Talk:Norcroft C compiler

Latest comment: 12 years ago by Trevj in topic A view from Codemist Ltd
edit

There's a reference to "re-source the Norcroft back end" in Tom Crick's 'A GCC Front End for BCPL', which may be worth including. --Trevj (talk) 14:35, 16 June 2011 (UTC)Reply

Website in infobox

edit

The currently stated site is the most appropriate one I know of. Perhaps there's something more obvious which I've missed. It's not listed there for WP:SPAM puropses, despite such an appearance from the URL. I hope this makes sense. Thanks. --Trevj (talk) 22:52, 16 June 2011 (UTC)Reply

Converting Programs to 32-bit

edit

This guide published by Castle Technology states it is "Based on a two part article which appeared in both Archive and Acorn User, November/December 2002." Therefore, if anyone reading this has access to the relevant publications it could be useful. I don't know whether they include any extra information, or if it's all in that doc. In any case, Archive and Acorn User would be preferable in terms of reliable sources. --Trevj (talk) 13:03, 17 August 2011 (UTC)Reply

This article is full of lies

edit

Most of which can be seen by just visiting Codemist's website. Norcroft was a licencable C front end developed at the university of Cambridge (and later also at the university of Bath) used in teaching, and Acorn licenced it. It targets many many architectures. The history and architectures sections of this article are completely made up. And many elements in the main section (including ownership etc) are misleading. Also, useful information such as the fact that ARM themselves abandoned the Norcroft frontend for EDG's are missing. Sort it out, guys. — Preceding unsigned comment added by 82.6.165.102 (talk) 15:53, 20 October 2011 (UTC)Reply

Well I made a start to fix it, you're correct, Norcroft is certainly not in anyway RISC OS/Acorn specific as the article suggested.--Flibble (talk) 16:22, 20 October 2011 (UTC)Reply
Thanks Flibble. User:82.6.165.102, please assume good faith: what you refer to as "lies" are actually misunderstandings. Did you read all of the article before contributing here? It did mention both Cambridge and Bath universities. And the 'Architectures' section contained a {{citation needed}} tag. I appreciate there were some inaccuracies, but if you'd checked the edit history you'd have seen that there's been only one real contributor to date. I'd suggest that any new-ish article worked on by a narrow breadth of contributors (especially for which there are few online sources) is likely to contain inaccuracies. Wikipedia is a collaborative project, so fix it rather than unconstructively stating that the article is full of lies! Your attitude in this case is exactly the reason why editors leave the Wikipedia project! I hope this makes sense to you. Thanks for reading. -- Trevj (talk) 04:56, 21 October 2011 (UTC)Reply

Acorn C/C++

edit

I'm confused. http://www.codemist.co.uk/ncc/index.html says Acorn Computers ARM C Compiler. This is the original ANSI C compiler known as Norcroft C. But that page is titled Codemist Compilers / Norcroft, which implies that all the other compilers (listed beneath Acorn on that page) also fall under the Norcroft title. (Which is ambiguous, if "Norcroft C" refers to the Acorn compiler.) Should there be subheadings within this article for each compiler? I've just emailed Codemist to request input (pointing to WP:COI and WP:V).

Additionally, I think that the 'History' and 'Uses' headings should be retained (perhaps as subheadings within 'Acorn C/C++'). Help:Section states A page can and should be divided into sections [...] Removing section headings could result in unstructured expansion of the article. While I can't find any policy as such, e.g. at MOS:HEAD, the purpose of sections is clearly to structure articles. WP:BODY also states Very short or very long sections and subsections in an article look cluttered and inhibit the flow of the prose. but (with the exception of the 'Architectures' seciton) I wouldn't consider them "very short". So where do we go from here? Thanks. -- Trevj (talk) 07:49, 21 October 2011 (UTC)Reply

A view from Codemist Ltd

edit

First I should make it clear I am a director of Codemist Ltd who wrote the Norcrift C Compiler and still sell it (see http://codemist.co.ukncc). The main page seems to talk as if this as a product of Acorn, when in fact it started when Arthur Norman and Alan Mycroft wanted a C compiler to run ML/LCF and the ARchimedes was new. The compiler was written by them, and then taken up by Acorn. For some years we developed it together, but we recently agreed to separate and ARMremoved our copyright notice. We have built Norcroft C compilers for a wide range of processors, including many for japan with an arrangement with a japanese software company, but recently mainly startup chip manufactures.

The emphasis on Acorn/ARM is strange to me, as it also formed the basis on the Transputer C compiler (via Perihelion Software Ltd), a Hitachi mainframe C, as well as other architectures (sorry if the signature comes out wrong) ~~ jpff 18:07, 23 October 2011 (UTC) ]Reply

That's very helpful. Thanks for the info, which I'm sure will be incorporated soon (preferably with sources, if they are readily available). The emphasis here isn't deliberate and has arisen as a result of lack of information/understanding. It should be noted that Wikipedia is a work in progress! -- Trevj (talk) 21:16, 23 October 2011 (UTC)Reply
I can see two ways forward, either we research and improve the article on Norcoft C to emphasise its portability and not over emphasise one particular implementation, or rename the article to 'Acorn Developer Tools' or similar if that was the original intention of the page?--Flibble (talk) 09:01, 24 October 2011 (UTC)Reply
Acorn C/C++ currently redirects here and is possibly a preferable title to 'Acorn Developer Tools', being the common name AFAIK. Researching Norcoft C to emphasise its portability makes sense. In the mean time, I've created a couple more headings. I'm not sure about the notability of the following, so have omitted them for the moment.
  • Hitachi for their mainframe (IBM compatible)
  • ICL KCM the Knowledge Crunching Machine
  • Matsushita Adenart experimental graphics processor
  • AIH324
  • NEC 860
Unless further info is included to balance the article (more background and generic info is needed), the Acorn stuff may have to be moved out as a spinoff, per WP:UNDUE. -- Trevj (talk) 13:44, 24 October 2011 (UTC)Reply
  Done as a spinoff. I hope this improves things a little. Further clarification welcomed. Thanks. -- Trevj (talk) 21:46, 21 November 2011 (UTC)Reply