search

fortran-lang / webpage

New Fortran webpage
https://fortran-lang.org/en
MIT License
48 stars 39 forks source link

Make the compiler descriptions more objective and less promotional #73

Open Beliavsky opened 3 years ago

Beliavsky commented 3 years ago

Some of the compiler descriptions read like advertising copy, because they were taken from advertising. For example,

"Absoft Our compilers build faster code more efficiently than ever before."

The "our" should be reworded since the fortran-lang site is not owned by a vendor.

The Lahey description has two exclamation points that don't below in a vendor-neutral site.

"The Cray Compiling Environment (CCE) is the cornerstone innovation of Cray’s adaptive computing paradigm." This is MBA-speak that conveys nothing to me.

In general, vendors will try to summarize the special features of their compilers in a few sentences, and such descriptions can be informative, but I think fortran-lang should tone them down a little.

vmagnin commented 3 years ago

I agree with @Beliavsky and the commercial style was one of the reason it was difficult to translate that page in French.

certik commented 3 years ago

In order to make this section fair, I still like this idea the most: give each vendor a space of 100 words or less to put anything they want for their compilers. That way they can advertise as they wish. We can perhaps put a paragraph at the very top explaining this.

The alternative is to start editing the paragraph for each compiler, but I don't have access to all of them, and I don't want to be in a position either to be the "editor" for how compilers are described and advertised.

By offloading this responsibility to the compiler vendors themselves, and just providing a venue and "rules" (100 words or less), and making those rules clear (currently it is not clear) seems better and ultimately fair I think.

Update: to explain why I don't want to be the "editor": first, I am in a conflict of interest, as I am the primary developer of the LFortran compiler. Second, even if I wasn't, I don't feel we should be "judging" or "comparing" commercial compilers (or even open source compilers). Rather, let the users decide. The only judging we made was to roughly order the compilers based on our perceived usage in the community of them. If there was some more objective measure, we could use that instead. But other than that, my vision was to provide the platform, so that at least all the compilers are now sitting at the same table. But other than that, let each compiler speak for themselves. To more objectively compare compilers, we created this benchmark repository: https://github.com/fortran-lang/benchmarks/, we haven't had time to actually populate it, but we will. Benchmarking objectively is very hard, see the discussion here: https://github.com/fortran-lang/benchmarks/issues/2 and at the other issues there. I think there is a way to do it, to simply show the numbers and let users decide. Another idea is an automated results of "Fortran features support", where we run each compiler with a test suite and report how well it is supported. But even if we had all that infrastructure for more objective comparisons (we currently don't), I still don't want to be writing the final "summary", but rather I just want users to try the compilers out (or not) and decide themselves.

CC @milancurcic.