Why didn't PL / I do this in scientific computing?

Question

PL / I is a very old language, but apparently it has not gained much popularity and appreciation as Fortran for scientific computing. Why is this? A search on the Internet shows that there are many PL / I codes for scientific computing. There is, among other things, a scientific computing library developed by IBM in PL / I.

But I did not find any free compilers for Windows for PL / I. Due to this lack of free compilers that PL / I did not attract the scientific community?

Thank you so much...

Answer 1

Why is this?

I think there are several reasons.

Perhaps because habit is second nature . When PL / I appeared, FORTRAN already existed for almost 10 years. When some new technology / language appears, you can begin to neglect the existing code base, causing its outdated code. But there is no reason to immediately start the conversion. This was especially a big problem in the 70s due to the lack of automatic converters and other tools.

The next reason may be that there is no silver bullet . PL / I tried to create such a silver bullet - a general-purpose language. They tried their best, you know everything else. =) PL / I was a monstrous language. Frow Wikipedia article :

Programmers were sharply divided into scientific programmers (who used Fortran) and business programmers (who used COBOL), with considerable tension and even hostility between groups. The PL / I syntax borrowed from both COBOL and the Fortran syntax. Therefore, instead of noting the features that make their work easier, Fortran programmers of the time noticed the COBOL syntax and it was believed that it was a business language, while COBOL programmers noticed the FORTRAN syntax and looked at it as a scientific language.

Also, “old people,” such as FORTRAN, are developing under the threat of death and have added features such as structured programming, object orientation, etc. This reduced the relative advantages of PL / I.

PS Also consider the part of the Wikipedia article that was already mentioned.

Answer 2

I worked with people at IBM who built PL / I optimization compilers and PL / I checkout compilers. I liked using PL / I, but it was a classic example of a “second system effect”: very complex and awkward, albeit powerful. Imagine a closed source version of Perl and you won’t be too far away. And the key is a closed source . Languages ​​take years, and by the time PL / I was stable in the mid-1970s, computing was already moving from mainframes to mini-computers. Then, in the late 1980s, microcomputers took over. The PL / I compilers were not designed for portable devices; they were launched right after the IBM 360, which was supposed to be the last computer that was ever needed - and IBM couldn’t catch much easier, easier than ported languages ​​on new platforms. (Not that I can imagine that IBM is trying to come up with a PL / I port for PDP-11.)

In short,

• No one but IBM is interested in making PL / I a success.
• IBM only cared about its own mainframe hardware.
• PL / It was hard for me to transfer to other equipment.
• By the time it became clear that the mainframe era was over, it was too late for PL / I to catch.

I liked the PL / I, but I did not miss it.

Answer 3

Agree with the above, it was something like a Swiss pocket knife with too many blades.

Regarding compiler availability, a project was developed to build the front end of PL / 1 for the gnu compiler. See http://pl1gcc.sourceforge.net/ (although it looks deprecated from the date of the last announcement). In addition, http://www.thefreecountry.com/compilers/miscellaneous.shtml mentions the Digital Research PL / 1 compiler for MSDOS, which was supposedly free for personal use.