Sunday, July 27, 2014

Bristol Design Workshop - NHS Citizen Conference 24th and 25th July

Notes

Background

To design something that connects the large number of public interested in NHS with the small number of Civil Service staff.
Discover – How you can get involved in discussion
Gather – When we want to say what needs to be discussed, gather people together. Solutions, Recommendations – how things can be better.
Assembly – Meeting every 6 months – discussion with board there amongst people to discuss the issues that have been discovered and gathered.
Presenting technology, interacting with citizens and with designers of the technology.

Discover

Bit of the whole system that connects people together. Including finding the people that can assist the offline people. Discover is the bit where NHS listens and learns.
Brings all the data together including from within NHS Citizen and from outside. For personal data stores – person decides what data to put on and who to give permission to access the data. Charities and community groups can help to find staff, volunteers, and friends and family who are online – to enable them to put their data and permissions online if they want to.
Community map of people who are on NHS Citizen including how to reach online connector people who can link to the offline people. This could be whole community.
Can be presented as Data Stories – helping people to make sense of data. If people discover a problem, many people may also notice that problem. These can be flagged and gathered – Gather flag.

Data

Related data is data not created in the NHS Citizen system. This is a shared resource inviting people to participate. It isn’t taking data or surveillance. This is a way of getting data into the Discover system either by direct participation or by searching the web for relevant information to create data stories. Does so via machine learning, natural language processing, and human involvement e.g. data journalists. This is to extract meaning and stories e.g. from Twitter.
This is pulling data from the web and converting it to meaningful information. I suggested pushing the data into the Discover system via such things as Twitter posts with a specific hashtag, texts and emails with a specific subject line. This could encourage people who don’t have computers or want to do this on the fly whilst busy and mobile, or who don’t want to use the Discover system (at least yet).  People using these tools that are outside Discover could receive messages from Discover to confirm their registration or information is accepted and give information about where it has gone in Discover.

NHS Gather

MyDesk – Discussion spaces – Tools for debating and prioritising ideas – To avoid flaming posts displayed as - Left side of window Support idea – Right side Oppose idea. This is a crowdsourcing tool.
The system for these ideas needs to be simple and quick, otherwise people don’t get involved.
Process is –
Exploration – Closing Statements – Move Forward – Locally or Nationally
Could have Community Moderators
Assembly meeting will report back if ideas or problems are being ignored by officials not being accountable. People will be able to search so they know whether to raise a flag or if this has been flagged by others.

Assembly

This is a 6 monthly meeting where the most important national ideas will be raised and discussed amongst people including the board.
The most important and popular ideas as discovered, gathered, and voted on would go to the Assembly and/or other relevant bodies to resolve them.
Discussion on what are the problems and solutions to ensure inclusion. I raised problems of cuts to libraries, advocacy, support workers, legal aid. Especially for people who don’t have the technology, excluded, poor, and/or don’t know how to use the NHS Citizen technology they are unable to express their views. Solutions I raised include Healthwatch, service user and community groups to provide a ‘bridge’ (someone else mentioned that idea too) A disability advocate raised issue of Equality Impact Assessment.

My Video interview – 2 minutes 19

Bristol Design Workshop- YouTube- was interviewed at this event in Bristol 25th July- https://www.youtube.com/watch?v=lEvHyTnqrpw

Sunday, March 16, 2014

Stephen Wolfram Wants To Make Computer Language More - Popular Science

I'm going to look into this to see if it's relevant to my idea - New post on my Computing Research blog - MOCHA - Massive Online Collaborative Hierarchical Application - http://userdrivenmodelling.blogspot.co.uk/2014/03/mocha-massive-online-collaborative.html - Semantic Web

Stephen Wolfram Wants To Make Computer Language More Human - http://www.popsci.com/article/technology/stephen-wolfram-wants-make-computer-language-more-human - Article and Video 12 minutes - demonstration

"At South By Southwest, programmer and scientist Stephen Wolfram showed off more of his Wolfram Language software."

"Last month, Stephen Wolfram--scientist, founder of Wolfram Research, and the closest thing Big Data has to a rock star--unveiled Wolfram Language, a "symbolic" computer language more than 25 years in the making."

There is a 12 minute video presentation I will watch. This looks like the kind of highly structured search but human language based search that could link with highly structured but human language based code I am working on, in order to provide a collaborative application of related nodes/objects.

Wednesday, March 12, 2014

MOCHA - Massive Online Collaborative Hierarchical Application

My new idea based on my PhD research - MOCHA - Massive Online Collaborative Hierarchical Application

I literally dreamt up this idea this morning, writing it down a scrap of paper when I woke. This idea is based on my PhD (links to that below). The idea is for a Semantic Web Hierarchical Programme - run by a search algorithm, which is expanded to provide the services of an interpreter/compiler.

This is a particularly appropriate subject on the 25th anniversary of the World Wide Web -

BBC Video-Sir Tim Berners-Lee: Web needs its own Magna Carta - http://www.bbc.co.uk/news/world-26536354 - Sir Tim Berners-Lee on world wide web at 25

BBC- Sir Tim Berners-Lee: World wide web needs bill of rights - http://www.bbc.co.uk/news/uk-26540635  - Berners-Lee seeks web 'Magna Carta' - Video & Article

What I want to research is how to provide structured Semantic Web code information to a web search algorithm. The code would be in a combination of English language and mathematical equations, the same way as I used for my PhD research. The hierarchical aspect of this is that from the root node/page, every other page would have a position in a tree. Each node would contain the following information -


  • A name of the page
  • List of Sub pages if any
  • Equations/Code if any
  • Values if any
This would be scalable to very large trees.

As a simple example we could have page
  • Name - 'Energy'
  • Sub pages - 'Mass' 'Light Speed'
  • Equations - 'Energy = Mass * (Light Speed) Squared
  • Values - Null

Sub pages
  • Name - Mass
  • Sub pages - Null
  • Equations - Null
  • Value - 3 (Kg)

  • Name - Light Speed
  • Sub pages - Null
  • Equations - Null
  • Value - 300,000 (Km/h)

Of course the algorithm would need to deal with units in the long run, but leaving that aside, the algorithm just needs to follow a logical path through each page, and each page points to the next page.

At the end of a branch the algorithm needs to return to the root of the branch and move to the next branch (if any). That algorithm would work for any sized tree (hence would work for massive trees, as was the case for my PhD - 2012).

What would then be possible is to enable a search engine such as that which ranks pages to instead perform this search and calculation. As an example a calculation could be performed and continuously collaboratively refined on the predicted cost of a complex engineering project. Each participant would enter their own data on a page or branch. Whenever their figures or equations change, this would cascade into the overall predicted cost.

I'm going to work on expanding this post and providing updates in this blog until I have enough information to turn this into a paper and/or presentation.

Below are the main links to my previous research that this idea is based on - 

My PhD thesis on University Repository at http://eprints.uwe.ac.uk/17918/ - User driven modelling: Visualisation & systematic interaction for end-user.

Journal paper JVLC subscription required -  http://www.sciencedirect.com/science/article/pii/S1045926X12000572 - 'User-Driven Modelling: Visualisation & Systematic Interaction for end-user programming'.

Journal paper JVLC open access - http://eprints.uwe.ac.uk/17817/ - 'User-Driven Modelling: Visualisation & Systematic Interaction for end-user programming'.