I saw Jean-Marie Hullot's original "InterfaceBuilder
" for ExperLisp?
, an obscure variety of LispLanguage
on the Macintosh around 1985. It immediately struck me that he had simplified interface construction enormously, compared to Smalltalk or anything else then available, and that this should be applied at once to the appropriate object languages in the C family, for commercial gain! When SteveJobs
picked him up and called the NeXT front end by the same InterfaceBuilder
name I assumed Jean Marie was really key to the process. His own account in 1989 certainly gave me this impression (see above). On the other hand, I know that NeXT had many more developers than JMH and that in any important project everyone rates the relative contributions of everyone else rather differently than they do themselves. (In mining exploration, they have so few economic successes to share out that the joke goes: "What's the definition of a mine?". Answer: "A hole in the ground with a liar at the top.") One reason for caring is that JMH was JustaProgrammer
and in my mind could be held up as something of an icon in that category, someone who had a superb feel for external (HumanComputerInteraction
) as well as internal design issues. But I know none of the other people in person. Any comments? -- RichardDrake
I'm not familiar with JMH's work, but it doesn't surprise me. A lot of really interesting UserInterface
work was going on in the industry during the mid-80's, and Next tried very hard to locate and take advantage of the best of it. I think that "InterfaceBuilder
" was also the name PPI used for their own GraphicalUserInterface
tool, and Next bought access to that when they negotiated their use of ObjectiveCee
. I was a research scientist on the staff of an institute (IRIS) at Brown University at the time, and we were also exploring similar things. It was one of those periods when the same ideas were being pursued in parallel by lots of individuals, and so innovations tended to surface in wavefronts of parallel but independent development. -- TomStambaugh
Other names involved may be: BillParkhurst?
was also an influence on the genesis of VisualAge
. I met JMH and saw InterfaceBuilder
on a visit to Expertelligence. The next time I met him was on a visit to NeXT. I was very attracted to the model of putting a GUI together out of components and hooking it up to objects to implement the application logic. The thing I didn't like, being a big fan of IDEs was the process of being dumped into the old world text and command line based development tools for Objective-C programming. It just felt like you were stepping over a precipice.
About the same time we got our hands on reasonably good Smalltalk implementations like SmalltalkV for the PC class machines of that time. So I put together a small team of three and in less than a week we had a prototype of an InterfaceBuilder
UI editor running in the Smalltalk IDE, which considerably leveled the floor. Shortly after that we got an opportunity to demonstrate this prototype, which I dubbed ApplicationBuilder?
who was the IBM Executive who was responsible for application development tools at IBM. Earl saw the benefit of the Smalltalk IDE which led to IBM eventually putting out VisualAge
including Smalltalk. The development team took that prototype and greatly expanded the programming by wires metaphor, too far if you ask me. -- RickDeNatale
See also: NextStep