>Peter brings up an excellent point about larger software companies. A situation I ran into lately got me thinking about this:
>
>- A client bought a semi-custom solution based on Microsoft Dynamics, with .Net customizations for $X
>
>- After using it for a year, it turns out Microsoft has ceased support for the version of Dynamics on which the current version is based. To continue to use it would require upgrading to the latest version of Dynamics and significant rewriting of the .Net customizations. Even getting a "deal" recognizing the short useful life of the current version, the cost to bring the solution forward would be ~60% of the original purchase cost (call this $Y)
>
>- Instead, the client will likely go with a field-for-field replacement written in Java (I don't yet know the exact details). Total cost something like 30% of $Y (not $X)
>
>The kicker is the guy offering the Java solution is the same guy who supports the current solution - an independent developer. IOW he's dropping Dynamics/.Net in favour of Java.

But as the dev already knows topic, customer and pitfalls that calculation makes sense. it does not reflect Java being better equipped for the job, but the dev wanting to switch horses but still remembering the other ride...