[Gllug] open source centric ICT in Schools from Sept 2012 ?

[Alain Williams]

On Fri, Jan 13, 2012 at 12:44:58PM +0000, James Courtier-Dutton wrote:
> On 13 January 2012 11:01, Alain Williams <addw at phcomp.co.uk> wrote:
> >
> > I have always believed that open standards are much more important than
> > open source[**]. If you have open standards (ie documentation good enough to
> > allow independent implementation) that is freely available then you have
> > the potential to stop one company (or cartel) from dominating.
> >
> 
> In reality, the documentation of any protocol or file format is rarely
> good enough to implement a new version of software to work with it.
> The document will invariably be interpreted by different people in
> different ways.
> You then need to go to interoperability workshops and do lots of tests
> to make it work fully.

+1

> The best documentation is a full open source implementation, including
> all error paths.

It may be a help, it depends on the quality of the code. I do not think that
code is a substitute for good file_format/protcol/... documentation.

> So, I disagree with your view that open standards is enough.

Not what I said. I said that open standards are more important than open source.

An example of that is the definition of the Unix file system. This is an
API: open/close/create/readdir/... How it is implemented is, at the application
programming level of zero interest. This means that your application will
work on very different file systems: sysv, ext2, reiser, ...

This illustrates that documentation can work at different levels.

Also: if you have a reference base in code then the temptation is to implement
the same thing rather than something that is different and perhaps better in
some way.

> Open standards with an open source reference implementation is really
> the minimum requirement.

Minimum requirement for what ?
However I suspect that I disagree with that statement how ever you answer
the question.

> Well defined interfaces is what is really needed.

Which is what I have been saying.

> This is why XML is quite good, because it forces people to fully
> define the interface between two software components.

Really ?

> Take the new National NHS computer system that failed quite
> spectacularly, never going into the production use.
> The main reason it failed is because the project was too big and had
> too many dependencies.

I am not qualified to comment on any of this. Do you have personal experience
of the NHS IT project ?

-- 
Alain Williams
Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer.
+44 (0) 787 668 0256  http://www.phcomp.co.uk/
Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php
#include <std_disclaimer.h>
--
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug