This post is nearly devoid of information. At least an example of what the author means by modeling data about people v.s. stimulating people might have helped. Most problems I encounter don't relate to people at all but maybe I'm to just replace 'people' with 'things'?
I'm puzzled why this is currently at the top of HN. I'm assuming the author must be someone well known who I've missed out on, as that seems to be the only way posts devoid of value push forward around here.
I'm not criticizing you at all. You're of course free to write whatever you want. HN's tendency to upvote stuff from well known people just because they are well known is a little annoyance I have with the site.
If that were the case for me then anything I would write would get upvoted in bulk, but that's just not been the case. I wrote it because I thought it might be useful to some people and presumably it has been for those who've upvoted it. On the other hand, those who have found it worthless have expressed so very strongly in this comment thread. So it goes.
I do agree with bluehex: an example would definitely helped. It seems you have a new perspective on the topic, which is great but the current version sounds more mystic than informative. Here are one most confusing thought: Why are you bringing in people into the equation? Are these end-users? Or programmers? If they are end-users, why do they care what programming paradigm is used?
The idea about mixing OO and FP at different levels is a good one. Have had similar experience.
I still don't understand. If you're writing SimCity ("simulated-people") you should use OOP and if you're writing genealogy software ("data about people") you should use functional programming!? I assume I misunderstood what you meant because otherwise I have no idea how you arrived to that conclusion.
No need to apologize. How do you factor in relationships (between objects)? I feel that with the different kind of relationships between objects, the kind of operations would change, and therefore different paradigms would provide different strengths.
I do think it is the case for certain high profile people (the mega super stars). Regardless, it's not important enough to warrant a drawn out debate here. Sorry to see you're getting such a binary reaction on this post. Like I said before, my concern with it is not aimed at you.
Yeah, "devoid of value" is a little harsh. I apologize for that. But it's not up to the blog author to ensure their post has enough value. In this context that's the job of HN.
It's definitely a very small post(I expected more when I opened it) but I assumed that people upvoted it because they were interested in the discussion that it would produce. There's definitely plenty to say on the subject of functional vs object-oriented programming, even if this article doesn't happen to say much.
I do it fairly often depending on the topic. There are topics for which I am more interested in the discussion than in the actual article.
For an article like this I'm split. I'm a passionate programmer and would love to read a good article on the tradeoffs between functional and object-oriented styles. I'd also love to read a good discussion about the two styles.
For some topics, say, a site being down, I don't really care about the discussion in most circumstances.
For many other topics, I'm more interested in a good discussion about the topic than I am in any particular article on the topic. It's harder for me to articulate which topics those are, but I know them when I see them.
Simula is regarded the first object oriented language and the purpose of the language was simulation.
The idea of using an object oriented approach to simulation goes back to at least the sixties.
The way I read his post is that working on sets of information tends to be easier in functional languages event based simulations tends to be easier in object oriented languages.
