code to a project.
It is also worth noting that although millions of developers build on top of or around open-source software, the number of contributors to the core projects is only a fraction of this. Open-source companies hire professional developers to enhance the product, fix bugs and ensure that a high quality product is delivered to their user community. Open-source development is a meritocracy and subject to the same rigorous software development methodologies as closed source development projects are.
6. Open Source software is not secure
A common misunderstanding is that open-source software is more vulnerable to exploitation than closed-source software simply because code that is exposed is more likely to be hacked. Open source software is secure by design and uses standard software development methodologies and secure coding techniques. All source code changes are subjected to rigorous peer review before acceptance. Because of this review, more proactive checking for vulnerabilities occurs, bringing to light any exploits in the code and providing built-in security from the ground up.
It has actually been shown that the number of vulnerabilities uncovered in closed-source software is significantly higher than that of open-source products. The Linux Kernel has been found to contain one per cent of the bugs that a typical closed source product would contain.6
Summary
The open-source market has evolved and grown in leaps and bounds: indeed, it has reached the point where many customers do not see a software purchase as being a choice between open and closed, but a question of which product meets their needs, will deliver the best performance and receive the best support. This realisation that open-source is not only a viable option for large enterprises and small businesses alike, but a real threat to their bottom line, has spurned a partnership and acquisition frenzy amongst the closed-source community, including the Oracle-SleepyCat acquisition and Microsoft-Novell joint venture, as they try to buy themselves into the market.
However, what the closed-source players lack is the recognition that the open-source movement is a result of years of collaboration with the customer, involving them in every step of development and delivering premier class support and services—a completely different business model to what they are accustomed to. So today, the open source database providers like Ingres, who have over 30 years relational database heritage, are set to be the disruptive technology of the future as they not only have a solid, proven and reliable product, but deliver unsurpassed levels of support and services, unmatched by the closed-source players.
The author Emma McGrattan is responsible for the development and integration of the Ingres database and associated products and technologies. A leading authority in DBMS technologies, Emma has been instrumental in the ongoing success of the Ingres product line.
She joined Ingres as Senior Vice President of Engineering from CA where she held a similar position responsible for the Ingres family of relational database management products. Emma started her career with Ingres in 1992, and has held a variety of senior development and management positions.
Born in Ireland, she earned a bachelor's degree in electronic engineering from Dublin City University.
