Liberation software

[This is a first draft! Will be used for the talk and then refined].

I’m talking this morning at Kitware, a US company (SME) spun out of GE and making enormous contributions to Openness. It creates and distributes a widely used toolkit (VTK, http://www.vtk.org/ ), and the molecular visualizer and builder Avogadro . They’ve invited me to Clifton Park, NY and I’m launching my ideas on Liberation Software.

Liberation Software is the concept that software is a critical element of fighting the digital battle (see Eben Moglen’s Guardian article). There are many things software can do:

• Publication of information (my first HTTP server gave me immense power)
• Discovery and linking of information
• Creation of communities
• Speaking truth to power (e.g. online petitions)

Today I’m going to talk about software that liberates knowledge. I’m a bit rushed because I only got my ideas together at breakfast and I haven’t drawn a good diagram. I’ll summarize the problems, then the liberation. There’s probably a bias on scholarly publication but it’s meant to be general.

Problems

• Gatekeepers. An excellent example is STM publishers who create “paywalls” (usually 30-50 USD to read a single paper for a day). No rights are given and control is absolute. There are reverse walls – the difficulty of getting your voice heard.
• Technological mismatch. Typical examples are binary files (sometimes encrypted). These can often be deciphered with enough effort.
• Missing metadata. The information only works with a given program which “knows what the data means”. Or the creator uses abbreviations on their Excel columns
• Apathy. Many people don’t care to make their material available at all
• Lawyers, Politicians, lobbyists. By far the hardest to tackle. Organizations assert they own the info and will sue/jail people who have succeed in the earlier steps.

Solutions

[Placeholder until I get time to draw it.]

Components :

• Crawler: visits legally visible sites
  
• Format Converter: deciphers what it finds
  
• Semantifier: makes semantic sense out of data.
  
• Repository: easy storage of semi-structured information (RDF/XML/Mongo, etc.)
  
• Services: (often domain specific) addition/filtering/transformation/conversion of information (taxonomy, content-mining, annotation…)
  
• Browser: Human interacts with information
  
• Javascript in browser to extract and transform received information
  
• Amanuensis: “desktop” companion providing close coupling with human. (e.g. AMI2, Avogadro). Lightweight “AI”.
  
• LinkedOpenData Cloud everywhere.
  

The emphasis is to transfer power from the gatekeepers to the human-amanuensis partnership.