Genetic Programming for finding trading system algorithm

I've been thinking about using Genetic Programming (GP) approach to find new trading system algorithm. There is no simple relation between such algorithm and outcome of the system (profit/loss benchmarks on historical data). To evaluate usefulness of algorithm, it have to be tested on real data. So its good target for that kind of automation.

I have been using Genetic Algorithms (GA) and even built some kind of GA framework at the university. Now I'm browsing through resources to refresh basics and find out what is new in the subject. I'm reviewing also GP frameworks to use (choosing possibly the simplest one). Results could be interesting - but it needs some integration work to take off.

Chrome OS - the dawn of net(only)books

Google is making good strategic move to get control and share of netbooks, and maybe nettops market. The incoming era of ubiquitous mobile internet connection for everyone, makes concept of thin clients and network systems plausible. The idea probably will be mainly accepted by mainstream population (look at software usage patterns and emerging fast booting systems from BIOS on desktops for basic web/email operations). If the big G stays behind it so we can expect hardware producers answer. Looking at the Android history it can be massive adoption of "new" OS concept.

There were similiar succesful concepts like Eye OS, but I tented to use separate specific web services best suited to one task at once like email, documents editor and sharing etc. Google already has those applications, so its also reason why it could be available as Chrome OS asset.

Some experts are calling that future is only in weba pplications. But some more advanced or geeky users it could be wrong way. It's ok to have most of tools as online service, but there is still place for lower level native (and offline) platform applications. Either Chrome OS "native applications" will be sufficient then, or Chrome OS will be installed as second system on more "fat" devices.

One thing is for sure - current web applications technology is still evolving, so this is only beginning.

Is OpenOffice good tool for advanced applications?

I had to make some "serious" and personalized reports in "office" format. One time task, no need for fancy reporting software.I had to acquire one of the MS Office licensed copies or just use newest OpenOffice 3.1 which I use everyday. I'm a fan of Open Office as a tool for simple documents and spreadsheets - there is no real advantage in choosing other solutions. But new task was more demanding: preprocessing table of data, making nice template and then generating personalized documents.

Data part preparig was rather tedious - fixing some errors, cleaning and adding missing values. The OO database fits well in the OO environment but it lacks more advanced data operations, like joining two or more tables, grouping etc. So I decided to switch to external database (sqlite in that case) and connect it to OpenOffice through ODBC Driver.

I have tried to make nice report with graphs etc., but I couldn't find a way to make it work with mail merge.
I used some scripting to manually copy template and insert some data, but unfortunately I couldn't set up some details, like print ranges and merging documents into one.

I dropped graphs and just placed more data in simple tables. That kind of document was easy to personalize using "Mail Merge" feature. I was afraid it wouldn't work well, remembering bugs in OO 2.x and hacking to get work around. No more problems this time. Good work OO folks!

It took me more time than I planned due to finding ways to do the work, but I'm still the OpenOffice (and family) proponent :). There are features I didn't even mentioned like possibilities of external scripting in server mode, making it even more universal tool (for hardcore coders).