Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.forth > #33483

Re: There is such a thing as Chaos

Show all headers | View raw

Elizabeth D. Rather wrote:

[%X]

>> Does that expectation of the unexpected in regards to Forth actually
>> have something to do with Chaos or is this a design issue with Forth?
>> e.g., your failure to implement safety checks, therefore your program
>> crashed spectacularly.
> 
> Except that it doesn't work that way. As Paul Bennett (who's whole
> business involves highly reliable systems) has repeatedly described,
> Forth programs are exceptionally easy to test thoroughly, and hence have
> a reputation for being extremely reliable.

Thanks for the mention Elizabeth.

> I had many years of experience using native Forths before using Forth on
> OSs. With a native Forth, the programmer (programmer*s*, if it's a team)
> have complete control of the software, and since it is so simple, they
> can test it thoroughly. Introduce a host OS like Windows and all bets
> are off. There's too much code with too many authors, and no adequate
> test procedures. I remember when Windoes NT came out, it was reputed to
> be the most secure Windows ever. It was released with over 65,000 known
> bugs. Chaos.

Interestingly, I am currently reading a book called "The Art of Readable 
Code (Theory and Practice)" by Dustin Boswell and Trevor Fouches. Most of 
their code examples are in the C, C++, Java, Python and HTML languages. Just 
got to the section on generating tests.

One of the reasons native Forth on a processor is so good is just because 
you have complete control. Additionally, Forth programmers are not having to 
keep in mind the need to use the "Safe Sub-set" of a language. You can begin 
with a small, fully understood and very basic environment and grow it to the 
needs of the application. You do not rely on one humungus compiler but on a 
tool-box full of very small and simple compilers that are selected for the 
precise purpose for which they were created. If you do not have a compiler 
specific to your needs it is often just a small effort to creat what you do 
need.

When it comes to testing we do not usually have to sit down and create even 
more code to accomplish a test of what we have just written. We may do 
eventually but it is not necessary at that instant it was written as we can 
just put the values on the stack and run the code. I go much further by 
insisting on Static Analysis, Functional Testing and Limits Testing for my 
code (the Limits Test is a concerted effort to get code to mis-behave if it 
will). By this end, it will keep the chaos portion in the hardware (random 
failures) rather than in the software (which tend to be systematic 
failures).

I am presenting a paper at this years EuroForth on the subject of "High 
Integrity Systems Component Oriented Development Environments" where I will 
go further into the techniques I use. Perhaps I will see you in Majorca.

-- 
********************************************************************
Paul E. Bennett IEng MIET.....<email://Paul_E.Bennett@topmail.co.uk>
Forth based HIDECS Consultancy.............<http://www.hidecs.co.uk>
Mob: +44 (0)7811-639972
Tel: +44 (0)1235-510979
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
********************************************************************

Back to comp.lang.forth | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

There is such a thing as Chaos Jason Damisch <jasondamisch@yahoo.com> - 2014-09-20 10:54 -0700
  There is such a thing as Chaos Mark Wills <markwills1970@gmail.com> - 2014-09-20 11:29 -0700
  Re: There is such a thing as Chaos humptydumpty <ouatubi@gmail.com> - 2014-09-20 12:54 -0700
    Re: There is such a thing as Chaos Jason Damisch <jasondamisch@yahoo.com> - 2014-09-20 13:40 -0700
      Re: There is such a thing as Chaos albert@spenarnc.xs4all.nl (Albert van der Horst) - 2014-09-21 11:29 +0000
        Re: There is such a thing as Chaos rickman <gnuarm@gmail.com> - 2014-09-21 07:56 -0400
  Re: There is such a thing as Chaos "Rod Pemberton" <buzzoff@nonamewhichexists.cmm> - 2014-09-20 18:32 -0400
    Re: There is such a thing as Chaos "Elizabeth D. Rather" <erather@forth.com> - 2014-09-20 13:22 -1000
      Re: There is such a thing as Chaos Jason Damisch <jasondamisch@yahoo.com> - 2014-09-20 17:01 -0700
      Re: There is such a thing as Chaos Paul E Bennett <Paul_E.Bennett@topmail.co.uk> - 2014-09-21 09:48 +0100
    Re: There is such a thing as Chaos Jason Damisch <jasondamisch@yahoo.com> - 2014-09-20 16:57 -0700
      Re: There is such a thing as Chaos "Rod Pemberton" <buzzoff@nonamewhichexists.cmm> - 2014-09-21 20:49 -0400
      Re: There is such a thing as Chaos Brad Eckert <hwfwguy@gmail.com> - 2014-09-27 18:06 -0700
  Re: There is such a thing as Chaos Whammo <sidciavic@gmail.com> - 2014-09-22 20:16 -0700
    Re: There is such a thing as Chaos rickman <gnuarm@gmail.com> - 2014-09-23 01:08 -0400
      Re: There is such a thing as Chaos Mark Wills <markwills1970@gmail.com> - 2014-09-23 00:28 -0700
        Re: There is such a thing as Chaos hughaguilar96@yahoo.com - 2014-09-23 20:19 -0700
  Re: There is such a thing as Chaos Brad Eckert <hwfwguy@gmail.com> - 2014-09-27 17:34 -0700

csiph-web