[racket-dev] Downloading DrRacket for Mac is hard?

From: Matthias Felleisen (matthias at ccs.neu.edu)
Date: Fri Aug 12 19:22:32 EDT 2011

Eli, I must say I appreciate some of your concerns but some 
I don't comprehend. 

(1) My hunch is that most of our downloaders -- especially
technically unsavy (what's the right word here?) people -- 
download one and only one thing from us. 

(2) We need to accommodate them mostly, not the people who
understand language packages. 

(3) If it works for FF/Mozilla, why should it not work for us? 
Could we write a program that 'steals and adapts' their dowmload
code on a regular base and thus keeps ours uptodate for the next
N years? 

On Aug 12, 2011, at 6:55 PM, Eli Barzilay wrote:

> Yesterday, Guillaume Marceau wrote:
>> Or we can trust that the Mozilla Foundation's user interface
>> designers has already done the experiment.  They have some of the
>> best people of the industry working for them, including Aza Raskin,
>> son of Jef Raskin, one of the original designer of the Macintosh. I
>> see no reason to deviate from their design choice.
>>   [big:]         DrRacket
>>   [almost as large:]   Free Download
>>   [small and grey:]  5.1.2 for Windows, English (US)
>>   [outside of the button, small and light-grey:] All Systems & Languages
> Borrowing too much from the Firefox download is not a good idea, since
> there are some major differences.  In the above, the obvious things
> are "Free" which they have a good reason to emphasize, and we have a
> good reason to not even mention; also, we have some multilingual
> support in drracket, but nothing at the installer level (except for
> the Mac "installers" that have no text ATM).
> But those are all in the cosmetics department.  The main thing that
> I'd like to have is a more reliable way to do the platform detection,
> and the difference between racket and FF is a substantial one: we have
> a lot of downloads, but that's nothing compared to the masses that
> install FF.  So it makes a bit more sense to "just hack something that
> works", and rely on near-immediate user backlash when the guess is
> wrong.
> -- 
>          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
>                    http://barzilay.org/                   Maze is Life!
> _________________________________________________
>  For list-related administrative tasks:
>  http://lists.racket-lang.org/listinfo/dev

Posted on the dev mailing list.