3 ms·

You can't compare Constitutions to programming languages in terms of the prevailing consensus on their utility. While the latter has a vast test bed of real wo

You can't compare Constitutions to programming languages in terms of the prevailing consensus on their utility. While the latter has a vast test bed of real world utilization that reveals PL efficacy, the former can only be evaluated from the standpoint of theory, with a heavy heap of ideology.

crazygringo 4y ago
You don't think academics are constantly comparing the real world utilization of constitutional forms...? Allow me to introduce to you, comparative politics: https://en.wikipedia.org/wiki/Comparative_politics https://en.wikipedia.org/wiki/Comparative_politics Lest you think this is ideological or theoretical, no -- it's entirely about collecting data of real-world political operation, how similar or different constitutional structures result in similar or different outcomes given similar or different populations and histories.
- ETH_start 4y ago
  The real world data is extremely limited, given there are only ~200 countries in the modern era, and the huge span of time over which the full impact of a constitution is seen. It's incomparable to PLs, which have perhaps billions of case studies on which comparative analysis can be done.
  - PaulDavisThe1st 4y ago
    Remind me how many PL's there are again.
    - funcDropShadow 4y ago
      Not GP here. I would estimate the number of programming languages is in the thousands, but the number of languages with multiple implementations and long-term maintained programs written in it is probably below one hundred. All the others do not contribute a meaningful body of real-world data for the purpose to learn from programming language design. But they might have contributed knowledge of about the design space of programming languages.
  - crazygringo 4y ago
    Actually the more I think about it, the more useful I do find the comparison between constitutions and programming languages. Because there are probably only ~200 PL's with truly significant usage, and they tend to have a variety of purposes and be used in historically different contexts -- much like constitutions. The same arguments that programmers have over declarative vs imperative is much like the arguments political scientists have over parliamentary vs presidential. Static vs dynamic type checking is much like civil law vs common law. Perhaps the rise of congressionally-created federal agencies is akin to the rise of object-oriented programming? And there's similarly nothing "scientific" in proving that one programming language is "better" than another, but people have strong opinions, and experience will tell you which ones are good ideas to pick for a project and which ones are not. By the way, I'm not sure you know what a case study is -- it's a report written by a person. Nobody's written billions of those. And you don't do comparative analysis on case studies -- a comparative analysis is a case study, where the cases are countries (or programming languages).
edgecasestdio 4y ago
I'm also wary to use programming languages as a metaphor to advance a view about writing constitutions. I can agree with the broad point that inertia is a major cause for constitutions to stay the same over time. But I disagree with the original commenter's implication that more recently-written constitutions tend to be better than older ones due to lessons from the past, because of the idea (perhaps relating to yours) that contemporary political and ideological considerations can introduce weaknesses into more recently-written constitutions, which overlook potential lessons from the past.
- WalterBright 4y ago
  People don't learn very well from the past.

crazygringo 4y ago

You don't think academics are constantly comparing the real world utilization of constitutional forms...? Allow me to introduce to you, comparative politics: https://en.wikipedia.org/wiki/Comparative_politics https://en.wikipedia.org/wiki/Comparative_politics Lest you think this is ideological or theoretical, no -- it's entirely about collecting data of real-world political operation, how similar or different constitutional structures result in similar or different outcomes given similar or different populations and histories.

ETH_start 4y ago
The real world data is extremely limited, given there are only ~200 countries in the modern era, and the huge span of time over which the full impact of a constitution is seen. It's incomparable to PLs, which have perhaps billions of case studies on which comparative analysis can be done.
- PaulDavisThe1st 4y ago
  Remind me how many PL's there are again.
  - funcDropShadow 4y ago
    Not GP here. I would estimate the number of programming languages is in the thousands, but the number of languages with multiple implementations and long-term maintained programs written in it is probably below one hundred. All the others do not contribute a meaningful body of real-world data for the purpose to learn from programming language design. But they might have contributed knowledge of about the design space of programming languages.
- crazygringo 4y ago
  Actually the more I think about it, the more useful I do find the comparison between constitutions and programming languages. Because there are probably only ~200 PL's with truly significant usage, and they tend to have a variety of purposes and be used in historically different contexts -- much like constitutions. The same arguments that programmers have over declarative vs imperative is much like the arguments political scientists have over parliamentary vs presidential. Static vs dynamic type checking is much like civil law vs common law. Perhaps the rise of congressionally-created federal agencies is akin to the rise of object-oriented programming? And there's similarly nothing "scientific" in proving that one programming language is "better" than another, but people have strong opinions, and experience will tell you which ones are good ideas to pick for a project and which ones are not. By the way, I'm not sure you know what a case study is -- it's a report written by a person. Nobody's written billions of those. And you don't do comparative analysis on case studies -- a comparative analysis is a case study, where the cases are countries (or programming languages).

edgecasestdio 4y ago

I'm also wary to use programming languages as a metaphor to advance a view about writing constitutions. I can agree with the broad point that inertia is a major cause for constitutions to stay the same over time. But I disagree with the original commenter's implication that more recently-written constitutions tend to be better than older ones due to lessons from the past, because of the idea (perhaps relating to yours) that contemporary political and ideological considerations can introduce weaknesses into more recently-written constitutions, which overlook potential lessons from the past.

WalterBright 4y ago
People don't learn very well from the past.