[totalzeroone]

I seem to remember that the author only require that the forked Shen satisfies a number of published tests, that is to guarantee that the forked program function like the original Shen in the basis case. A forked program could extended the original one with new capabilities, libraries and ideas. I think that the author of Shen believe that free sofware should not be free as in beer, now he earn some money selling the Shen book. Only my two cents.

[jcrubino]

Shen and Qi are written by an academic. More likely than trying to sell the book is the that author is trying to ensure that the core theories that got put into the languages stay in the language with out alterations to the core principles... what exactly those are I never delved into enough for comment but the documents and history point to a very principled author and language being created.

[totalzeroone]

After searching for the official specification (a term used formally in the license), the only thing I could find is the web page for the documentation that is called the official shen standard doc, http://www.shenlanguage.org/learn-shen/shendoc.htm, so I suppose the official specification is the content of this page. As I don't have the book of Shen I can't tell if there is further documentation about the "specification of the language".

The author in the license page claims that he is not going to make a fork illegal by changing the standard in such a way as to make the fork retrospectively illegal.

There was one author who made a close source port of Shen in java but then there were big problems. Perhaps he could give a lot of information of what kind of problems are you to encounter when you try to develop a close source product using Shen. The last think I can remember was that he bought the book of Shen for her 16 years old daughter and he was emotionally very interested in Shen but then all went down. [By the way, I apologize for not putting the 's i the third person )). in my other posts.

[totalzeroone]

here is the link https://groups.google.com/forum/#!topic/qilang/yL2Eps2cnwY

[MTarver]

The issue here was not that the port was closed; but that people helped to build it and gave their time for free on the assumption it was going to be free and readable. My post

https://groups.google.com/forum/?hl=en#!topic/qilang/PJyM6xC...

opened this topic. The rest is history.

I have a clear policy of not contributing my time to closed (for money) software unless I am being paid as a consultant. I do not have any objection as such to closed source software.

Mark

[totalzeroone]

Hello Mark. Could you give a link to the "specification of Shen", that is, is there a battery of test that for a fork of Shen certifies that the fork is a conforming with the specification fork?, is the "official standard documentation web page of Shen the text of the specification alluded in the license?

Thanks for you work in Shen.

[MTarver]

The license was introduced before Shen was issued in September 2011 (the license came out in June) and for a year or so the only specification was Shendoc (now Shendoc 16). The understanding was that Shen was specified in that document and what was not covered by Shendoc was covered by 'Functional Programming in Qi'. Later a hurriedly introduced text 'The Book of Shen (first edition)' (TBoS) was produced to fill a gap (2012) and this year (January 2014) a more thorough 2nd edition of TBoS (> 400 pages) was published which fixes the language standard very thoroughly. This is currently the canonical standard. You can find a link to that book on the Shen home page.

http://www.fast-print.net/bookshop/1506/the-book-of-shen-sec...

Shen is now very stable and has been for nearly two years. At my suggestion, I posited that it might be better to move the standard to a computable series of tests and this was floated to the 2011 committee that is responsible jointly for all the ports.

http://shenlanguage.org/2011committee.html

Such a change requires the unanimous consent of all the people involved and it seems we have this and a reworded simplified license.

The only obstacle is the work needed to put this test suite together. I've suggested that this suite might be assembled in Github, though for legal reasons the final version must be put in a publicly accessible but tamper-proof place.

Since the type-integrity given out by the system is not better than the strength of the kernel, we take kernel work very seriously. There is already a suite of 126 tests that I run every Shen port through and 2011 members echo these tests. But this informal test suite needs to be amped up to several hundred tests to approach what I consider to be an adequate test suite. It is very boring but important work. So far I have begun assembling all the programs in TBoS into this suite.

These license issues really only affect people who are deeply involved in kernel work and as far as application programmers are concerned, I doubt that it affects them much at all. As far as graphics, concurrency, FFI etc. and add-ons are concerned there are no restrictions. Likewise none on closed source work.

[MTarver]

If you want to see the tests that are currently used; they are in the source code download.

http://www.shenlanguage.org/Download/download.html

See Test Programs/Readme.shen