[FoRK] Should Exist: a new UNIX shell

Jeff Bone <jbone at place.org> on Mon Jun 4 06:45:16 PDT 2007

On Jun 3, 2007, at 11:43 PM, Lucas Gonze wrote:

> On 6/3/07, Jeff Bone <jbone at place.org> wrote:
>> doing so, at least once a day I find myself forced to actually drop
>> down to some other tool just because I have some relatively simple
>> --- yet more complicated than is suitable for e.g. bc --- numerical
>> or otherwise complex computation I need to do.  Or I need a data
>> structure more complicated than delimited line-oriented records.  Or
>> I've got to join data that's coming from two different places /
>> APIs / tools in two different formats.
>
> Fair enough.  But I don't see enough of a pattern between these
> examples to know how the shell should change.
>
> A thing about the shell is that it's a crazy ball of mud built up over
> 30 years or so.  If you read the bash source you'll find things like
> patches to handle the limited physical bandwidth in a VT100.  (I
> realize that on FoRK there are probably at least ten people who have
> read the source already, very likely including yourself, Bone).
> There's a lot of value in that, so any solution with a reasonable
> chance of success needs to either be a huge improvement or build on
> top of the original.

You do realize that bash isn't the original by a long shot, yes?  I  
haven't looked at the bash source in some time (since trying to  
figure out how to hack variable disciplines into it several years  
ago) but you're exactly right in that it's a giant hairball.  But  
it's a hairball that's been stolen in bits and pieces from lots of  
places.

I realize there's some value in all that, and all the other shells  
too;  the question is the question of any long-term legacy --- is the  
value in the lessons learned or in the way those lessons are  
instantiated in code.  Particularly for user-facing code, there's  
often a point where the value shifts from the latter to the former...

jb


More information about the FoRK mailing list