Why is Ada considered "too specialized" for scientific use
in [ADA]
|
From: Colin Paul Gloster on 17 May 2010 06:26 On Fri, 14 May 2010, Robin sent: |------------------------------------------------------------------------| |""Colin Paul Gloster" <Colin_Paul_Gloster(a)ACM.org> wrote in message | |news:alpine.LNX.2.00.1004132014460.3668(a)Bluewhite64.example.net... | | | || I met someone today who described himself as "an ordinary FORTRAN | || programmer" who advocated C for the practical reason that libraries | || are designed for C. He claimed that small tasks are good for multicore| || and large tasks are good for GPUs. | | | |I think you will fnd that libraries are also designed for Fortran." | |------------------------------------------------------------------------| They certainly are. He uses code based on LAPACK. If you are aware of Fortran bindings to GPUs which you would care to inform me of, then I could mention to him. Maybe he already knows about them, maybe not, but I have already informed you of the reason he gave for advocating C.
From: robin on 17 May 2010 05:23 "Shmuel (Seymour J.) Metz" <spamtrap(a)library.lspace.org.invalid> wrote in message news:4bef48fb$11$fuzhry+tra$mr2ice(a)news.patriot.net... | In <4bed3524$0$67490$c30e37c6(a)exi-reader.telstra.net>, on 05/14/2010 | at 08:50 PM, "robin" <robin51(a)dodo.com.au> said: | | >Here's another example. | | No. It's another example of an algorithm that was first implemented in a language other than Algol -- -and more specifically, in a language at a lower level than Algol. So, the correct answer is therefore "yes". | >Don Shell published his algorithm in machine code. | | No. Probably CAGE. Possibly SAP. Either you didn't read the article or | you have no idea of what machine code is. To be sure, I know what machine code is. I used the term in the general sense. Here, the intent was to point out that the algorithm was not first implemented in Algol.
From: Shmuel Metz on 17 May 2010 06:05 In <4bf10c9c$0$89663$c30e37c6(a)exi-reader.telstra.net>, on 05/17/2010 at 07:23 PM, "robin" <robin51(a)dodo.com.au> said: >It's another example of an algorithm that was first implemented in a >language other than Algol K3wl, David. Why do you persist in debunking claims that nobody has made while ignoring the actual issues in dispute? >So, the correct answer is therefore "yes". Unfortunately it's the answer to a question that nobody asked. It's not the correct answer to what you actually posted. >To be sure, I know what machine code is. The evidence suggests otherwise. >I used the term in the general sense. The "general sense" would have been machine language on more than just the 704. Pointing to assembler code as an example of machine language just makes you look less than Frank. >Here, the intent was to point out that the algorithm was not first >implemented in Algol. Which is totally irrelevant to the issues in dispute. -- Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel> Unsolicited bulk E-mail subject to legal action. I reserve the right to publicly post or ridicule any abusive E-mail. Reply to domain Patriot dot net user shmuel+news to contact me. Do not reply to spamtrap(a)library.lspace.org
From: robin on 17 May 2010 06:09 "Shmuel (Seymour J.) Metz" <spamtrap(a)library.lspace.org.invalid> wrote in message news:4bf114cf$4$fuzhry+tra$mr2ice(a)news.patriot.net... | In <4bf10c9c$0$89663$c30e37c6(a)exi-reader.telstra.net>, on 05/17/2010 | at 07:23 PM, "robin" <robin51(a)dodo.com.au> said: | | >It's another example of an algorithm that was first implemented in a | >language other than Algol | | K3wl, David. Why do you persist in debunking claims that nobody has | made while ignoring the actual issues in dispute? | | >So, the correct answer is therefore "yes". | | Unfortunately it's the answer to a question that nobody asked. Go back and look at the postings. You will find that it is. | It's not the correct answer to what you actually posted. You are mistaken. What side of the bed did you get out of this morning? | >To be sure, I know what machine code is. | | The evidence suggests otherwise. | | >I used the term in the general sense. | | The "general sense" would have been machine language on more than just | the 704. Pointing to assembler code as an example of machine language | just makes you look less than Frank. | | >Here, the intent was to point out that the algorithm was not first | >implemented in Algol. | | Which is totally irrelevant to the issues in dispute. Which it isn't. See above.
From: Dan Nagle on 17 May 2010 09:36
Hello, On 2010-05-17 06:26:35 -0400, Colin Paul Gloster <Colin_Paul_Gloster(a)ACM.org> said: > They certainly are. He uses code based on LAPACK. If you are aware of > Fortran bindings to GPUs which you would care to inform me of, then I > could mention to him. Maybe he already knows about them, maybe not, > but I have already informed you of the reason he gave for advocating > C. There is a standard way of calling C functions from Fortran. It is considered an important feature of Fortran 2003, and is widely implemented. Therefore, the I-need-a-C-library reason for not using Fortran is not very strong (nor is that a very strong reason for not using Ada). Of course, YMMV. People make emotional decisions, and then backfill with rational (sounding) reasons. -- Cheers! Dan Nagle |