Kósa Dávid wrote:
> Since we are networkers, why don't we go and connect the available tool
I think that this is in many ways a worthy long-term endeavor, although
(as others have ably pointed out) there is also some strength to be had
> There are certain steps to be taken.
> 1. Collect the functionality of the available software tools.
> 2. Examine the underlying data structures and create a generic and flexible
> structure that could be an interchange format for all software tool. This
> generic network data structure, that would be supported by SNA software
> packages, would elliminate the need of conversions.
> 3. As a first step toward connectivity, SNA software creators could add
> support for the generic data structure
> After having a catalogue of functionality and a common data format,
> researchers could make much more complex analysises.
Well, there was some discussion of this sort of thing at the latest
Sunbelt. One result of these conversations was the emergence of an
alliance between a number of researchers at various places to (among
other things) form a unified SNA toolkit...elements of this will be in C
and R (aka GNU S), although there will probably be interoperability with
other languages as well (via common standards, though it's still very
early in the game). I do not think that our goal is to produce The
Ultimate SNA Package (TM), but to rather to synchronize our own projects
(several of which have fairly specific objectives) and to see where
things lead. For my part, I have written a draft of a generic class
specification for network data in R, which could probably be extended to
other languages as well. (Ulrik, I shall have to see what you folx have
come up with -- as usual, I see that we are on a similar wavelength.)
Depending on discussions within the group, something like this may
become part of a larger set of standards on which our own tools will be
based; these standards, of course, will be made available in some easily
accessible form to aid other researchers who might want to make use of them.
(The others may have more to say on this matter, and I will leave it to
them to do so. :-))
> From technological point of view, there can be many different type of
> implementation, let me outline just one: SNA programs could be operated as
> web services, therefore a researcher would not have to buy all the software
> packages, but could lease the software and could be charged after the number
> of nodes and the functionality (structural equivalence, power position,
> historical analysis ...), research type (education, commercial, ...) etc.
> Developing an accurate pricing model which is beneficial for both parties is
> a part of the project.
Jonathon Cummings has a web-based network analysis system at
www.netvis.org; you might be interested in checking this out. As far as
pricing models go, my own is fairly simple (0$), and thus I am not too
concerned with whether 'tis better to charge users by betweenness or