Exit words from Matti Vuori

Hi, “pervasivians” and others,

This department is like any other in the sense that people enter and leave often without any notice. At least many new workers are announced via this blog — which is an improvement over how things were when I arrived. But of those who leave, we rarely hear a thing. Today is the last day of my contract, so here are a couple of last words from me. This is not a proper post mortem, only only some notes that could be of “human interest”.

I joined you 5,5 years ago, for three reasons. First, that was the darkest time ever in the ICT job market in Tampere and I had been laid out from one company, which left little interest in the then very depressed, apathetic private sector. I knew docent Mika Katara, who invited me to TUT to work on a project about development of safety-critical systems for a couple of months. After that the job continued as a teaching associate, with the main focus of working in research projects and in teaching software testing, with a requirement of doing a dissertation on the side. The third reason for joining the department was that I was interested to see what the department does and how they do it. I was particularly interested in joining a workplace with professor Ilkka Haikala, whom I had learned to know a little before, but unknown to me, he had passed recently. That is one lesson to every expert organisation: is is good to have interesting gurus that tempt outsiders, especially at times when there are no fantastic domains of activity.

Now the contract, with a small extension, is done, just as is the dissertation about testing competences that respond to the changes in our environment (the defense following later) – well, actually it is just as much about product development competences, which start in the development of great concepts. Also the local private sector seems very much alive – both in business sense and psychologically it is active and energetic. So it is a good time to leave.

The good things here were the opportunity to use my knowledge in teaching and in research, the new learnings from TUT and Inforte seminars about topics varying from patenting via startups to cognitive computing, and the involvement in developing the university collaboration with Demola in the InnoPilotti project, as well as many other things. The whole world is now hyped about robots and in one project  we reseacher using them for testing.

Oh, one thing to mention is that when I arrived at TUT, Tensu soon asked me to enter, after him, as TUT’s representative the board of Pitky, which I’m currently chairing. It has been nice – and has given the opportunity to arrange the TUT/Pitky collaborative Testauspäivä a couple of times. Now TUT should get a slot in the board as I’m not representing you. TUT understands the importance of local collaboration with industry better than any other university and Pitky has always been an important operative and symbolic channel for that (societies and individuals live off symbols and meanings). “Act local” is the final and brightest measure of science and maintaining good practices is the measure of civilisation.

Ok, so that’s that. Now I’ll need to find work in somewhere else.

Thanks and goodbye to all,

Matti Vuori

http://www.mattivuori.net
matti.vuori@mattivuori.net
Twitter: @Matti_Vuori
050 3605429

Posted in Uncategorized | Leave a comment

Hyvää joulua! Happy holidays!

Laitoksella arkinen aherrus alkaa hiljetä, kun tutkijat ja opettajat suuntaavat joulua juhlimaan. Ahkerin saa vielä huomenaamulla avata luukun 23 laitosta ilahduttaneesta joulukalenterista, mutta 24. luukku taitaa jälleen jäädä jo perinteiseksi arvoitukseksi, ehkä jopa ensi vuoteen saakka.

Hyvää joulua!

Hyvää joulua ensi vuoden kalenteria odotellen!

Adventtikalenterin myötä Tietotekniikan laitos toivottaa kaikille hyvää joulua ja iloista uutta vuotta 2017. Suurkiitos yhteistyökumppaneille kuluvasta vuodesta.

The Department of Pervasive Computing wishes everybody Happy Holidays and a Happy New Year 2017.

Posted in wishes | Leave a comment

Warm thanks to guest lecturers on software testing course!

The software testing course was once again held this fall. As usual, there were guest lectures by people from various companies and organizations.

This year we had great lectures from Arto Stenberg from Symbio, Antti Kervinen from Intel, Tuomas Lunti from Nokia, Kaisa Tirkkonen from Kela and Juha-Markus Aalto from Qentinel.

Warmest thank you for all the guest lecturers!

BR,

Antti Jääskeläinen & course staff

 

Posted in Uncategorized | Leave a comment

INFORTE workshop on Artificial Cognition

Our department hosted the INFORTE workshop on Artificial Cognition on 1.-2.12.2016. The workshop’s main organizer was our professor emeritus Kai Koskimies, who became interested in bio-inspired computing related topics during his last years in active work.

acsemma-kaitsu

Prof. Kai Koskimies opening the workshop

The two-day workshop was divided roughly into two – during the first day the experts gave introductions and tutorials into their topics, while the second day was reserved for applications and more advanced discussions. Both days ended with an hour long panel discussion.

