Because I monitor GIS press releases like a maniac, I see a lot of announcements for newly-minted GIS extensions.  Every time I see one, I think: “Who will ever buy that?”  But someone must buy them; there must be a market, or else developers wouldn’t keep developing them, right?

Not sure.

I remember the early days of ArcView “extensibility” – first with Avenue scripts, then with full-blown extensions.  Then came ArcGIS, and extension fever spread like, well … fever.  There must be hundreds, if not thousands of (Arc)GIS extensions in existence.  To what end?

I know of very few successful third-party GIS extensions.  One example is EditTools from SpatialTechniques.  This useful utility gives the ArcView user a quasi-ArcEditor functionality.  In other words, it saves its users money.  This makes sense.  

On the other hand, I know of many unsuccessful GIS extensions.  Not necessarily because they were poorly written (that may be a contributing, but not a main factor).  There are plenty of commercially-unsuccessful GIS extensions primarily because their authors overestimated (or perhaps did not even research) the market.  One example – a GIS extension designed for urban planners purportedly facilitates the development of municipal or regional master plans.  A once-a-decade activity.  How many copies do you think it will sell?  To whom?

Another issue plaguing third-party extensions is the breakneck pace of core technology development.  I am having a hard time keeping up with my core ArcGIS patches.  There’s never knowing which patch will break which extension.  It’s like a daisy-chain reaction leading to a game of Russian Roulette.  Unless you are geek to the bone, this exercise can become very tiring very soon.  Not to mention unproductive.

Very coincidentally, Directions Magazine is running a poll on the subject.  Go vote.  I’ll be watching the results.  Like a maniac.