[Tinyos Core WG] Unit of time in TOS2

[Cory Sharp]

I'll actually give Rob S most the credit :).  - Cory

On 10/19/06, Joe Polastre <joe at polastre.com> wrote:
>
> Martin and Cory came up with the idea of applying the binary prefixes
> to negative exponents at the Moteiv offices earlier this week.  Cory
> also had a post about it to the 2.0 working group a few days ago.
>
> -Joe
>
> On 10/19/06, Martin Turon <mturon at xbow.com> wrote:
> >
> >
> >
> >
> > I agree that TMilbi is the right naming.
> >
> >  Is Martin L. the first to define negative exponent binary
> prefixes?  This
> > is a cool idea.  It would be good to use an "official" name, or to
> propose
> > the names to NIST/SI or even just on wikipedia.
> >
> >  Martin
> >
> >
> >  -----Original Message-----
> >  From: tinyos-2.0wg-bounces at Mail.Millennium.Berkeley.EDU on
> > behalf of Martin Leopold
> >  Sent: Thu 10/19/2006 12:23 PM
> >  To: tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> >  Subject: [Tinyos Core WG] Unit of time in TOS2
> >
> >  Hi All.
> >  While reading the Timer TEP (102) a question came up regarding the use
> >  units to denote the precision for the timers. The TEP is confusing on
> >  the part on what TMilli means - is it 1/1024 s or is it 1/1000 s (as
> the
> >  name suggests). The TEP should be clear and I believe Cory has already
> >  fixed that, but I think the question is a little deeper - mixing 1000
> >  and 1024 units could in the end lead to misunderstandings.
> >
> >  I'm not trying to say that the timer unit should be ms, but that it
> >  would be elegant if the name reflected what you really get - there is
> >  plenty of good reason for using 1/1024 s instead of ms. My feeling is
> >  that even if it's put in the FAQ, tutorials and documentation, at some
> >  point some poor bloke is going to come along and use the TMilli as ms
> in
> >  a calculation, which is wrong.
> >
> >  So what are the thoughts on this? Is it enough to be precise in the
> >  documentation, or is a name change appropriate?
> >
> >  I ran into Cory and we talked about what the name should be (if it were
> >  to be changed). And resembling the KiB, MiB seems like a good idea, but
> >  unfortunately they don't define negative exponents. So in the spirit of
> >  the kibi/mebi units milli would become milbi and micro would be micbi
> >  and the precision tag names TMilbi/TMicbi.
> >   http://physics.nist.gov/cuu/Units/binary.html
> >
> >  --
> >  Regards Martin Leopold.
> >  Dept. of Computer Science, University of Copenhagen
> >  http://www.diku.dk/~leopold
> >
> >
> >
> >  _______________________________________________
> >  Tinyos-2.0wg mailing list
> >  Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> >
> https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-2.0wg
> >
> >
> >
> >
> > _______________________________________________
> > Tinyos-2.0wg mailing list
> > Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> >
> https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-2.0wg
> >
> >
> _______________________________________________
> Tinyos-2.0wg mailing list
> Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-2.0wg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.millennium.berkeley.edu/pipermail/tinyos-2.0wg/attachments/20061019/fa81a421/attachment.htm