Volodymyr Obrizan: personal website - 25

KhNURE’s Achievements on the Market

2016-09-25T18:27:00+03:00

Recently I came across a recording of a roundtable on the strategy for Kharkiv’s development, in particular IT technologies. During the discussion, KhNURE employees disclose financial figures. See the excerpt from 30:00.

Transcript:

Eduard Samson:
— Look at the volume of industry-funded research topics or contracts. Do you know the figure for your university?
Stanislav Ignatiev:
— Yes, of course.
— And how much is it?
— Last year it was 12 million. Those are state-budget-funded topics.
— No, no, no, industry-funded! State-budget-funded is what I said.
— This year, less than half a million.
— Less than half a million? Well, let’s divide that by 25. Twenty thousand. A cheap car. And this huge university earned 20,000 dollars on the market!

What is this about? University employees get their salaries from the state budget for education and science. They get, of course, a humiliatingly small amount. For example, a full-time senior lecturer takes home around 2,800 hryvnias. But the university can carry out independent commercial activity. In the dialogue, that is “industry-funded research topics.” The law does not forbid a university from going to the market, selling its services or goods there at market prices, and paying university employees market salaries. But in 25 years of independence and a market economy, the university still hasn’t managed to restructure itself.

Richard Sheridan’s Book “Joy, Inc.”

2016-01-25T22:40:00+02:00

I recently read Richard Sheridan’s book “Joy, Inc.” (93 UAH in electronic form).

In the book, Richard Sheridan talks about the history and day-to-day life of his company, Menlo Innovations", where he works as director and founder.

Hands up, who has their dream job? Yeah, a forest of hands. Well then, read how Sheridan did it.

His company does custom software development. One striking feature: they practice pair programming. And not the way you do it once a year, but every day. That is, this company has exactly half as many computers as employees. What’s more, the pairs change every week. This way, after a month, every person on the project knows every part of the system.

How does this affect employee happiness? Very simply: at the very least, you can go on vacation at any moment; you will never be the one person who knows a key part of the system and without whom it’s impossible to do a release. Secondly, with pair programming, the quality of implementation gets much better. In the end, everyone is happy.

Overall, it was interesting to read about this and a lot more. For company directors and founders — strictly required reading.

Summary of the Twelfth Meeting of the "Python at APVT" Club

2016-01-25T16:00:00+02:00

On January 23, 2016, the twelfth meeting of the "Python at APVT" club took place. The meeting was held in room 318 from 11:15 to 13:00.

We discussed the basic principles of project management:

Projects should be done in pairs.
Use open repositories.
Do code reviews.
Write tests.

The meeting minutes were recorded on http://python.mass.education/project

The main question: "Where to start?" Answer:

Installing Python 2.7.
Installing a development environment, for example, PyCharm.
Installing the Google App Engine Python SDK: https://cloud.google.com/appengine/downloads?hl=en
Going through the GAE tutorial: https://cloud.google.com/appengine/docs/python/gettingstartedpython27/usingwebapp

This should be enough for the first week.

Among the proposed projects:

A simple blog.
An online store.
A wiki.
Online checking of Python assignments.

At the next meeting, the teams will talk about their progress.

Good luck! :)

Previous meetings:

How to get news about meetings of the "Python at APVT" club?

Subscription to news about meetings of the "Python at APVT" club: http://bit.ly/python-apvt-reg The mailing list will be the main channel for information about meetings.

VKontakte and Facebook groups: https://vk.com/python.apvt and https://www.facebook.com/groups/python.apvt respectively.

A Q&A site about Python and the club: http://bit.ly/python-apvt-qa

Questions? You can contact me personally: volodymyr.obrizan@gmail.com

The first meeting of the "Python at APVT" club took place. Summary

2015-10-25T16:23:00+02:00

On October 24, 2015, the first meeting of the "Python at APVT" club took place. The meeting was held in room 318 of the APVT department from 11:15 to 12:50.

At the beginning, everyone introduced themselves and talked about their relationship with Python. I was pleasantly surprised to see not only people just starting to learn this language, but also guys with more than 5 years of experience.

The day before, I tried my luck and wrote to Guido van Rossum — the inventor of the Python language — asking him to write a short address on the occasion of the club’s founding.

And Guido replied:

The patriarch’s blessing had been received; we could begin.

Brief notes on what we discussed at the meeting.

Principles of the club’s work

— active self-education versus passive contemplation;
— organizing horizontal connections among club members.

Let me emphasize once again — there will be no lectures or labs. There will be short talks and master classes. The main focus is self-education and communication with like-minded people.

Course on Coursera

On October 27, a free online course starts on Coursera: https://www.coursera.org/learn/python We recommended registering for this course and starting to take it. The level is beginner; no knowledge of the language is required in order to start studying it.

Books

Mark Lutz, "Learning Python", 1280 pp. Market price: 1200 UAH. A very thick book with a detailed presentation.

