In my day job, I get to see a lot of innovative geo-related software and services, and the O’Reilly Radar does a great job of tracking innovations in this space. SimpleGeo, WeoGeo, ESRI, Loki, Cloudmade, Quantum GIS, GeoCommons, etc are just some examples of companies/organizations/open source projects doing very interesting things around geospatial data of all kinds. There are a number of good open source efforts around geo-data and visualization, and I am almost certain I am missing a ton. These toolkits allow people to do interesting things.

So where am I going with this? Somehow there seems to be a lack of similar interesting things with scientific data. Admittedly that is a gross generalization, but outside of things like Rich Apodaca’s many projects, there don’t seem to be that many API driven, mashup oriented scientific applications out there. One would think there is more interest and more potential users of scientific data than geospatial data, or maybe I am just too biased. Mapping data is more general and interpretation is somewhat more simple. Geo-data is easier to mash into other data, and with all the Maps APIs out there (and Google’s role in making Maps so accessible should not be minimized either), it is no wonder that there is so much interest in mapping. But I digress. What I want to see is a lot of apps and services around scientific data, and not just targeted at other scientists, but developers in general. I think in the sciences we tend to think only other scientists can handle scientific data, but there are enough smart people out there who, given the right data sources, can do some pretty cool things with data. I wonder if the problem is that we first need to have a science data platform. I don’t think our current data sources do a good job of being a platform for people to build upon.

Related articles by Zemanta

• Great Introductory GIS Tutorial And Screencast Seriesm, Discovering Qgis – an open-source mapping app for windows and macs! (rafer.tumblr.com)
• Geographic Information Systems (GIS) Software Development Firm Avencia Incorporated Announces New Company Name – Azavea Inc (eon.businesswire.com)
• Quorum Introduces Latest Version of Geospatial Information System – Quorum GIS (eon.businesswire.com)
• SkyHook and SimpleGeo Present SpotRank, Now You Can Always Find Where The People Are (radar.oreilly.com)

The other day I tweeting a link to a GigaOm article on How everything is becoming a game. In the tweet I jokingly referred to Matthew Sobol, the creator of the Daemon (if you haven’t read the book, do it now), but in reality this trend towards the gaming experience might just a very real one. In the sciences we have various screensaver products like folding@home that try and engage the user and even more directly we’ve seen foldit, an attempt to make protein structure a true game (foldit could be a great starting point for high school teaching), but when Matthew Ingram writes about the game experience, he is not writing about using video games per se, but rather the system of rewards and points that is such a core part of game play.

In a similar vein, Lee Sheldon, a professor at Indiana University has introduced the gaming experience in his game design courses. For game design, it’s not that much of a stretch. After all game designers should understand those systems to implement them in their games, but what about the sciences. Is there a mechanism by which we can reward students for achievements? For example, you could have a choice of assignments, with a varying degree of difficulty and different achievement scores. You earn these micropoints along the way and in the end you tally up your score and then perhaps a regular grading system could come into play. This is just one possibility. In a response to my tweet on Friendfeed, Ruchira Datta points to Stuart Yeh who has shown that rapid assessment (e.g. with immediate feedback and leveling up) is a very effective way to enhance student performance. I have a feeling this is not the last we’ve heard of this topic.

Quite honestly, I like the idea. It keeps you engaged, and if done right, it could be a load of fun. I’d encourage teachers to try it out.

Related articles by Zemanta

• Professor Ditches Grades For XP System (games.slashdot.org)
• Build Your Own Sheldon Syllabus (terranova.blogs.com)
• A School Where Kids Are Taught Through Video Games [Education] (kotaku.com)

In CACM there is an interesting article on assessing computational thinking. As you know, I am a fan, but have always wondered about how it might be taught from an early age. In this article Judy Robertson talks about a test piloted with 11-12 year olds that uses visual systems to assess computational thinking, e.g. simple Boolean operators and identifying conditions. In her own words, the goal of these tests is to see if the “viewer can identify rules that govern the behavior of members of different classes and then extrapolate from these rules”. Fascinating stuff and it got me thinking. How could we teach scientific thinking to 11-12 year olds and what would that mean?

In my opinion scientific thinking essentially should follow a path that teaches children to do the following

• Ask questions about natural phenomena
• Try and come up with a theory that might be the reason behind it
• Test it

I am not an educator, so I don’t have any bright ideas about how we would teach that. But I do know this, when I was a kid my dad brought me books about a couple of kids who were always curious about things, and encouraged me to try and find out how things worked. I am convinced that those early teachings (I was 5) are why I got so interested in science and curious about trying to figure out how things worked. If we can get kids interested in that, we will have a population that doesn’t have to be scientists, but has the core principles of inquiry built into their thinking. It will result in a populace that might be more skeptical, but also more enlightened. Maybe I am biased but I think we will have a better society. It doesn’t mean we all become analytical automatons, but there is nothing wrong with curiosity, and the ability to try and piece together puzzles. So maybe we should be pushing for scientific thinking education, not just teach kids about the wonders of science.