The first talk was given by Prof. David Vernon from University of Skövde, Sweden, as he took us through an “Extreme Tutorial” to Artificial Cognition. Prof. Vernon returned to the stage later in the day, as he had agreed to also take care of  Prog. Giulio Sandini’s topics, as prof. Sandini had had to cancel his participation at the last minute. 4 presentations in 2 days then for Prof. Vernon, we applaude him!

Prof. David Vernon

Prof. David Vernon

The second talker was Prof. Marja-Leena Linne from TUT, who works in neuroscience. She concentrated on the biological aspects of cognition, with her talk “Neurobiology of learning”.

wp_000349

Prof. Marja-Leena Linne

The last talk was given by Prof. Steven Furber, from University of Manchester, whose topic was intelligent machines.

Prof. Steven Furber

Prof. Steven Furber

 

The organizers wanted to call this a workshop instead of a seminar to encourage discussions, and they succeeded in that. The fully-booked workshop had a very lively and participating audience, and all talkers got to answer questions both during and after their presentations.  There also rarely was a straightforward answer to a question within this topic, so one question lead to many more and there was interesting discussions and even debates going on throughout the two days.

wp_000374 wp_000361

The finale for the workshop was the closing panel where the topic was nothing less than “Future of intelligent machines”.  At 4 pm on a dark November Friday, after two days of intense workshopping and powerpoint slides one might expect that there is no energy left to discuss, but so it went that the one hour timeslot was not even enough.

Prof. Koskimies leading the panel

Prof. Koskimies leading the panel

 

Posted in events | Tagged , | Leave a comment

Algorithms, Software design and Project planning, all wrapped into one PhD thesis

Friday 2.12.2016 was the day for MSc Sriharsha “Harsha” Vathsavayi’s thesis defence. The title of the thesis was “Applying Genetic Algorithms for Software Design and Project Planning”.

Opponent Prof. Ivan Porres, kustos Prof. Kari Systä and the PhD candidate Sriharsha Vathsavayi

Opponent Prof. Ivan Porres, kustos Prof. Kari Systä and the PhD candidate Sriharsha Vathsavayi

In practice, Harsha had used said algorithms to automate software design by automatically finding optimal configurations of software architecture patterns. He had then added project planning to the picture by also considering work allocation to teams and scheduling alongside the software design. On top of that, Harsha had added the concept of distance so the approach was applicable to distributed software development where the work would be done at different sites. The work included developing quite a complicated tool to work with the algorithm and all the architectural and project planning concepts.

While the topic of the thesis may seem wide, using the words of the opponent, Professor Ivan Porres from Åbo Akademi, Harsha had “elegantly combined all these themes” to make a coherent thesis.

Harsha had already practiced his defence in the rehearsal defence (arranged and blogged about 2 weeks ago), and we could clearly see that practice makes perfect, as he very smoothly delivered answers to all the opponent had to ask.

harshanvaitos1_pien

Congratulations Harsha!

Posted in PhD thesis, research, thesis worker | Leave a comment

Profes’16: a Norwegian Take on Product Focused Software Process Improvement

At the end of November, our researchers focusing on modern software systems development and engineering headed to lovely Trondheim to attend the 17th international conference on product-focused software process improvement, or PROFES. While our flight there did not go according to plan when a delayed flight rerouted our entire five researcher delegation from TUT out of Stockholm first to Oslo and then, finally, to Trondheim, the conference was amazing!

Nidaros Cathedral by early morning pictured by one early bird member of Pervasive running team

Nidaros Cathedral by early morning pictured by one early bird member of Pervasive running team. Yes! it was THAT dark a seven AM.

We had researchers contributing to four papers (our contributors are emboldened)…

  • Eight Paths of Innovations in a Lean Startup Manner: A Case Study by Mikko Raatikainen, Marko Komssi, Harri Kiljander, Laura Hokkanen, Jukka Märijärvi and Omar Mohout.
  • Log File Analyzing in
Intelligent Transportation Systems Development by Esa Heikkinen and Timo D. Hämäläinen.
  • The Developers Dilemma: Perfect Product Development or Fast Business Validation? by Henri Terho, Sampo Suonsyrjä and Kari Systä.
  • Supporting management of hybrid OSS communities – A stake-holder analysis approach by Hanna Mäenpää, Tero Kojo, Terhi Kilamo, Myriam Munezero, Mikko Nurminen, Tomi Männistö and Fabian Fagerholm

…and a poster

  • Internationally Distributed Software Development: On the Impact of Distance Based on a Case Study by Harri Sten, Hannu Jaakkola and Kari Systä.

