Truck tailing

Wednesday, May 28th, 2008 by hinrich

Tailing a truck

Prices for oil are increasing at a faster pace than auto manufacturers keep up with enhancing the fuel efficiency of cars. Certainly one way of reducing fuel cost is to drive at slower speeds. This strategy is useful for small distances where the extra time needed is not an issue. Still, high oil prices could force people to reduce costs by lowering speed and spending more time on the road. For the sake of traffic safety, let's hope nobody comes up with the idea of truck tailing to profit from a truck's slipstream to reduce fuel consumption...

Posted in Green

Fuel control instead of cruise control

Wednesday, May 21th, 2008 by hinrich

Cruise Control

Cruise control is becoming a standard feature in modern cars. However, with green cars such as the hybrid Toyota Prius, the use of the cruise control is quite fuel inefficient - especially in areas with lots of small hills. Here the cruise control wastes energy by trying to keep the speed when going uphill. Why not design a "fuel control" / "gas control"? The concept would be to minimize drastic accelerations and decelerations to keep a constant fuel consumption rather than keeping a constant speed. Of course, such a system should not allow a car to suddenly drive too slow or too fast. Still, this should be technically perfectly feasible.

Posted in Green

Green computing at home

Monday, May 19th, 2008 by hinrich

Wyse Winterm 9235LE

Why does an increase in computing power have to be coupled to an increase in power consumption? Yes, having nice looking interfaces, transparent windows and three dimensional animations are all very impressive. I would rather like to see a simple user interface where the computer focusses on the main tasks. For quite some time I have a Wyse Winterm 9235LE in "always on" mode. When I need to look at my calendar, check my emails, look at the latest RSS feeds, I just turn on the monitor and move the mouse to get the terminal out of sleep mode. And the server of this blog site is a NSLU2 which works like a charm since April 1st 2008: website monitoring service

Posted in Green

The next big thing

Thursday, May 15th, 2008 by hinrich

At work I am responsible for an Affymetrix microarray platform. Over the years I have seen many new chip designs and molecular biology applications appear for this platform. Of course, one major driver are the technological possibilities provided by e.g. miniaturization, increase in computing power, etc. As new molecular biology technologies emerge, biotechnology companies and scientists develop new applications for these technologies. Once there is a large enough crowd that follows, the technology becomes viable and there is time to mature the application / optimize the reagents. However, more and more new technologies initially lack suitable analysis algorithms - partially because they lack data from "real world" examples or are optimized on artificial experiments that do not resemble settings that would be used when applying the technology to answer a biological question. By the time scientists involved in developing data analysis approaches have caught up and one knows the advantages and the limitations of a given technology, the technology crowd has moved on to the next big thing. What is it these days? Tiling arrays for annotation-free, genome-wide transcript discovery? Copy number analysis? MicroRNAs? Deep sequencing? The danger is not only that such novel technologies get over-hyped initially, but also that there is insufficient support for a matured technology later as the next big thing promises technologically so much more...

Posted in Science

Data disaster II

Tuesday, May 13th, 2008 by hinrich

We should not only keep an eye on our short term data security. Also thinking about long term use of data should be an issue. While many people will think about a last will the older they get, only few people will think about what will happen to their data once they die. What data would you want to hand over to a relative - like an old family chair? Or is there no need? Is data really only for the living? Think about inheriting a 500 GB hard disk from your mother - not filled with one big video, but rather with thousands of pictures, emails, text files, etc. Are you going to look at all of these? Or would you rather just keep copying (= preserving) them, until "someday" you might find time to sort them out? I think, we should early on start keeping a central place for access to our online identity (e.g. a KeePass file on your home PC), so that our relatives can access our data. Furthermore, we should have in one very obvious location a folder containing a limited number of files, we consider very important about ourselves, which we would like to had over to be preserved for future generations.

Posted in Computer

Knowledge in PDFs

Thursday, May 08th, 2008 by hinrich


Most of the scientific journals nowadays have electronic versions of their articles available in PDF format. While this makes it much easier to do literature research - textual content of PDFs can be extracted and indexed - it stops at this stage. I see a problem in the fact that PDFs are a very nice format to provide people with a document that looks exactly like a printed version (i.e. it has the same layout), while it is not easy to work further with those documents electronically (e.g. the text of the article is mixed with text from tables, figure legends, page numbers, page headers & footers, etc.). We need to be able to use computer technology (such as natural language processing / text mining) to assist us in producing knowledge based on the individual scientific findings presented in different articles. If the PDF document could be adjusted so that text would be readily available, this could prove to be very useful.

Posted in Science