[SystemSafety] Logic

[Andrew Rae]

Peter,
I think it is important, from a practical and political point of view, to
acknowledge the competing demands on time and space in the undergraduate
computing curriculum.

Just to be clear up front - I think there is a persuasive case that a
degree course should have a substantial amount of formal logic in it even
once you've considered the competing demands - but the
practical reason why more time is not given is not that people think that
logic is not important, but that it is not important enough compared to the
competing demands. Just making a case for how
important it is won't win the argument.

As an illustration, the first recourse anyone will have when you demand
more compulsory logic modules is likely to be "okay, what are we going to
take away?" Inevitably the thing that you want to remove
(or give less time to, or be optional instead of compulsory) will also be
something that someone else thinks is _essential_ for a modern computer
scientist.

I've heard very similar discussions focussed on:
   - practical programming skills
   - teamwork
   - ethics
   - basic safety and risk management
   - non procedural programming ("no, we can't just talk about the
principles, they need to be intimately familiar with at least one
functional language and at least one ... ")
   - software project management (including process management, measurement
etc.)
   - human factors and interface design
   - understanding physical hardware and its implications for software
design

That's leaving aside the arguments that students are only employable if
they have used specific programming languages / tools, or understand
networking, or mobile devices, or ....

All of that needs to fit in with the fact that these are kids getting an
education. If they don't have an opportunity to spend a substantial amount
of time throughout the course coding or building cool stuff,
they aren't going to absorb the other things you're trying to teach them.

None of this says you can't teach more logic, but it hopefully makes clear
that the argument isn't _about_ logic. It's about how much can you
reasonably fit into the program, and where is it going to fit.

Drew


My system safety podcast: http://disastercast.co.uk
My phone number: +44 (0) 7783 446 814
University of York disclaimer:
http://www.york.ac.uk/docs/disclaimer/email.htm


On 18 February 2014 14:55, Peter Bernard Ladkin <ladkin at rvs.uni-bielefeld.de
> wrote:

> Derek,
>
> On 2014-02-18 15:36 , Derek M Jones wrote:
> > Formal logic is all well and good for small systems but it does
> > not scale.  I think you should explain this important issue to
> > your students.
>
> Sorry, I think that's nonsense. So I won't be explaining it to any
> students (except as an example of
> nonsense).
>
> The Praxis/Altran publications referenced by Martyn, including the
> Tokeneer work, are some of the
> evidence as to why it's nonsense. Others are referenced in the paper (see
> comments on requirements
> consistency checking. Michael Jackson references papers by Mats Heimdahl
> and David Harel with
> pertinent examples in aerospace critical systems).
>
> PBL
>
> Prof. Peter Bernard Ladkin, Faculty of Technology, University of
> Bielefeld, 33594 Bielefeld, Germany
> Tel+msg +49 (0)521 880 7319  www.rvs.uni-bielefeld.de
>
>
>
>
> _______________________________________________
> The System Safety Mailing List
> systemsafety at TechFak.Uni-Bielefeld.DE
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20140218/808eb8b8/attachment-0001.html>