Sunday, March 20, 2011

Misko in Prague

Misko Hevery, top Google programmer is coming to Prague tomorrow. He is going to give two presentations on Monday and Tuesday. He will focus on the Test Driven Design (TDD) methodology and philosophy.  The technology will be presented on real examples. TDD is a standard Google technology used to develop all their products. Misko is an evangelist and propagator of this technology for several years. He has the hands-on experience as well as the experience sharing his views with developers. The TDD is one of the key elements of Google success. There are not too many opportunities to see presentations about proven top notch technologies delivered by top experts, do not miss this unique chance and join us for the presentations.

We will meet at the CVUT, Karlovo namesti 13. Here is the plan:
  • 21.03.2011 Monday 18:00 19:30 KN:E-301   
  • 22.03.2011, Tuesday, 18:00 19:30 KN:E-107
    • Psychology of TDD: We explain the reasoning behind TDD and  common pitfalls and why it is hard to introduce it into organization. (Slides here:
    • This is the theory behind the TDD demo

Friday, March 18, 2011

Cloud Computing Center

the Cloud Computing Center (3C) is starting. It is going to be a platform for MSc, doctoral students, university staff and industry experts. Platform for all who are interested and enthusiastic about cloud computing. The core of the 3C are PhD students and the target is education, research and commercialization.
The Czech universities are typically focused on basic research with very weak links to the industry. This is not only a problem of the universities. The Czech SMBs do not have enough funding and internal resources to work with universities. The global IT players do not conduct research in the Czech Republic. They view us only as a pool of good programmers. In many cases companies also do not believe the university can effectively help them solving practical problems in their business.
On the other hand the university staff is interested in pure research and academia people do not want to get their hands dirty with the everyday programming etc. The academia people are not looking for
 research funding in the industry they concentrate on governmental funding. On top of this academia people and industry speak different languages, they have problems understand each other. 
3C has the ambition to bridge this gap at least in one segment of IT. If we manage to get university working with industry it will be beneficial for all society, industry and academia too. The best benefit for the university research, I believe, is the motivation and feedback coming from users, from industry and society. The benefit for the industry is also clear, they get the best innovation and technology.
You may ask, what is the difference between 3C and software company? The software companies have business motivations, they need to earn money. They have the urge of executing projects in the most efficient way to stay competitive. This is limiting a space for programmers to test new approaches or develop innovative solutions. The main 3C vision is the research as a driver for innovation, practical implementation and evangelization  of latest technologies, paradigms, frameworks etc. 3C researchers need to prove their quality by writing top scientific contributions as well as deploying and commercializing their ideas in the industry. Similarly, I want to evaluate the students not only based on the quality and number of publications, but also on their impact to industry. After all this should be easy to measure in terms of funding coming from the industry. In some sense 3C should act as a technological start-up.
Next time I would like to point out the main technical directions for 3C.

Tuesday, March 15, 2011

Eucalyptus-next step

We are starting the first jobs on successfully installed Eucalyptus cloud. The primary intent is to use the cloud for image processing, testing new algorithms. The CMP group is using some Terra Bytes of images for testing or compare new algorithms. Typically the algorithm is implemented in Matlab. To ease the provisioning of the tasks we are building standard Eucalyptus Machine Image (EMI) with Debian Linux and installed Matlab. This will allow researchers simply start the image, log-in through ssh and start the script with the experiment. The pictures are stored on our servers. The volumes must be mounted to the EMI instance before the experiment starts. We are in the middle of the initial tests. For the basic management we use the Eucalyptus euca2tools and the Hybridfox plug-in for Firefox browser. We are looking for testers to prove that our environment is stable and accumulate some experience. While doing these first steps we have started to look at ways how to make the process of testing and developing new algorithms easier. Especially the UI for starting the Matlab algorithm, monitoring and collecting the results will help. We are reviewing several tools and libraries, such as the RightScale tools and Gem for Ruby. I will report as soon as we have first batch experiments successfully running.

Tuesday, March 8, 2011

I am on a Jury

I am sensing something is going on in the Czechia. After we started meetings in our eClub I am meeting more and more people interested in startups. There are two activities similar to eClub at the Faculty of Economics in Prague Cach Flow and BizIT. There was a two days competition ipo48. And finally today Pavel Curda announced an API MushUp Contest. In addition I am on the jury. Great, I am really pleased to see around people trying to discover new opportunities and enthusiastic enough to help growing them. I am eagerly looking forward to see the results of eClub competition and API Mushup Contest too. I have also learned a lesson the ipo48 ant the Mashup Contest have English pages. This is encouraging me to switch the eClub web to English soon. Many friends of mine, not speaking Czech, want to to read it and contribute too.
If you are interested do not wait join us, cooperate, work on your own ideas, and most of all enjoy the adventures journey.