1. 8

  2. 8

    It’s really hard to judge this article since it doesn’t have a timestamp on it; its arguments would be sound for 2005 or so but seem to leave a lot out if read in a modern light.

    I think this argument has been expressed better by others tho:

    1. 5

      I went up one level and the article is listed under 2009. It would be nice if we didn’t have to go fishing for the date. Very important for putting in context IMO.

      1. 2

        Thanks, I’ve suggested adding ‘(2008)’ to the title!

      2. 2

        I agree, timestamp your hot takes people!

        A commenter mentions “Fortran 2008” so maybe that can anchor the piece, but who knows, it might have been floating around since 2005 attracting minor outrage from time to time.

      3. 6

        The first stand-out is COBOL. Unfortunately, its design to be humanly-readable by business people was its downfall. Businesses found that it was possible to hire programmers to look after their computers. Programmers would then gravitate to languages designed for them, rather than their managers.

        I … don’t think that was the major issue with COBOL.

        The “human-readable” syntax turned out to not be very readable. COBOL has a reputation for being incredibly difficult to debug and maintain, in no small part because the syntax is so convoluted.

        And COBOL is not dead, at least as of 2012. There may be some selection bias, but:

        Is Cobol being used in your organization to develop new business applications?

        Yes: 53%

        Given that this article is likely from 2009ish, I’d say the author was talking out of their depth.

        1. 6

          Looking at MicroFocus’ annual report (pdf), the unit that does COBOL- and mainframe-related software is doing $528 million a year if I skimmed it right. That’s just one company, albeit a major one, in the COBOL space. If that’s dead, I wouldn’t mind being even 1% closer to that kind of dead before the end of the year. ;)

          1. 3

            For proprietary COBOL, MicroFocus has practically cornered the market, and their licensing fees are insane.

            1. 1

              That’s why I picked them. I figured it would be representative of the significance of and money in COBOL market.

              1. 1

                That was a field report, not a quibble.

        2. 4

          It’s an interesting read, but I think the problem with this article is just the problem with narratives in general: they oversimplify things to the point that they are often more wrong than they are right. You need to have a lot of data before you can truly explain things accurately. For example, the first line and very crux of the article:

          why the lisp computer language is no longer widely used

          At what point was this ever true? According to my data, the growth in the number of people who can program in lisps has never declined. In fact, it’s steadily risen. Of course, the % of programmers who can program in Lisp versus program in other languages perhaps has possibly declined (my database isn’t good enough yet to make such an assertion), but Lisp in absolute numbers has continued to grow.

          1. 4

            That’s kind of my reaction as well; I’m not sure I buy the premise. Besides Lisp “proper”, the article also seems to be looking at broad families of descendants when determining Lisp to have “failed”, but I’m not sure I agree with the grouping. For example, it judges that Algol successors succeeded, unlike Lisp successors. I can accept that C is an Algol successor, and succeeded, so I agree that Algol has been successful in a sense (through descendants, even though Algol itself was never a big hit).

            But where would you place something like Python? Viewed from the lens of circa-1975 programming languages, Python is a lot more like a Lisp than like anything else of that era… certainly more than it could be called a successor of Algol, Cobol, or Fortran.

            I think some of this is that people put too much emphasis on the syntax, so it’s “curly brace languages succeeded, while paren languages failed”. But there’s a lot more to these language families than syntax. Some of the things that distinguished Lisp compared to its contemporaries in that era were: garbage collection, dynamic typing, reflection, closures, common use of higher-order functions, etc.; and all these have gone on to quite a bit of success.

            1. 1

              Python is a lot more like a Lisp

              This is a good point! And the same would go for Javascript, which is a lot more Lisp-like even though it has C/Java-like syntax.

              IMO think Lisp has wildly succeeded, and its best days are still ahead.