someone whose career, for whatever strange reason, turned out to involve both legacy code and troubled projects
I don't understand why you keep saying this like it makes you unique. I'd say 90% of programming contracting thrives on cleaning up other people's shit. If not that, it's the beginning 1/3rd, if you're lucky, of a relationship between a client and a consulting firm.
Three thousand years of beautiful tradition, from Moses to Sandy Koufax, you're god damn right I'm living in the fucking past
| [reply] |
And also, it’s fast!
Perl is a lot of things but it's definitely not fast:
https://benchmarksgame-team.pages.debian.net/benchmarksgame/performance/spectralnorm.html
^ Here, even on a benchmark that favors Perl with respect to other competitors like Python and Ruby, VisualWorks Smalltalk somehow outstrips Perl, which is very unusual for any Smalltalk implementation. Of course, "usual suspects" like C, C++, Java, OCaml, SBCL etc. and actually even Node.js—surprisingly fast—are superior to any of these (in execution speed).
It didn’t seem to bother the language designers in the slightest that nearly six thousand source-files which produce hundreds of millions of dollars in revenue each year might simply “stop working” and with no path forward other than heroics.
The PHP development team made a hard choice. If they're getting rid of utter crap from previous versions the sacrifice may be worth it. Of course it would have been even better had Rasmus Lerdorf not, for example, used strlen() as a function hashing mechanism and then chosen function names so as to balance the hash buckets in the first place (and no I am not making any of that up; here he is talking about this ... unusual design choice) but time travel is not available to these people.
My take on programming-languages is very much like that bumper-sticker that you have probably already seen: “COEXIST.” They’re all here, they’re all a little bit different (and purposely so), they’re all well worth studying, and none will ever supersede another.
That is not clear. QBASIC for example is pretty much a nonentity at this point. There were a whole bunch of programming languages from the Soviet era that are basically also nonentities at this point save maybe for computer historians. In the future, when people are extensively cyborgified, they may have far different ideas about communicating with machines than we do currently.
| [reply] |
