The presentation first covers the use of the laboratory knowledge management system SMIRP for nanotechnology applications during the period of 1999-2001 at Drexel University. The exporting of single experiments from SMIRP and publication to the Chemistry Preprint Archive is then described followed by the evolution to Open Notebook Science in 2005. Abstraction of semantic structure from ONS projects in the areas of drug discovery and solubility is then detailed as an efficient mechanism to provide web services and machine readable data feeds.

This was a terrific opportunity to tie together my current ONS projects with my work in nanotechnology about 10 years ago, when the focus was to capture laboratory information in a structured format so that autonomous agent could begin to replace human workflows. I found it really interesting that the most active workflows back then were related to processing reference information. It took a team of students to find, photocopy and scan many of our key papers, with all the problems that come with training and managing new students. Today, obtaining relevant papers and extracting metadata is not so much of a challenge with tools like Mendeley. I ended the talk with a mention of our use of Mendeley tags to share dynamic links of article collections.

Another important development over the course of the past decade is the availability of free and hosted tools to easily communicate research. This includes wikis, blogs, Nature Precedings, institutional repositories, Google Spreadsheets and many others. It also includes some failed attempts like the Chemistry Preprint Archive.

I didn't anticipate in the late 90s just how crucial openness would prove to be for the evolution of the automation of the scientific process. It isn't my impression that there is currently a consensus on this point. Obviously it is possible to leverage automation in very clever ways for private use. But I think that exponential impact requires very low barriers to contribution (human or not) that can only be achieved with openness and transparency.

I have been very impressed with the ideas and projects discussed at this conference. Open sharing of nanotechnology data and integration of resources are clearly high priority items for many in this community.

As we have shown with our Reaction Attempts and ONS Solubility projects, abstracting meaningful semantic structure is necessarily field specific. One of the exciting opportunities to result from this meeting is finding ways to interconnect our solubility dataset with the nanotechnology community resources. I have met with a few people who would like to collaborate on this and I will be sure to report on our progress.