So all in all a very solid performance! Especially prof. Tommi Mikkonen saved the day when one researcher fell ill right before the conference. He delivered a perfect talk from his student’s paper.

The conference also provided good opportunities to network with other researchers for example by a Pokemon printing 3D-printer or a magician with a few cards (and a scary Houdini trick) up his sleave.

You can see many great photos of the conference here. Try to spot a familiar face!

Posted in conference, research | Leave a comment

Conference Tour of the Navigation (Sensor) Group

This is a three-in-one report about the conference trips of the Navigation (Sensor) Group in the Department of Pervasive Computing.

Our visiting researcher Jan Rácko made a great study on pedestrian navigation with smartphones and we published together a paper Pedestrian Dead Reckoning with Particle Filter for Handheld Smartphone (Collin, J. T., Perttula, A. S., Parviainen, J. T., Racko, J. & Brida, P. 4 Oct 2016 Proceedings of IPIN 2016 Conference) which Jan presented at IPIN 2016, in Madrid, Spain. IPIN has become maybe the best conference on indoor navigation and it was very good place to show what we are doing here at TUT.

Next week it was time to fly to Budabest, to IEEE International Conference on Systems, Man, and Cybernetics (SMC 2016), where Jussi Parviainen was presenting our inertial dice, with the paper entitled Real-time Implementation Of Dice Unloading Algorithm (Vassilyev, A., Parviainen, J., Collin, J. & Takala, J. 2016 2016 IEEE International Conference on Systems, Man, and Cybernetics SMC 2016)

Random numbers and statistics sometimes involve tedious experiments. In the second half of 19th century there was Rudolf Wolf with 100 000 throws of a single die (result 16632, 17700, 15183, 14393, 17707, 18385), later there were books consisting only random numbers on sale. Then became computers and pseudorandom numbers that save a lot of work. But we wanted to go back a bit because 1) throwing dice is fun 2) pseudorandom numbers are pseudo. So we combined dice and computer using inertial sensors. Obviously attaching anything to dice makes it a bit unbalanced. To avoid performing 100 000 throws for correcting the bias we employed Neumann’s method. The method requires ~3 throws and then the computer gives a one single unbiased result.

Finally, Arto flew to Orlando to present a paper Pedestrian Detection with High Resolution Inertial Measurement Unit (Perttula, A. S., Parviainen, J. T. & Collin, J. T. 2016 IEEE Sensors Conference 2016) at IEEE Sensors 2016 conference. In this paper we used our 32-fold inertial measurement unit array to detect passengers on a bus. Inertial sensors are becoming very sensitive and we are interested on figuring out what kind of events we can recognize from the data. Passenger footsteps were detected successfully and the research continues. The work is related to the Living Lab Bus project.

Postprints of the papers will be in TUTCRIS soon.

Figure 1. Landscape at SMC2016, photo by Jussi Parviainen

Figure 2. IEEE president Barry L. Shoop at SMC2016 Opening Ceremony

Figure 3. Arto presenting a poster at SENSORS 2016

Text: Jussi Collin

Posted in conference, research | Tagged | Leave a comment

Rehearsal defence

The day of one’s thesis defense is a pivotal point in every Ph.D.’s career. It is a day one waits anxiously for  – both as it finally seals the deal on many years of gruesome labor and also because it feels extremely daunting, having to publically answer tough questions on the thing everyone expects the candidate to be an expert on. When our SWEng professor learned that Business Information Managers have a tradition of practicing for that big day he thought hey, such a good idea should also be tried out at our department.
The idea is that the candidate who will be next in line will get to practice defending his thesis, and another candidate whose thesis is also finished and just waiting for examination will be acting as the opponent (a valuable learning opportunity there as well!).

