Через пятнадцать минут стартует мой короткий доклад про роботов на конференции ЦИПР, в прямом эфире: https://youtu.be/YotpvnE3ma8 (запись тоже будет позже, но это не точно)

"И снова здравствуйте!" (2022) это наш ответ Тарантине. Достойный ответ. Сериал динамичный, жесткий, одновременно добрый и романтичный, с прекрасной игрой прекрасных актеров. Кино о смысле жизни. Хочется пересмотреть еще раз.

New post on my blog: Reflection Means Hidden Coupling. It argues against reflection in Java, as well as all other languages.

На прошлой неделе на конференции ЦИПР в Нижнем Новгороде я студентам рассказывал о роботах в software engineering, видео уже готово: https://youtu.be/LG9CQrqzG2o

Еще один короткий доклад на конференции ЦИПР и тоже про роботов для программистов. В этот раз доклад был для бизнес аудитории, не для студентов. Видео здесь: https://www.youtube.com/watch?v=KCLECWhB1-w

M197: Nothing hurts best performers as much as the lack of honesty from the management about the work of low performers. If you want your best players to be motivated, invest all your energy into building a system of measuring performance correctly, without a fear to offend or discourage anyone. Watch it.

This is the result of the previous Twitter poll, which confirms that if a bug report is hard to understand, its author is guilty. Here is a new Sunday Twitter poll, please vote. Its results will illustrate the blog post I recently published about reflection.

A new post on my blog: The Principle of One. It may sound like a too simple/short content for a blog post, but I insist. This principle deserves a blog post :) By the way, the movie is also very good.

M198: If you are an expert in software development, you most definitely contribute to open source projects, simply because you see many issues in them and know how to fix them. If you don't, you are not an expert. Watch it.

We started to form the Program Committee for ICCQ 2023, our third computer science conference dedicated to code quality, program analysis, software maintenance, and program verification. The previous two events were supported by IEEE Computer Society. The proceedings were published by IEEE Xplore. This event will be hosted by St. Petersburg University. The Chair of the Program Committee this year is Prof. Andrey Terekhov. If you want to join our PC or you know someone who might be interested, don't hesitate to email: [email protected].

M199: Coding without unit tests similar to building a house without a safety net: you can do this, but your productivity will be extremely low. You will mostly be driven by fear. Can you afford this? Watch it.

N16: новый юбилейный выпуск черно-белых новостей из мира айти на моем YouTube канале: смотрите.

A few days ago I sent out a newsletter, which is supposed to be monthly, but happened to be annual. The previous one I sent exactly 12 months ago. Some of you, I believe, received it. There are 4100+ subscribers in the list. From now on I will try to sent it monthly. You are welcome to subscribe, just fill out this web form. You can unsubscribe with a single click.

We finished forming the Program Committee of ICCQ'2023 conference, which we organize for the third time. The committee includes 25 professors from 14 countries: Russia, USA, Spain, Chile, Italy, Czech Republic, Israel, Germany, Japan, Austria, Belgium, France, South Korea, and The Netherlands. These people will review the papers submitted to the conference and will decide which of them deserve to be accepted and published (usually we reject 70%+ thus keeping the quality bar high).

At the moment we are trying to invite an interesting and famous keynote speaker (stay tuned, I will announce soon). Also, we are forming a Steering Committee. This should be finished soon and then we'll start promoting the event in computer science research community.

If your research is somehow relevant to the problem of code quality, you are more than welcome to submit it to ICCQ. If you are interested in keeping an eye on ICCQ, follow us on Twitter and subscribe to our YouTube channel.

New post on my blog: Automated Tests Are the Safety Net that Saves You (it received some positive response on Reddit)

A few open source Java projects (EO, Artipie, Polystat) are looking for freelancers. It's fully remote work in pay-per-result mode. You will work with GitHub issues assigned personally to you and will get paid when issues are closed. It's up to you how much time to spend on us: one hour a week or twelve hours per day. You will be paid for the amount of results you deliver. No meetings, no reports, no status updates! Just coding. Text me if interested and get three things ready: your resume, your GitHub account, your Stackoverflow account (if they are not impressive enough, better don't bother).

New post on my blog: Command Line Default Options in Linearized Plain Text

Очень понравился сериал Экпроприатор (2017) от Владимира Краснопольского. На удивление реалистичное и глубокое кино о том, что такое хорошо, и что такое плохо. Слегка театральная режиссура диалогов поначалу создает негативное впечатление, но посмотрите дальше — привыкнете и понравится.

For the third time IEEE Computer Society agreed to support ICCQ conference! ICCQ'23 is our third international conference about code quality organized and hosted in Russia. What does IEEE support means for us and for the authors of the papers that we accept? First and foremost, the papers will be published by IEEE Xplore and automatically indexed by Scopus, Web of Science, Google Scholar, and others.

If you are a computer scientist and your research focus is on programming languages, program analysis, code maintenance, or defect detection — submit your paper to ICCQ'23. The deadline is the 18th of December.

A new post on my blog: The Code and Its Tests in Different Pull Requests The idea is simple: always submit changes to the code separately from the changes to its unit tests. Simply put, in one pull request you modify the tests, maybe marking some of them as “disabled.” You merge this pull request and then make the second one, modifying the code and most probably removing the “disabled” markers from the tests. You don’t touch the body of the tests in the second pull request.