"Programming for Kids. An Illustrated Guide to the Scratch and Python Languages". An approach to learning the language with illustrations and explanations. I have already done a short review of this book. Market price of the book: 550 UAH.

Mark Pilgrim’s "Dive into Python" and its Russian translation: "Into the Depths of the Python Language". Online book, free.

Kenneth Lambert, “Fundamentals of Python: From First Programs Through Data Structures.”

Software

Installing the Python interpreter: http://www.python.org It is available for any platform. It is preferable to install the very latest version of the language: 3.5. You need to learn this version of the language, but be ready to adapt in some respects to version 2.6 or 2.7. It is important to write code in a style that is supported by all versions of the language.

Recommended development environment: JetBrains PyCharm Edu. This is a free development environment; it includes 50 simple beginner-level exercises for learning the basics of the language.

You can use any other development environments, including a simple console.

Where to ask questions?

In English, you can ask questions on http://www.stackoverflow.com Russian equivalent: http://ru.stackoverflow.com

A mailing list has been created where people can communicate in a closer circle: https://groups.google.com/forum/#!forum/pythonapvt You can ask questions both by email and through the web interface. Please subscribe.

Group photo as a keepsake:

How to get news about meetings of the "Python at APVT" club?

Subscription to news about meetings of the "Python at APVT" club: http://bit.ly/python-apvt-reg The mailing list will be the main channel for information about meetings.

VKontakte and Facebook groups: https://vk.com/python.apvt and https://www.facebook.com/groups/python.apvt, respectively.

Site for questions and answers about Python and the club: https://groups.google.com/forum/#!forum/pythonapvt

Questions? You can contact me personally: volodymyr.obrizan@gmail.com

How Is VKontakte Technologically Better Than Facebook?

2015-07-25T19:46:00+03:00

Few people know that Facebook is a service for feeding the masses shit with a shovel. The shovel is so big that there is enough for everyone, and so there is only one. You have only the illusion of content personalization based on your likes and friends. But in reality there is no personalization at all, and if you want it, you will simply be denied the service of any content whatsoever.

By default, Facebook shows so-called "top-content". Supposedly, these are the posts that are most popular among users. Since top content is generated automatically, it’s obvious that there is nothing interesting there for me personally, because Facebook will never write an algorithm that would guess what I like. To make me click ads more often — they’ll write one; to make me like the posts — no.

Facebook has a "recent stories" mode, i.e. all posts and events in chronological order. Naturally, this mode is more attractive, because I don’t want to miss something important in my friends’ lives that Facebook’s algorithm won’t consider important.

But even here there is the drawback that my friends not only publish information from themselves, but also like other people’s posts, repost, and comment. VKontakte and Facebook have a very useful feature — hide posts by a particular author.

I highly recommend this feature for maintaining information hygiene. A person, as an individual, is what he reads. Over the past few years I have banned 4,000 VKontakte public pages. Now I simply don’t see reposts from those public pages. Despite the fact that I have about 500 VKontakte friends, it’s enough for me to log in once a week for 15 minutes to see all my friends’ news from the past week.

Recently I decided to filter Facebook too. After several weeks of filtering, to my surprise my personalized feed turned out to be broken. They might show me 10 entries, maybe 5, or maybe just one. When I scroll through the posts, this one hangs at the end, and nothing else loads:

The attempt to personalize the content, to opt out of what I’m not interested in, ended in failure. Now I don’t see any content at all. I tried clearing cookies, tried two browsers: Yandex and Safari. Same result.

Technologically, VKontakte is better than Facebook. They don’t have this problem.

Innovations at KhNURE

2011-02-25T11:51:00+02:00

Innovative technologies at KhNURE:

Now security can watch amateur soccer instead of doing their actual duties.

Why is this monitor sitting out in plain view of everyone, and not in a special room under the eyes of a dedicated employee? Why are there only two cameras, not a couple dozen? Why do people freely get into the university who steal mobile phones and wallets from the labs?

Artificial Intelligence Reinvents the Wheel!

2011-01-25T22:07:00+02:00

The other day I saw a link to an amusing Flash thing: http://megaswf.com/serve/102223/ It gives a clear demonstration of the essence of genetic (evolutionary) algorithms.

The idea is simple. You need to make a bicycle so it travels as far as possible. The computer knows nothing about what a bicycle is supposed to be like. It gets fed various bizarre shapes at random. Among the parameters: body size, wheel size, distance between the wheels, shock absorber position. Maybe they’re something else, but judging by the logic of it, something along those lines.

In the early generations the bicycles come out as a total train wreck. The first generation shows a sample of 20 random bicycles. For example, these:

On the left, in the column, you can see experiments with the latest newborn freak, namely: how many notional meters of track it traveled. The most interesting part starts when the generation changes. At that point the very worst ones (the ones that traveled the least) are discarded (“This is Sparta!!!”), and the best prototypes are “crossed,” i.e., they exchange some parameters, for example the body shape or the wheel diameter.

