Bob Nemec, the new director of STIC, recently made a point in a blog post that echoes a point I made recently on the Squeak-Dev mailing list.
Excerpt from what I said (e-mail sent Wed, 17 May 2006 -0700):
"Those of us `old timers` who went through the learning curve on Smalltalk years ago no longer clearly remember what it was like, nor what the conceptual stumbling blocks were. We have now fully internalized the `Smalltalk way,` and don't really grok why newcomers don't instantly realize why its so superior.
I think our attitude is wrong, even if we're `right.` We absolutely should enable more `traditional` approaches for doing Smalltalk programming. If the Mountain won't come to Mohammed, then Mohammed must go to the Mountain."
Excerpt from what Bob said:
"We had a rare sighting at last night’s meeting: a couple of Smalltalk newbies. One of them, Chris, was very willing to talk about his frustrations in getting into Smalltalk, which he’s been looking at for only a few weeks. He’s a webmaster with an interest in Seaside. In his opinion, the Smalltalk community presents a significant barrier to entry by coming across as arrogant and aloof. It was difficult defending some of his criticisms; in many ways, he was right. I had thought those days were behind us.
He made some very good points. Basically, we don’t provide the baby steps needed to learn our environment. There are virtually no easy to find introduction materials on the web for people poking around with Smalltalk for the first time. The fact that Smalltalk was able to do things 15 years ago that some languages struggle with today is irrelevant to him. He wants to learn, not be preached to."
I would add that educational materials accessible to beginners is only one approach. Another is to enable newbies to use Smalltalk in ways that much more resemble the tools and environments they already understand. Just because we know that such "traditional" tooling is actually inferior is no reason not to provide it--especially when that's one of the main stumbling blocks to early/easy acceptance by newcomers.
Who knows, perhaps we will discover that the "traditional" tooling has certain advantages, and/or that there is a synergistic effect when the traditional Smalltalk approach is combined with the traditional Unix/Windows approach.
But the bottom line is simply that Smalltalk needs to get rid of its "holier than thou" attitude, stop mumbling about all the great forward-looking achievements of its past, and start dealing with the issues of today and planning for its future.