News for hackers. Stuff that matters.
Re:Bootstrapping trust in compilers by _lce0 on Tuesday December 08, 2015 @09:46AM (#10696578)
Given that today hardware is cheap and fast, why are we still using compiled languages?

Plus using languages that requires source code at runtime help lower the barrier for newcomers.

Re:Bootstrapping trust in compilers by dave2000 on Tuesday December 08, 2015 @05:01PM (#10699923)
1) nobody cares about newcomers. fuck 'em 2) you're crazy if you think there's no advantage in compiling code. scripts are fast, compile code is faster. we love faster code, don't we kids? 3) if you wanna be a html hairdressing, sit and ponce about with some lovely, lovely html, this week's javascript framework and css. some of us are writing games for complicated consoles, (limited) android devices or emulating system calls in code converted from old, shitty languages into c++.

come back in 30 years.

Re:Bootstrapping trust in compilers by paulddraper on Tuesday December 08, 2015 @11:03PM (#10701835)
Well, ultimately, some programs must be native.

But beyond that, off the top of my head, I'd say the primary reasons:

* performance

* memory usage (think resource restricted environments)

* performance

* no extra runtime needed

* performance

* start up overhead

* predictable performance (think real-time stuff)

* access to ubiquitous native APIs

* performance

Look at programs like Nginx or PostgreSQL.

They make crazy effective performance optimizations that are really only available to native code. Of course, not every program is a web server or a database, but they are compiled for good reasons, and those reasons apply to many other programs.

Re:Bootstrapping trust in compilers by rwallace on Tuesday December 08, 2015 @10:16AM (#10696784)
By all means use Python if that's what you prefer! But it doesn't solve the problem discussed in the article: what if the Python binary has been compromised?
Re:Bootstrapping trust in compilers by _lce0 on Tuesday December 08, 2015 @10:54AM (#10697079)
Yep I agree. I was talking about compiled _high-level_ languages.