I still don't understand the phenomenon of "worse is better".
My current best guess is a riff on Bob Martin's (?) observation about new programmers (noobs) are minted faster than wisdom of the ancients can be transmitted. An ever expanding circle of ignorance, the center occupied by cranky old people yelling "But you're doing it wrong!" to no one in particular.
So when 'C' emerged, amongst otherwise equal options (to noobs), perhaps it appeared to have both the 'portable' bit fiddly bits (vs ASM) and the structured programming bits of Pascal. And natch, noobs wouldn't be immediately aware of the broken glass and tiger traps.
Further, like with all grifts, there's a turf grab for the new hotness. Something along the lines of "Turbo Pascal was for casuals. Real programmers, like you, should use Watcom's professional new 'C' compiler."
--
I was so enamored with object-oriented analysis and programming. A real zealot. Two (different) friends with experience were "meh". I vividly remember thinking "Those old people just don't get it."
Ahahahah. The folly of youth. Of course, they were proven right. (But also wrong in some ways.)
I try to remember young me now that I'm old, eschewing each new hotness, while trying to keep an open mind, gleaning whatever good bits tumble out.
Worse is better: This can be taken a number of ways. In terms of time (and money), "worse but I can afford it and get it today" beats "better but I can't actually get it". "Worse but easier to use" may also beat "better but harder to use". In terms of writing OSes (which was the use of C), the ability to flip bits and deal with raw memory was very useful. A safer tool might have been "better", but it also could have made it harder to write OSes.
The Fran Allen quote seems to be saying that C made all this promising research stop. Bell Labs did not have armed thugs going out pointing guns at peoples' heads. Why did the research stop, if it was so promising? Because the researchers (many, at least) decided it was less promising - or at least, that more promising options were now available. So if C did that, it did it by either showing that previous directions weren't promising, or by opening up better options. Either way, it's rather unfair to mourn the paths that were not pursued and blame C for it.
My current best guess is a riff on Bob Martin's (?) observation about new programmers (noobs) are minted faster than wisdom of the ancients can be transmitted. An ever expanding circle of ignorance, the center occupied by cranky old people yelling "But you're doing it wrong!" to no one in particular.
So when 'C' emerged, amongst otherwise equal options (to noobs), perhaps it appeared to have both the 'portable' bit fiddly bits (vs ASM) and the structured programming bits of Pascal. And natch, noobs wouldn't be immediately aware of the broken glass and tiger traps.
Further, like with all grifts, there's a turf grab for the new hotness. Something along the lines of "Turbo Pascal was for casuals. Real programmers, like you, should use Watcom's professional new 'C' compiler."
--
I was so enamored with object-oriented analysis and programming. A real zealot. Two (different) friends with experience were "meh". I vividly remember thinking "Those old people just don't get it."
Ahahahah. The folly of youth. Of course, they were proven right. (But also wrong in some ways.)
I try to remember young me now that I'm old, eschewing each new hotness, while trying to keep an open mind, gleaning whatever good bits tumble out.