Re: 7-ZIP for Mac OSX ???

[mdj <mdj.mdj@gmail.com>]

On Jul 3, 4:26 am, "Michael J. Mahon" <mjma...@aol.com> wrote:
> mdj wrote:
> > On Jun 30, 7:09 am, "sfahey" <sfa...@a2central.com.remove-k08-this>
> > wrote:
>
> >>  To: Garberstreet Electronics
> >>  Re: 7-ZIP for Mac OSX ???
> >>  By: Garberstreet Electronics to comp.sys.apple2 on Mon Jun 29 2009 03:06 pm
>
> >> > I don't have a Mac, so someone try this, please.
>
> >> >http://web.me.com/krmathis/
>
> >>Mac users will probably prefer this app.
>
> >>http://wakaba.c3.cx/s/apps/unarchiver.html
>
> >>It's free, open source and a whole lot more versatile.
>
> > What's wrong with plain old Zip ? Works everywhere, doesn't require me
> > to download extra tools.
>
> > And no, I don't care if <insert whiz-bang new tool> get's 2-3% better
> > compression. Wake me up when it's 25-50%
>
> > Not wanting to complain, mind. Just think it's easier to stick with
> > simple.
>
> My thoughts exactly.
>
> Occasional specialized filetypes have special redundancies that
> justify particular compressions (JPEG, MP3, etc.), but virtually
> everything else can be sufficiently compressed with a standard
> app--and the world is much better with fewer "standards".  ;-)
>
> The last thing we need is a bunch of people writing their own
> peculiar data compression programs for public distribution.

Indeed. Also, LZW and "deflate" class algorithms can be implemented on
machines with very modest resources, for example, Apple II's :-)

In order to get better results than those two, you need to spend
ridiculous amounts of CPU time, like for example the Burrows-Wheeler
transform used by bzip2 which is widely used in the Unix community.
Strangely, compression of this type can be parallelised, but that
feature still isn't in the standard distribution.

Beyond that, you've got PPM style algorithms which are effective, but
require the 2+Ghz processing and >1GB free ram characteristics of
modern machines. Interesting stuff academically speaking, but not
pragmatically useful for widespread public use.

Matt