Re: Is Linux right for Embedded?

[David Bolen <db3l@fitlinxx.com>]

pete@nospam <pete_member@newsguy.com> writes:

> In article <9pfuqg$ikl$1@nh.pace.co.uk>, "Marin says...
>  
> >Everything you say is true, but when talking to people who develop
> >throw-away code, the emphasis should be on time-to-market and reliability
> >rather than long-term benefits such as reduced maintenance or reusable code.
> 
> but but but, code REUSE is what will make time-to-market much smaller.
> 
> Ada is one of the best languages for designing code reuse, which means
> if you want very short time-to-market, then Ada is one of the best choices.

But most of the time, such reuse only occurs on the subsequent
projects, so if, as Marin suggested, you're talking about throwaway
code on a particular project, you expend energy designing/developing
it for reuse but then never get the benefit since it isn't reused.
Instead, it can prolong time-to-market for that first release.

And in some cases, missing that first time window of delivery means
that even if you did built a fantasically designed and reusable
system, the market isn't going to afford you the opportunity to make use
of it since you won't be around for that second release :-)

> Untill they see that code reuse == profit and shorter time-to-market.
> 
> of course, writing packages with the idea of REUSE requires more time
> and effort, but it will soon pays of (in the next project).

Precisely.  But if the primary goal is time-to-market and resources to
rebuild the next time are available, sometimes the prudent business
design is to build immediately and assume a large fraction is throwaway.

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l@fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/