For this pilot rehearsal defence the Ph.D. candidate was M.Sc. Sriharsha Vathsavayi (whose defense will be on Friday 2.12. in TB109, all very welcome to attend) and the opponent was M.Sc. Marko Leppänen. Both Sriharsha and Marko currently work in the industry, so this was a very nich opportunity for both them to get back to the academic mindset and for us to hear what they have been up to since finishing their studies.

reh01

Rehearsal defence is kicked off with a “lectio”

The audience got a very real feel of a defence on Harsha’s topic “Applying Genetic Algorithms for Software Design and Project Planning”, as Marko raised some interesting questions from the thesis, ranging from considering individual resources in disributed software development instead thinking only on team/site level to how usable the presented project/architecture optimization tool is.

reh03

Harsha nicely pondered and delivered his answers to them.

reh02

Harsha listening carefully to what Marko is asking

And professor Kari Systä, who orchestrated the rehearsal, also got to practice being kustos, and succeeded very well in keeping quiet.

In the end there were also some questions from the audience, which is not so common in a real defence.
Good luck to both candidates on their actual defences!

Photos by: Kari Systä

Posted in PhD thesis, research | Tagged , , , , | Leave a comment

Do you know what are Fatfonts? Or steamgraphs?

Neither did I, but then I went to VIS.

IEEE VIS2016 conference was held in October 23-28 at Baltimore, Maryland (USA). With five parallel tracks of visualization and data-analytics related topics, there was much to choose from and lots of running in between rooms.

I participated in BELIV workshop (Beyond Time And Errors: Novel Evaluation Methods For Visualization) to present a paper “Information visualization heuristics in practical expert evaluation”, written by Heli Väätäjä and myself from TUT, Tomi Heimonen from University of Wisconsin-Stevens Point, Katariina Tiitinen, Jaakko Hakulinen and Markku Turunen from UTA, and Harri Nieminen from Fastems. In the paper we presented an evaluation study of 10 heuristics intended for evaluation of information visualization services.

Baltimore's most famous dead poet.

Baltimore’s most famous dead poet.

During the conference I especially enjoyed a tutorial in Visualization Analysis and Design by Tamara Munzner. Slides are available here: https://www.cs.ubc.ca/~tmm/talks.html#halfdaycourse16

Checking the neighborhood...

Checking the neighborhood…

Another great presentation was by the closing speaker Jean-luc Doumont from Principiae, who presented his three laws of communication, including nice tips regarding paper writing and conference presentations. Some of Jean-luc’s materials are available here: http://www.principiae.be/lectures

Local playfield.

Local playfield.

Oh, the fat fonts? If you didn’t google it yet, here’s the link: http://fatfonts.org/

About steamgraphs you can read for example from Tamara Munzner’s slides.

Reporting from the land of hope and glory. And burgers,

Jari Varsaluoma, IHTE/TUT

Posted in Uncategorized | Leave a comment

ICTSS 2016 and Automated Model Construction

A while back I visited the 2016 International Conference on Testing Software and Systems. The conference was held at the Graz University of Technology in Austria. The university is just over 200 years old and right at the city center, very different from the TUT campus.

The main building of the university is impressive, but located on a narrow street.

The main building of the university is impressive, but located on a narrow street.

 

The lecture hall in which the papers were presented was also very stylish. One person commented that he felt like he was giving a presentation at Hogwarts.

The lecture hall in which the papers were presented was also very stylish. One person commented that he felt like he was giving a presentation at Hogwarts.

As is often the case, the keynote presentations had some of the most interesting ideas. Jeff Offutt proposed that test cases should not be inert clumps of code and data as they usually are, but smart actors who know why they exist, when they should be executed and so on. Nowadays this kind of information is in separate test management software, or only in testers’ heads. Gordon Fraser talked about gamification of testing and its teaching. In his Code Defenders game, some players add bugs into code and others try to write tests that detect them. At least early experiences were good: students enjoyed the game and managed to create high-quality test cases. Perhaps something to try out in the next year’s testing course?

Surprisingly many of the papers were related to mutation testing, which I had previously considered a somewhat marginal approach. Probably something I should take more seriously in the future.

My own paper outlined a method for producing test models automatically based on the GUI of the system under test and some pre-existing test cases, a hybrid of two earlier approaches. The outline was all I had when I submitted the paper, but by the time the conference was held, I had a prototype tool ready and was able to present some early results. Unfortunately, my presentation was at the very final session of the conference, so there wasn’t much chance to discuss the topic with other people afterward. The tool is available at http://www.cs.tut.fi/~jaaskel9/murphy.html if anyone’s interested.

A model extracted from GUI alone...

A model extracted from GUI alone…

 

... can grow quite a bit in coverage and quality when a few test cases are added into the mix.

… can grow quite a bit in coverage and quality when a few test cases are added into the mix.

Text by Antti Jääskeläinen

Posted in conference, research | Leave a comment