Digital Chemistry And The Internet
By Gordon Logan,
Managing Director of Thermo LabSystems Ltd. and Executive VP of Thermo BioAnalysis
In my last column, I put forward the notion that both the process control world and the clinical instrumentation world could be used as barometers for the success of lab automation technology in the coming years. Respondents were, for the most part, very supportive of the views I presented. In this column, I'll endeavour to continue with my forthright approach.
The Internet has become a great talking shop, but like most things, irritating opinion can overshadow healthy debate. I like the comment of Rafe Needleman, a judge of the annual Muddies (bestowed on the world's worst designed Web sites), who said, "There's so much trash out there, we all put our eyeballs on the line every day."
It is clear that the Internet will have a staggering effect on our lives, if we wish it to. According to industry analysts, Internet commerce will be worth over $1.2 trillion in the next four years. It will be interesting to see how far this will impinge on the lab environment. We now spend less time scouring scientific literature than ever before because the Internet has given us significantly faster access to our information sources. I doubt if the publication rate of peer-reviewed work has increased at anywhere near the same rate. Is this so bad? Of course not! Our information society abandoned the concept of "expert" long ago. Now, everyone is an amateur polymath who generally knows a little about a great deal. We work in global teams with appropriate expertise available instantly. Even though the quantity of peer-reviewed work has not increased substantially, Internet tools have allowed us to better mine the available data and improve our general expertise.
To a large extent, the investigative process in the analytical sciences has been reconciled to the digital age. There have been enormous advances in the use of chemometrics, electronic library systems, and molecular and chemical modelling that are now part of general scientific practice. With the explosion of data generation and the flattening of our access to data, we are in need of robust filtering and database management on an enterprise-wide scale.
Many of us are familiar with autonomous agents, where fuzzy logic algorithms are used to assist in decision support. A number of products that use these agents are about to reach us, including IBM's MailCat. This program will pre-select an appropriate receiving mail folder on a computer desktop by "reading" the contents of the mail and sorting it appropriately. Systems that use advanced and adaptive control coupled with artificial intelligence, pattern recognition, and neural nets will come into their own in the not-too-distant future to analyze both small information groups and terabyte data sets. Without these tools, many people (myself included) find that the Internet solves only half the information puzzle: It's immediate, but doesn't provide information relevancy.
Fortunately, when it comes to IT issues such as the Internet, analytical scientists are standing near the middle of Main Street. We use IT as a service to support business endeavours and not as a primary business function. Therefore, we can afford to take a step back and learn from others who are pioneering IT solutions for their applications in other fields. To utilize our data most effectively, our business segment needs the equivalent of commercially available search engines to assist in the intelligence-gathering exercise. As some companies have begun creating these engines, it is clear that we are at the start of a revolution of sorts in the applied sciences.
"Stop right there, Logan!" came a cry from a reader (I trust I've more than one but I can't be too careful). "Not another technology revolution! The last seven have passed through this town and I'm still here!"
OK, OK. Point well taken. What we're dealing with is change in a series of small stepwise movements. The effect they have on our day-to-day operations tends to be gradual and almost always at the expense of technology that has matured and is no longer providing a value-added service.
Here's a simple example: Numerous groups are trying to develop and use commercial electronic laboratory notebooks (ELNs) to improve the efficiency of the knowledge-management process. While these are being developed, the individuality of the bench chemist's premiere tool—the hardbound notebook—is also being preserved. Like all new-for-old technology, truly dynamic thinking and technology alternatives take a while to mature.
After a deeper look, it's easy to see that hardbound notebooks have been preserved because bench scientists always choose the most appropriate information medium for their needs—this choice will always lie with the end user. Ironically, if ELNs that look and feel like electronic versions of hardbound lab notebooks are developed to suit the wishes of bench scientists, they will (of course) fail. This is because there will not be a convincing reason to abandon the traditional tool which has served us all so well for so long. Even as we preserve traditional hardbound lab notebooks, the technology is almost upon us to combine PDA and mobile telephony. This will be done using real-time operating systems and data searching hooked into distributed universal databases of petabyte size, allowing instant retrieval and update.
The role of LabSystems and other suppliers is to assist in the data-transfer process and remove encumbered information systems, easing the route to decisions. The IT business has yet to establish itself as a value-added service provider in the area of decision support—our technology remains better at recovering information than providing information enhancement. Microsoft, asks, "Where do you want to go today?" as one of its marketing slogans. What we would wish for is the next step: To be taken where we wish to go without the IT system asking in the first place.
QED Immediately relevant!
Cheers.
Please send comments or questions to columnist Gordon Logan at glogan@vertical.net.