lua-users home
lua-l archive

[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index] [Thread Index]

On 5/19/2011 12:43 PM, Isaac Gouy wrote:
> It's unfortunate not to notice that those "Which programming languages
> are fastest?" pages, say again-and-again that isn't the question being
> answered.

I can't get to the site at all right now (don't know if it's down or if
the connection I'm on isn't good enough -- public Internet access isn't
always reliable), but I've seen the disclaimers all over the site, sure.

And I get where the sentiment comes from. The best way to find out how
fast a language is is to implement your program in it.

But when it comes right down to it there are too many languages to try
to be able to DO that, and the only reasonable way to narrow that list
down is to look at benchmarks. You're lying to yourself if you think
that a disclaimer means that people won't use the benchmark to determine
which languages are fastest. If performance is an issue for someone,
then they need to see things like the Shootout to help eliminate
obviously bad options. Even if the list narrows the tests down to 2-3
languages, it's useful to help people find the fastest languages. Period.

And by deciding to exclude LuaJIT, you're deceiving the next generation
of developers into thinking that Lua isn't a contender, when (for a lot
of cases) it clearly should be on the table.

It's not whether we "like" the choice or not. It's whether the choice
helps people. It clearly doesn't, and may hurt them.