[agl-discussions] First stab at vehicle signal specification.
mfeuer1 at jaguarlandrover.com
Wed Mar 16 12:04:48 EDT 2016
On Wed, Mar 16, 2016 at 1:50 AM, José Bollo <jose.bollo at iot.bzh> wrote:
> 2016-03-15 21:49 GMT+01:00 Streif, Rudolf <rstreif at jaguarlandrover.com>:
> > Thanks, Jose. Please see inlined below.
> >>>> few remarks:
> >>>> - why a type so fine that int16 or uint16 when min and max are
> >>>> available? integer versus decimal would be enough.
> >>> I am not quite sure if I understand you correctly. Are you suggesting
> >>> rather than specifying integer and floating point types with multiple
> >>> precisions to just use one type for each and leave it to the
> >>> to choose the correct type according to the programming language
> >> Yes you understood my few words with the little exception that floating
> >> point is not decimal. When defining decimal numbers, the count of
> >> decimal cares. Telling floating point number just don't care. But IMHO
> >> should be part of the specification to tell the correct count of
> decimal (if
> >> any).
> > You are talking about the precision? I am not yet convinced if that adds
> > more value than complexity. If you want more precision you could change
> > unit when using integers, for example m/s instead of km/h. Or you could
> > floating point numbers. What example do you have in mind where this
> would be
> > beneficial?
> Hello Rudolf,
> There are two concerns:
> 1) specifying int16 that I think is to precise. As you wrote, it can
> be a detail of implementation.
I humbly disagree. One of the key use cases of VSS is to provide structure
for interprocess communication. Having each signal explicitly typed
provides information to the IPC system (DBUS, etc) on how to encode each
> 2) specifying the precision. An integer is a decimal number with no
> digit on right of the dot. Specifying how many digit are meaningful
> after the dot is not so complicated and is a major improvement when
> designing either program or interfaces. Programs because it allows to
> provide accuracy for numbers derived through formulae (mean,
> integral, ..) and to encode numbers as integers. Interfaces because it
> allows the designer to have an idea of what is meaningful to display.
Using implicit typing would require an IPC to transfer all values in a
generic format, such as in a string, and then have the receiver convert it
to an explicit type (float, double, int32, etc). This adds both to payload
and computing cost, which I believe would negatively impact the performance
of a high-frequency distribution system.
> Just my 2 cents.
> Best regards
> (remainer snipped because I fully agree with Don Ries)
*Head System Architect - Open Source Projects**Jaguar Land Rover*
*Email*: mfeuer1 at jaguarlandrover.com
*Mobile*: +1 949 294 7871
Jaguar Land Rover North America, LLC
1419 NW 14th Ave, Portland, OR 97209
Jaguar Land Rover Limited
Registered Office: Abbey Road, Whitley, Coventry CV3 4LF
Registered in England No: 1672070
This e-mail and any attachments contain confidential information for a
specific individual and purpose. The information is private and privileged
and intended solely for the use of the individual to whom it is addressed.
If you are not the intended recipient, please e-mail us immediately. We
apologise for any inconvenience caused but you are hereby notified that any
disclosure, copying or distribution or the taking of any action in reliance
on the information contained herein is strictly prohibited.
This e-mail does not constitute an order for goods or services unless
accompanied by an official purchase order.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the genivi-projects