[Rcpp-devel] Mac/Windows/Linux and C++/Java
carp at alias-i.com
Mon May 7 17:17:09 CEST 2012
On 5/7/12 1:05 AM, Michael Hannon wrote:
> Steve Lianoglou<mailinglist.honeypot at gmail.com> wrote:
I spent 1997--2011 developing C and Java (not C++)
on Unix/Windows/Linux and moved to Mac OS X and C++ a year
and a half ago.
> I *am* curious to know what is so nice about Mac OS X.
1. It's a form of Unix, making the shell more integrated
than in Cygwin on Windows.
2. I really like their hardware (screens, weight, keyboards,
trackpad gestures (partly software)).
3. There's currently less malware than on Windows.
4. It plays nicely with iPhone and iPad. I don't even have
a cell phone, but I LOVE my iPad. (I don't know how well Windows or
Linux run iTunes -- Apple's worse than Microsoft in trying to lock
you in and control your content -- I don't buy content from
iTunes, but you need it to sync and setup the iPad.)
5. All the cool kids have Macs :-) (They've taken over
computer science departments and continue to get more
popular in stats, and there's a definite network effect
in OS use -- it helps to have users around to ask
Every platform's easy to develop on if you only need to
get things to the "works on my machine" stage. Every
platform's a pain for developing cross-platform C++.
Especially if you need multiple tool chains for different
Linux seems very crude as a desktop environment compared
to either Windows or OS X. I especially hate its poor
font support (yes, I've had experts sit next to me and
fiddle all the controls, including installing my beloved
Lucida console font). Historically, installing drivers
was a major pain, especially a few years ago when 64-bit
support was minimal for notebooks.
I blame C++ for being a moving target, particularly the
transition to C++11 features, its lack of regulation for
basic data type sizes, and the uneven support of templates
in the compilers. I went through several major revisions
of Java (1.4 to 1.7) and they were all relatively seamless
and backward compatible (only one problem ever in an obscure
member declaration in an XML parsing API that was broken
R's very strict in its compatibility requirements for CRAN.
i386 and PowerPC? Really?
> ... I
> can see that Macs are useful if you have to deal with, say, Adobe products,
> etc. But I've never been able to fathom their appeal for work in math,
> statistics, science, etc. (My wife has multiple Macs, and she loves the
> things, even for her statistics research, but I don't seem to "get it".)
I didn't find Photoshop any easier to use on the Mac than
on Windows, but I'm a fairly casual user.
More information about the Rcpp-devel