From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=unavailable autolearn_force=no version=3.4.4 Path: eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!feeder.eternal-september.org!news.unit0.net!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail From: Niklas Holsti Newsgroups: comp.lang.ada Subject: Re: How to get Ada to ?cross the chasm?? Date: Thu, 3 May 2018 01:48:56 +0300 Organization: Tidorum Ltd Message-ID: References: <1c73f159-eae4-4ae7-a348-03964b007197@googlegroups.com> <2653d61a-c271-40de-833a-02f5408b3045@googlegroups.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit X-Trace: individual.net +2Xs8KKO01sQxYHxYcLnZQD/VaVaImmK91EtCRqwxr9uvAjbyg Cancel-Lock: sha1:CfEO5dc5fsVtsiRraP1424YDKww= User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:45.0) Gecko/20100101 Thunderbird/45.8.0 In-Reply-To: <2653d61a-c271-40de-833a-02f5408b3045@googlegroups.com> Xref: reader02.eternal-september.org comp.lang.ada:51929 Date: 2018-05-03T01:48:56+03:00 List-Id: On 18-05-02 20:10 , Dan'l Miller wrote: > On Wednesday, May 2, 2018 at 9:39:23 AM UTC-5, ric....@gmail.com > wrote: >> ... We use Ada commercially, and simply build FSF GNAT, which >> comes with the full run-time exception. It takes a bit of tinkering >> to get it going, but what doesn't in the open source world? > > Do you ever notice any bugs in the FSF Ada compiler or its runtime > (especially bugs that are absent in GNAT GPL Community Edition)? Perhaps the question was not addressed generally, but my experience, over some 10 years of using the FSF GNAT (Debian and MinGW) to develop a commercial SW product was the following: I found one problem that I suspect to be an error in the implementation of Unbounded_String, but which was not serious enough for me to track it down and report it. However, I was conservative, and did not start using new Ada features until they had been available in GNAT for a couple of years. > Other people are saying that, yes, narrowly speaking FSF GNAT > legalistically is an option due to more-favorable licensing of the > run-time, I found the FSF GNAT option viable, and am grateful to the people who made and make it available for commercial non-FOSS SW. > but they are effectively compelled to copy bug fixes from > GNAT GPL Community Edition's source code to fix those bugs. Or they can avoid using the buggy feature. > When those bugs are in the runtime, then the only legal right one > has to copy/distribute those snippets of AdaCore source code is > GPLv3, not the Runtime Exception or LGPL or any other nonviral > license. I doubt that. IANAL, but I think most bug fixes are so small that copyright would not apply. If the correction to "N := N + 1" is to change it to "N := N - 1", I do not believe that AdaCore could claim copyright to the correction. Copyright also cannot be claimed on things that can essentially be done in only one way, that is, when the "expression" (the code) of the idea is implicit in the idea itself. A direct importation of entire modules or larger pieces of code from the AdaCore run-time to the FSF run-time could be a violation of copyright. A re-implemenation of the module, even using the same algorithms as used by AdaCore, would not, AIUI, be a copyright violation. This is not about patentable ideas, but about expression of ideas. A more severe limitation of the FSF GNAT is the non-availability of run-times for non-PC systems, such as embedded microcontrollers. Not all embedded systems can be implemented with Linux or MS Windows on x86. This is an area where amateurs could contribute, and several have. -- Niklas Holsti Tidorum Ltd niklas holsti tidorum fi . @ .