I flew back from SF to Heathrow yesterday and, as usual, was hacking on the laptop (with difficulty as Virgin doesn’t seem to give enough space to open a laptop). After a while my neighbour (S) asked:
S: “do you mind saying what you do?”
P: “not at all – I’m a chemist at Cambridge, working in informatics”
S: “I saw you using Eclipse. I’m a software developer for [a company that tests devices] and we use it a lot”.
P: “Ah, we not only use it for testing, we have a rich client (Bioclipse) – it’s great – do you want to see it working?”
P shows Steve Bioclipse, Jmol and he is impressed…
S: “I’d like to help with that – I enjoy software challenges and would like to do some of that in my spare time”
S is a computer scientist and I tell him that Miguel Howard – a compsci – created the stunning graphics on Jmol without knowing any significant amount of chemistry. And we started to discuss the sorts of contribution that can be made towards the Blue Obelisk movement. S would be happy to look at developing machine learning methods, distributed threading, image processing, etc. where the project didn’t require an in depth knowledge of chemistry and where the project is reasonably self-contained.
So – by chance – we have a volunteer who can bring a lot to the project. I shall mail the BO suggesting that we promote a wish-list on the site that such volunteers can choose from. There must be quite a lot of potential volunteers out there in all sorts of disciplines who would enjoy – and that is the key word – contributing to the BO. Because it’s FUN – not a bar on a Gantt chart. I understand S – Often I tackle software problems for relaxation – it’s creative and fun (if there isn’t a deadline). The satisfaction of creating a bit of working software can be enormous (although the pain of debugging it is the opposite!).
So if you think you have something to contribute to BO, have a look at the site, download and try the software and see if it excites you.
I have worked with virtual volunteers for many years. It’s a different ethos from a managed project – you have no right to expect anything from anyone. The demands of real life (RL) must take precedence. Sometimes there are problem with employers or supervisors. And many volunteers offer something and then are never heard from again. But we now have a substantial software and information architecture and there are opportunities in many aspects besides software: BO Data curation, tutorials, advocacy, documentation (not as boring as it sounds if you don’t do it all at once, or if you have a new slant on it), examples, integration, mashups, applications, etc.
http://www.blueobelisk.org
-
Recent Posts
-
Recent Comments
- pm286 on ContentMine at IFLA2017: The future of Libraries and Scholarly Communications
- Hiperterminal on ContentMine at IFLA2017: The future of Libraries and Scholarly Communications
- Next steps for Text & Data Mining | Unlocking Research on Text and Data Mining: Overview
- Publishers prioritize “self-plagiarism” detection over allowing new discoveries | Alex Holcombe's blog on Text and Data Mining: Overview
- Kytriya on Let’s get rid of CC-NC and CC-ND NOW! It really matters
-
Archives
- June 2018
- April 2018
- September 2017
- August 2017
- July 2017
- November 2016
- July 2016
- May 2016
- April 2016
- December 2015
- November 2015
- September 2015
- May 2015
- April 2015
- January 2015
- December 2014
- November 2014
- September 2014
- August 2014
- July 2014
- June 2014
- May 2014
- April 2014
- March 2014
- February 2014
- January 2014
- December 2013
- November 2013
- October 2013
- September 2013
- August 2013
- July 2013
- May 2013
- April 2013
- March 2013
- February 2013
- January 2013
- December 2012
- November 2012
- October 2012
- September 2012
- August 2012
- July 2012
- June 2012
- May 2012
- April 2012
- March 2012
- February 2012
- January 2012
- December 2011
- November 2011
- October 2011
- September 2011
- August 2011
- July 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- November 2010
- October 2010
- September 2010
- August 2010
- July 2010
- June 2010
- May 2010
- April 2010
- August 2009
- July 2009
- June 2009
- May 2009
- April 2009
- March 2009
- August 2008
- July 2008
- June 2008
- May 2008
- April 2008
- March 2008
- February 2008
- January 2008
- December 2007
- November 2007
- October 2007
- September 2007
- August 2007
- July 2007
- June 2007
- May 2007
- April 2007
- December 2006
- November 2006
- October 2006
- September 2006
-
Categories
- "virtual communities"
- ahm2007
- berlin5
- blueobelisk
- chemistry
- crystaleye
- cyberscience
- data
- etd2007
- fun
- general
- idcc3
- jisc-theorem
- mkm2007
- nmr
- open issues
- open notebook science
- oscar
- programming for scientists
- publishing
- puzzles
- repositories
- scifoo
- semanticWeb
- theses
- Uncategorized
- www2007
- XML
- xtech2007
-
Meta
If he’s interested in machine learning and image processing, perhaps he might be interested in something like 2D image -> structure code? It’s actually not much chemistry for a first go — turn it into a graph, recognize a few typical element characters, know that a vertex without a label is a carbon atom.
Other bits are obviously complicated — dealing with overlapping bonds, pseudo-3D representations, etc.
But I don’t think anything in open source chemistry offers even a framework for this yet.
I had a hack at image2structure a few years back. My approach was to try to find consistent local patterns that represented common artifacts of the drawing program. This is rings, “Me”, “COOH”, etc. I reckoned that this would allow identification of core fragments and decorations, especially if people were using the same tool consistently. My problem cam in that the images were mor blurred than they looked – it could be quite difficult to separate characters even after some image processing. Maybe it’s better now.
Mixtures of text and graphics will always be a problem as positioning will be arbitrary.