[TYPES] Binary Reflective Type System for Communications

[Kathleen Fisher]

Hi David,

The work you are describing sounds a little bit like the PADS language  
I have been developing (along with a bunch of others.)  You can find  
information on PADS on the project web site (www.padsproj.org).

The 2006 POPL paper "The next 700 data description languages" defines  
a semantics for PADS and similar data description languages.

The 2005 PLDI paper "PADS: Processing Arbitrary Data Streams"  
describes the design and implementation of the original PADS system.

Many of the examples for PADS are for ASCII data sources, but PADS can  
also describe binary data by using base types specific to a binary  
encoding.

If you have questions or comments, I'd be happy to hear them.

Kathleen

On Oct 15, 2009, at 4:15 AM, David Ryan wrote:

> [ The Types Forum, http://lists.seas.upenn.edu/mailman/listinfo/types-list 
>  ]
>
> Hi,
>
> Over the past six years I have been developing an extensible type
> system for describing the format of binary data.  The core concept of
> this type system is the definition of 35 data types which are used to
> define the same set of data types.  This type system may then be
> extended to with the ability to define new types or define new meta
> data to define new types. This has been created with me having done
> little type system theory other than compiler design, etc.
>
> Having created this self referencing type system I would now like to
> discover the type theory which best explains it.  I am sure that a
> type system which has no external references will have been proposed
> and that other examples of this type of system will be available in
> computer science.  Can anyone point me in the right direction of what
> I should look at first?
>
> If anyone is interested, the following blog entry contains the 35 data
> types and corresponding encoding:
>
> http://blog.livemedia.com.au/2009/09/ietf-and-squeezing-meta-dictionary.html
>
> Additional information is available at
>
> www.einet.com.au
>
> Any help would be greatly appreciated.
>
> Thanks,
> David.