I saw this article, and my knee-jerk reaction was to look for whatever popular fad's proliferation on HN spawned it, but there isn't any, and there's only one erlang article on the home page at the moment, so I guess this is just an actual erlang article.
Seeing that it's actually a month old, perhaps somebody can explain this?
-- When enabled on the fix_alloc allocator, a strategy
for management of fix blocks will be used.
> I saw this article, and my knee-jerk reaction was to look for whatever popular fad's proliferation on HN spawned it,
I dunno. Reliable backend technology should always be popular if you ask me ;-)
> What is the different strategy?
No strategy or it wasn't configurable. fix_alloc could just dole out fixed sized block and then never return them to the OS. It could be use for say storing atoms. So maybe it went from not having a strategy to having one.
Erlang has a set of allocators. Each one of them has a set of strategies and implementation of those strategies see from erts_alloc docs for more info.
> I dunno. Reliable backend technology should always be popular if you ask me ;-)
Just looking at your 'days registered', I dunno if you were around at the time or not. I don't know if I had actually even registered at the time, but whenever HN would get too popular, or get press from sources that didn't quite fit within the HN ethos, the core community would band together and upvote the entire front page to just being erlang articles, presumably as a self-filtering mechanism.
If you had just come here from Digg, you probably weren't the sort of chap to hang around for a bunch of erlang news. If you were in the minority of those who actually would stay around for a bunch of erlang news, then you'd probably fit right in.
The community is far more diluted nowadays, and I don't even know if we could erlang-bomb the homepage the way it used to be done.
Seeing that it's actually a month old, perhaps somebody can explain this?
What is the different strategy?