When two standards are better than one

This week, while I was in San Francisco, Sun held a round-table meeting to discuss its postion on open file formats.  The meeting was held only a few hours after a hearing on the hotly contested matter was held in Massachusetts as that state looks to decide if it is going to move forward in support of just the OpenDocument Format, or if Microsoft's Open XML will be added to its list of approved file formats.  Based on the content of the meeting, Sun is clearly in the camp that one standard is better than two while Microsoft sees no problem with supporting an extra one.  I'm in the "one" camp.  Always have been.  Never liked the messes that were caused by things like DVD+R and DVD-R.  Imagine if we had two standards for TCP/IP (the protocol that makes the Internet tick).  Or two standards for Web browsing.  Whoops, we do.  And that's gotten us into several messes that some people would rather do without. 

There's a reason one format won the VHS/BetaMax war.  Sure, two or more standards can survive in a market.  But just because they can doesn't always mean it's a good idea.  In most cases where two more standards are surviving in a market, it's not because the buyers got together and said "You know what would be great Phil? If we had two or more standards that were incompatible with each other for this one thing." In just about every case I can think of, it was the sellers that thought it was a good idea. Sellers with their own agendas. I can't tell you how many times I've had to explain the differences between the GSM, CDMA, and iDEN flavors of cell phone technology and why they don't interoperate.  When I'm done explaining it, it's not uncommon to get "That's dumb. Who thought of that?" as a response. 

Two or more standards in a market often ends up being a source of unncessary confusion.  Two or more standards (for the same thing) inside one enterprise makes even less sense. If you ask me. Why bother setting standards at all? Because the greatest thing about them is there's so many?   The reasons enterprises standardize is for interoperation with other systems. Not only does that interoperation equate to more seemless sharing of data (not necessarily as important for a whole market as it is for a single enterprise), it's what facilitates easy switching to a another solution should you become dissatisfied with your current solution because of its costs, peformance, stability, or security.  For example, if certain bloggers are now thoroughly dissatisfied with SixApart's TypePad now that that service has essentially crashed for both publishing and viewing of recent posts (again).

Massachusetts is apparently now open to having two standards for the same thing (Office productivity file formats which is not what PDF is for).  If it's going to do that and throw away the true benefits of interoperation, it might as well not have a standard at all which is why I think it should pick one and stick with it.  The same way five separate library associations are apparently doing the same.  Never mind which one they picked.  I'm purposely staying away from the issue of which one to pick.  That's for a separate blog.  Massachusetts is apparently open to going with two.  So says the Boston Globe.  I think that's a mistake for Massachusetts.  I think that's a mistake for you.