Like Java, Shen is a The Right Thing vs. Worse is Better language. I think the concerns expressed in in the Reddit discussion are overblown, you have to assume (or not) a certain degree of good faith on the part of those responsible for maintaining the language's correctness.
Sure, the execution of the license is vague (and note we're talking about concerns for people doing ports, not those using a blessed port), but don't forget how explicit Sun's Java license was and how they simply refused to play ball with Apache.
And some of the explicit concerns (e.g. 10 vs. 12 bytes per int) are just plain silly, given that there are blessed existing and in progress ports to several versions of Scheme and to Javascript.
The presentation on the site doesn't do the language and its design justice, so I'll fill in a bit here.
Qi is a programming language with a "powerful" static type system, which can be useful. Other generally usable languages with "powerful" static typing include Haskell and Ocaml. Some people want "powerful" static typing. If you are one of those, then like me, you should be interested in Qi.
Qi has been implemented in Common Lisp. Qi code is translated into Lisp code, and Qi code can also call Lisp code - there is interoperability. If you code in Lisp or intend to do so, you can use Qi as well, and still reuse existing code and functionality.
However, few people use Common Lisp. Some of the most popular platforms do not support Common Lisp.
Shen is intended to be the next evolution of Qi; Shen is a "redesign" of Qi to reduce the effort required to implement Qi on a language. The idea is to implement Shen on popular languages, like Javascript, Python, etc.
There are three notable benefits that we can look forward to:
1. Single language across many platforms.
2. High-level language available on platforms, like the web browser, mobile devices, and so on. Haskell today doesn't run on mobile phones.
3. Qi/Shen is like a cleaned-up Lisp. Some people prefer that to languages like Javascript, to Dalvik Java, and so on.
What's pending is its implementation on the non-Lisp languages. Performance and reliability could be concerns, given the differences between its original host language, Lisp, and other languages. If the performance and correctness tests for these turn out well, there is good reason to be excited.
The release of Shen on Common Lisp is of lesser consequence. Qi already runs on Common Lisp. Releasing Shen (which is mostly a more "portable" Qi, but retaining the syntax and power) on CL just proves that Shen has been completed with nearly no errors.
Currently, there's a group working to implement it on Javascript, but they postponed the delivery to late October.
Its success would be noteworthy.
Qi is a lisp like language with a type system similar to Haskell, except that it's fully Turing Complete (the type system). Shen is the next version which was meant to be more accessible to regular users.
Shen was designed to not only run on Common Lisps like Qi, but also on Javascript (released mid-October) and Scheme. Practically all Qi programs will run in Shen, but not all Shen programs will run in Qi.
http://www.reddit.com/r/programming/comments/ibqp1/meet_the_...
It's one thing if one doesn't want a collaborative project, but that goes against (at least the stated) goals listed on the license page.