Vendor Lock-In, Consumer Choice, and Competition
One of the things Henry Chesbrough discusses in his book Open Business Models: How to Thrive in the New Innovation Landscape is how early adopters of a technology often get “marooned,” or trapped on a particular application or platform. This is a well-known issue in the software industry where it’s called “vendor lock-in.” Vendor lock-in is a powerful force, and I don’t think Chesbrough spends enough time talking about just how abusive this practice it really is.
In software, the most common form of vendor lock-in is when stored data is non-migrable. When your data is non-migrable, it is stored in a proprietary format that cannot be converted completely and reliably to any other format. Microsoft Office is an excellent example of a product that exercises vendor lock-in in this way. Once you’ve saved your data in .doc format, it’s impossible to get it out. There are some programs today that get close — the open-source OpenOffice.org, Sun Microssystems’ StarOffice, IBM’s own Lotus Symphony, and even online applicationslike Google Docs or Zoho — but even with these you will never be able to recover 100% of your formatting, content, or layout when you open a .doc document. Office also does not export into any external formats that retain 100% fidelity and editability. We’ll talk a little more about Office later.
If you’re a carpenter and you choose a tool, say a screwdriver, you’re always free to choose another screwdriver later because toolmakers today use open standards. Open standards are a set of publicly-available conventions that specify how a product or technology behaves. For example, all phillips-head screws use the same set of patterns on the screw head according to an open standard, which means that any phillips-head screwdriver built to that well-known, openly-available standard can be used with that screw. This tool interoperability is critical to the freedom of the consumer to choose another product at any time. A carpenter who chooses a Craftsman screwdriver on Monday is free to choose a Stanley screwdriver on Tuesday, and due to the open standards for screws, he can be assured that it will just work.
In business, software is nothing but a tool to get a job done. However, whereas in carpentry we can change tools relatively cheaply, in business changing tools is often extremely expensive or impossible because the open standards that enable competition are not available. This inability to move from one software product to another forces businesses to remain with the same — often inferior — piece of software, ostensibly forever. An important side-effect is that once a piece of software that practices vendor lock-in has achieved a critical mass of market adoption, it is impossible for any other product to compete effectively because they cannot interpret the dominant application’s data format. This is a serious double-whammy for software businesses, but the consumer is the real loser.
Vendor lock-in is one wound time does not heal. The longer a business stays with an application, the more data they produce on that platform. The more data they produce on the platform, the more data they have that cannot be migrated to a competing product. The more data they have that cannot be migrated, the more painful switching becomes. Microsoft Office is an especially insidious example of vendor lock-in because of the sheer length of time it has retained a functional monopoly on the market, and the value of the data — tens of thousands of businesses for tens of years — that it is effectively holding hostage.
It is impractical (and unfair, I think) to require a company to release its own proprietary formats so that competitors may support them immediately. Time has demonstrated that reverse engineering complex data formats also isn’t feasible due to the rate at which the formats change. (Additionally, many software End-User License Agreements specifically forbid reverse-engineering a file format or program! But I digress.) There is another good solution, though, which is to require all competitors in a field to make available a feature to export their stored data data in a different, open format with 100% fidelity, which all competitors in a field can read and write. Open Office XML promised to be this format for the Office Suite market, but popular opinion indicates that it has failed.
It is my opinion that Microsoft has intentionally fostered vendor lock-in on their Office suite. At first blush, it looks suspiciously like Microsoft has had a hand in the failure of Open Office XML. For example, Norway has come forward and claimed that there were “serious irregularities” with the ISO voting process for approving Open Office XML due to Microsoft influence. While Microsoft released specifications their binary file formats in early 2008 — you can read them here if you have a couple years and about a ton of aspirin — I suspect this was actually due to political pressure from the European Union, and not out of the goodness of Microsoft’s heart. It’s also interesting to note that Microsoft is in little danger of competition even with the file formats released due to the gratuitous, mind-numbing length and complexity of the standards. I wonder if the specifications were not intentionally made to be complex before they were released, but Joel Spolsky actually weighed in on Microsoft’s side on this issue, and his argument makes sense, so I’m willing to give them the benefit of the doubt here. Regardless, though, it’s reasonably clear that Microsoft is protecting its vendor lock-in jealously.
Open Office XML and the opening of the Office formats are actually just punctuation in the interesting story about Microsoft’s long-running anti-trust battle in Europe. I’ll talk more about that in another post.
What’s everyone’s experience with vendor lock-in? Where you work, is vendor lock-in having a negative effect? Do you have data you wish you could migrate, but can’t due to the inability of competitors to handle the formats?