I think that a language for building packets should be done before a graphical editor. I would hope that the language(s) for specifying packets (or their attributes) would be more powerful than any graphical interface...

From: Biot Olivier <[EMAIL PROTECTED]>
> ... And if we *really* don't know what to do for the coming 18 months,
>  then we could start developing a graphical protocol (and packet)
> construction tool. I think it must be reasonably simple to define
> basic building blocks with which to construct packets (always-present
> typed fields, TLV fields, variable field). It'd be graeat if we were
> able to drag-and-drop protocol field types to a canvas, define field
> dependencies etc in a graphical interface. Different protocol PDU
> types can then be defined by e.g., assigning specific values to
> fields. Excluding illegal PDUs can be performed by e.g. using the
> "reserved" attribute.





Reply via email to