Radek Falhar

Přijďte na panelovou diskusi o roli Česka v Evropě. Máme usilovat o hlubší spolupráci? Nebo se více soustředit na sebe? Máme aktivně spoluutvářet evropský projekt, třeba i směrem k federalizaci? Nebo raději zachovat současný stav? Join for a panel discussion to discuss what role Czechia could play in Europe - whether we should seek more cooperation or rather concentrate on ourselves, whether we should strive for a more active role of shaping the European project maybe even towards federalisation or whether it is better to maintain the status quo.

When a group of people need to choose something, how do they decide what to pick? The differences between these voting methods may surprise you! youtu.be/JDCQQEeSQlo?si…

If you wrote the code, you cannot update the tracking ticket. We need separation of duties for compliance reasons. #NotMyAgile

How "optional" are development practices? Can I work in a sub-optimal way that creates trouble for the company, the team, and the customer if I feel it "works for me?" Can I do that even though better ways are known? Where do you see the boundary? What is the litmus?

Oh, we’re absolutely doing this agile. We just have a waterfall plan is all. #NotMyAgile

Inside-the-box Brainstorming

Software projects are clearly difficult things to do well, but large government software projects have such terribly high failure rates that statistically they are doomed to failure before they even begin. This one wasted $5m... WHY?! (Link to the video in my bio)

@tottinge The treatment of the actual development as if it were a black box, by so-called "agilists" is maddening. For every one social media post about how to write a good test (for example), there are 999 posts about how to have a fun sailboat format retrospective.

Sometimes people who don't understand software development AT ALL think that a development request is (or should be) just like placing an online order. "Adding new capabilities into a complex working system without errors or side-effects" vs "buying a shoe horn"

I often see organizations that have formed a team, but in reality what they call a team is just a group of individual workers who work in full isolation.

@thecaptain_nemo Could this be survivor bias? We only notice predictions that were correct. But there could be hundreds, or even thousands, predictions that turned out to be wrong, and were forgotten by history.

fucking ridiculous that Schrödinger was able to infer the existence of a DNA-like “aperiodic crystal” acting as a “code script” for life “built up of a very small number of different kinds of atoms” before we had almost any empirical evidence to directly support that

@BatsouElef Yes. And your question seems to imply "human intelligence" is somehow good. It isn't. Human intelligence is severely limited and imperfect. "Human intelligence" is a low bar to set for ability for problem solving and learning.

Will AI ever surpass human intelligence? Opinions 👇

@ryanels I really hope you have forgotten the /s . Some people might take you seriously.

That's why I test in production 😂🤭

@Hasen_Judi @VasiliyZukanov Could you please not group TDD with Microservices or Scrum? TDD is actually good.

(1) I don't think I can count all their names, but it's pretty obvious. It's the group of consultants who came up with all kinds of object oriented design patterns and architectural ideas and agile. (2) Everywhere you go looking for a job, their ideas are being employed, whether it's TDD, DDD, Microservice, Scrum, Clean Architecture (3) It's just a matter of fact: these ideas are used everywhere, and everywhere suffers from the problems mentioned. But it's more: you can actually perform rational analysis and see all the connections from the ideas to the bad outcomes.

The Robert Martin gang has basically had phenomenal success in marketing their ideas. Now every where you go, the ideas that came out of that gang is what drives the entire industry. Let's check out the outcome: - Software takes more resources than ever to develop (in terms of time, money, manpower). - More bugs that are so difficult to fix they go unaddressed for years. - Worse performance than ever despite incredible advances in the underlying hardware What would it take for the industry to throw these ideas behind?

PRs are fundamentally flawed. They do not represent the way someone gradually develops something in the order that matters. You just see a huge list of diffs in some random order. Even if you go commit by commit, most haven't cared enough to break the commits and organize them in a sensible way. I often feel PRs are someone's vomit thrown on you, and you trying to figure out what they had eaten.

This is an oddly common claim. There’s no reason devs can’t effectively QA each other’s work. Devs aren’t inherently “bad at QA”. Devs are just humans. They can do anything any other human can do.

@alexbunardzic I dont understand how batch size makes sense int this context. I feel that better words could be used to improve clarity?

My experience is that the major obstacle in practicing TDD is our innate propensity to overbid on work and go for the big batches. The only way to do TDD properly is to learn how to work in tiny batches.

It's not about how frequently we commit, it's about how frequently we run the code.

@ryanels I believe you have missed /s

That's what makes JavaScript great 😃