I have some experience with Forth; I have been programming a forth CPU in 2002 and got paid for it.(it's me who left a comment in the article) The article says

I have some experience with Forth; I have been programming a forth CPU in 2002 and got paid for it.(it's me who left a comment in the article) The article says "the main WikiReader app is written in C" and uses forth "for testing and apps"; a fine approach if you depend on core libraries which are in C. It'll be interesting to see which functionality is done in Forth. The article mentions that Forth is used to simplify debugging. This is traditional approach and this is what Forth proponents say. The truth is, it was more true back in the day when JTAG wasn't available. Now that JTAG is omnipresent and popular IDEs can use it, the use of fringe language such as Forth (or any software solution where hardware-assisted one is readily available!) is so much less justified. In the quoted letter, Christopher Hall, the developer, does a nice job of explaining the use of Forth in a project. But I think many people who got the chance to familiarize with Forth, feel like this is a mere justification. Like "c'mon you can tell them what you want, but between you and me, we know that this is a good shit and you just want to try it once again" Or, like, picture Paul Graham picking a language for his next project: "Should I use Lisp or Blub? Hmm, this question needs careful consideration." NB: Wikipedia articles for Forth and JTAG are good.