Another interesting wrinkle in this whole conversation is that research tools that are useful for normal people tend to eagerly escape the lab. This happened wi

Another interesting wrinkle in this whole conversation is that research tools that are useful for normal people tend to eagerly escape the lab. This happened with (off the top of my head only): SQL/RDBMS, the Internet, Email, garbage collected languages, object-orientation/actor model, research UNIX, and the WWW. The list is endless. When this happens, the researchers (who were the initial user base) basically lose control of the technology. This process can benefit the researchers or not, but is basically indifferent to their needs.