This seems kind of strange to me, because it basically implements a very small subset of the features you would expect in other languages. It's not even that it's a simplification of some common paradigm; it just focuses on some weirdly specific activities.
I'm the author of Gura. Thank you for commenting on this language.
I found some of the comments are mentioning about related points each other, so I issue an independent comment instead of responding to each.
Fortunately or unfortunately, I didn't know about Icon and APL at the time I decided to develop Gura. Now, I'm being encouraged by the fact that these languages have already paid efforts to sophisticate repeating process, because that teaches me there certainly are some people who need that as I do. Unintentionally, Gura ends up to be what tries to inherit these language's feature just like today's major OOP languages are affected by their ancestors like Smalltalk.
Do you think that an effort to sophisticate repeating process has much less significance than OOP paradigm? I don't. Repeat is a quite primitive part of programming and is likely be a critical bottleneck of performance. Gura's mapping feature is aimed to eliminate side effects from a repeating loop so that it can be optimized for higher performance with parallel computing.
Actually, the topic of parallel computing is currently just an idea, and you can't get any benefit regarding it for now. It's for the future. In addition to mapping features, Gura has some unique mechanisms like a flow control generating an iterator and a quoted value that can get an expression without evaluation. I'd be happy if some of them inspire you to try this language.