Related articles by Zemanta

• Testing 1, 2, 3: Creating Quizzes to Gauge Science Knowledge (learning.blogs.nytimes.com)
• Jon Udell’s anti-geek manifesto (mndoci.com)
• Playing games to learn? (annmic.wordpress.com)

Brad talks about trends in computing and software that have allowed data-driven companies like Flightcaster to get to market faster. He breaks down these trends into three chunks

• Storing data
• Processing data
• Learning from the data

I won’t go into the specifics, but you will hear about many of the patterns we have become used to, ranging from distributed, elastic computing infrastructures to finding appropriate data solutions to fit your needs and problem, to the advent of open source solutions to process large data sets. However, it is in the context of learning that he points out a significant challenge, one bioinformaticians and computational biologists are faced with today.

The data renaissance does present some difficult issues. There are not many places one can recieve a good education on working on these problems at large scale. Scaling our modeling and optimization algorithms is hard. We need to figure out how to partition and parallelize, or sometimes trade speed and scale for approximately correct calculations. Another issue is that we are often using simplistic models, albeit with pretty good results in many cases. We would like to move toward a deeper approximation of real intelligence.

Hopefully this will change in the coming years and we will not only do great science, but do it with the right, appropriately scalable tools.

I would probably add a fourth dimension to the three chunks aboce; sharing the data. Data sharing is critical to scientists. Data sharing at scale is a challenge that needs to be addressed using best practices in data delivery, data storage and data availability. APIs and open data are the name of the game.

I am not a fan of blindly trying to mine data, but data does make our science better. The better we get at harnessing it, the more successful we will be in doing out science.

Related articles by Zemanta

• Don’t move that data (mndoci.com)
• Scalable Informatics introduces new storage cluster (insidehpc.com)
• In science, data is nothing without purpose (mndoci.com)

On a related note, check out this talk by Paul Borrill on Rethinking time in distributed systems (slides and abstract). The slides that are of relevance here are slides 29, 30 and 31. Paul reminds us that we need to change the way we think about infrastructure, especially how large systems evolve and fail constantly. The approaches to address these challenges are those that most people who operate at scale understand well. Approaches like decentralizing everything, about building systems from independent, autonomous units that can be cloned and distributed at will. Perhaps most importantly, and this is something I see all the time from people who get it, require systems to manage their own configuration, healing, provisioning and migration. If humans are touching your systems, you are not going to succeed and it’s not what they should be doing anyway.

Related articles by Zemanta

• Opscode Equips DevOps to Automate the Cloud (datacenterknowledge.com)

IMO, the days of moving data sets over the wire are long gone. You can move slices around, but entire data sets? Just doesn’t make sense. It’s why I like the idea of data services and APIs that allow you to pull those slices out. It’s also the reason why I love the idea of elastic computing resources that can be spun up close to your data resources. And to really get going, you want your data going to multiple compute nodes on multiple data pipes. All of these things can be done today. They’re just not done as often as they should.

Related articles by Zemanta

• The sequencing market is beginning to shape out (mndoci.com)
• Data platforms for science – From data to work (mndoci.com)

As we think about the future of scientific publishing, with evolving stories, diverse viewpoints, different sources, there is plenty of room for innovation that borrows from things that have succeeded in other areas but providing some unique scientific aspects, and perhaps we can do things that others borrow from. We also have to be cognizant of the three markets; fellow scientists, scientifically literate people (e.g. engineers) and the general populace. What is provided to each must fundamentally be different, especially to the third category. Who’ll be the first to figure it out? Will we ever get to a place where, as a consumer, I head to “Nature Science News” for a roundup of the latest scientific findings (not early stage, this might be possible kind of stories which mainstream media latches on to), but stories with very solid scientific foundations made accessible to the general public? I think we can get there.

Related articles by Zemanta

• ScienceOnline2010 – Trust and Critical Thinking (video), Part 6 (scienceblogs.com)
• Managing Scientific Information Overload: Slideshow (scienceroll.com)
• Science Blogging as a Public Outreach Tool – Unfulfilled Potential or Unrealistic Expectation? (scholarlykitchen.sspnet.org)
• Science Journalism/Communication week in review (scienceblogs.com)