Some interesting variants turn up:

Or here’s an example of a bicycle that looks a lot like a motorcycle:

But if you wait, as I did, about nine hours, you may end up with a perfectly usable vehicle:

Conclusion (every lab report, term paper, thesis, or dissertation has to have a conclusion): genetic algorithms are widely used in many problems of the national and agricultural economy.

The Bunny in the Kharkiv Metro

2010-12-25T15:52:00+02:00

Recognize the bunny?

In the Kharkiv Metro, at Universytet station, a five-meter bunny of ill fate has been installed. If you walk up to it and pet it, it will do you-know-what.

— Where are you headed? — To the metro. — Not meant to be!

Experts, my ass

2010-12-25T11:53:00+02:00

And here it is Saturday again, and once again someone on the internet is wrong.

Browsing Habr in my spare time, I stumbled on a note about the case of some Zhukov. The story is murky; as I understood from the discussions, the dude wrote a little accounting app, took money for installing it on users’ computers, and a year later that app started printing accounting reports with something missing from them. That was how the author motivated users to pay him more money. Well, since accounting reports are not your little term paper, the whole thing turned against the author. Now there is a trial, during which everything is shaping up not in his favor.

As always, the “completely innocent” raised a howl to the heavens on the little internet. After all, it is well known that the internet is the abode of democracy and justice. The people will always put their chests forward in defense of Truth. The post itself is not interesting, but the links very much are.

Since when did a Wikipedia article start being called “expert opinion”? First, Wikipedia has never been and never will be a site that provides authoritative legal advice. And second, when I checked the edit statistics it turned out that all these “experts” are anonymous users, most likely one and the same person.

I can already sit here and see how, in the future, people will be posting links like this:

Expert opinion: http://linux.org.ru/Дело_Жукова

Or even like this:

Expert opinion: http://lurkmore.ru/Дело_Жукова

P. S. The article itself has been nominated for deletion. Most likely it will be deleted in the end.

Growth in the Use of C++ Synthesis

2009-09-25T18:46:00+03:00

Sean McCloud, a director at Mentor Graphics, reports that by the end of 2009 the number of companies using C++ synthesis for production will exceed 33%. In 2008, that figure was about 14%. The statistics were collected through an anonymous survey of 800 engineers. The survey was sponsored by Mentor Graphics.

The three reasons engineers choose C++ synthesis are: a) it’s easier to verify (68% of respondents); b) RTL is produced faster (64%); c) fewer people are needed, which is very relevant in a crisis.

And here’s some interesting statistics on the market size for high-level synthesis products and related services. These figures were collected by analyst Gary Smith and published in the July issue of “IEEE Design and Test of Computers.”

The chart is very interesting. Smith reports that in the ’90s users developed an interest in high-level synthesis, but the tools were far from being suitable for production. Because of that, trust in them was lost. In recent years, though, there has been a renewed flourishing of new-generation high-level synthesis programs. I know that many people are skeptical of C++ and SystemC as system description languages. We’ll see in a few years how it really turns out.

And acquaintances also tell me that all these numbers are marketing bullshit meant to lure in users, like, “look, half the market has already bought C++ synthesis, so come on, you buy it too”!

Sergey Mikhtonyuk’s Lecture on Computer Graphics

2009-05-25T16:10:00+03:00

On Wednesday (05/27/09), Sergey Mikhtonyuk will give a lecture on computer graphics. The program includes:

Graphic primitives
Transformations: Projection / World / View
Matrices
Z-Buffer & Stencil buffer
Lightning: Flat (Gouraud / Phong shading), Directional (point / spot lights)
Textures: Texture coordinates, Mip-mapping, Sampling, Cubic textures
Lighting models: Diffuse / Ambient / Specular
Shaders: Vertex / Pixel, Effects
Effect, postprocessing tutorial

Date: May 27, 2009 (Wednesday). Time: 11:30 AM–2:30 PM (three hours). Location: Room 318, KhNURE

Admission is free.

Lecture participants are expected to have a good knowledge of C++, the basics of computer graphics, and basic terms and principles. To participate, please send your full name, contact phone number, and group number in advance to volodymyr.obrizan@gmail.com.

Intel’s Parallel Programming Contest

2009-04-25T09:28:00+03:00

Since April, another Intel parallel programming contest has been under way. This year participants are invited to solve 12 problems in two stages. The first stage began on April 6 and will end on July 3, and the second stage will begin on August 17 and end on November 13.

Each solved problem is scored separately as follows:

0—100 points for a solved problem (correctness and execution speed are taken into account)
0—50 points for the explanatory note (up to 10 points for a description of the sequential algorithm, up to 30 points for a description of the parallel algorithm and the technologies used, up to 10 points for performance)
5 points for each forum post devoted to solving a specific problem (but no more than 25 points)

The participant who scores the most points for solving a single problem receives a certificate for $100. At the end of each stage, the grand prix for the highest total score is awarded: a netbook.

Of the problems open as of today, two are known: radix sort and solving Boolean equations. Most likely, in the future we can expect problems on graphs, arrays, and lists.