Infinite State Machine

I thought quite a bit about this extension of Conway's Law, which is taking the flexibility of the system under change into account. Here are some unsorted and maybe random thoughts from my experience of doing an inverse Conway maneuver on a larger scale at Flix. #SoftwareDesign

We'll deliver the finest #housemusic in #tokyo tonight! Entry is free - so if you are looking for a nice #tokyonightlife evening, come around! #tokyoclubbing #tokyohousemusic #tokyoparty #tokyomusic

@fduran @allenholub It's the curse of knowledge - as people who are very affiliated to the inner (and outer) workings of software, it's easy to say that these safeguards are unnecessary. But for the general user - holy moly NO! The amount of times I heard from users "but I thought that ..."!

@allenholub For destructive, irreversible actions I do want an “are you sure?”; I’d rather click once more and have this mild annoyance than losing data or hours of work.

A periodic reminder that a "do you really want to do that" dialog is an abomination. Yes, I want to do that, or I wouldn't have asked the first time! Also, by the time that dialog pops up, I'm already annoyed because it's in the way of me getting back to whatever I was doing before. I just want to get rid of the damn dialog. However, the default action is usually "don't do it," which I hit too often, so I have to start over. Forcing me to actually read and choose carefully just makes me more annoyed. In general, it's bad UX design to annoy your users. So, just do what I ask, but provide me with an undo. Yes, that's harder to program (watch me bring out the world's tiniest violin). Your (the programmer's) convenience doesn't matter to me (the user). Do what I need, not what's easier for you. Lazy thinking that prioritizes ease of implementation is behind a lot of truly repugnant UX.

@runarorama Isn't that inherent to TDD, where you don't have an idea of how the API will look like, and its up to you to design it? Then in the test you start writing the code as you want it, and then make the test pass?

There's a way of programming that I've been doing for decades and I realized recently it doesn't come naturally to many folks, so here's a pro tip: Write to the interface you'd like to have. That is, write the code you want, even if it doesn't work, then make it work.

@MarcJBrooker It's disturbing to see how many people think that Microservices were conceived as a solution to technical problems, and not organisational scaling issues.

I know this kind of junk is engagement bait, but it's sad that new systems engineers see 100 ignorant ill-informed takes for every bit of good architecture advice. Microservices are fine. Take your systems advice from people who build and run systems. Ignore influencers.

@VaughnVernon @ICooper If persistence was transparent, it would, IMO, not be the Active Record pattern anymore. The benefits of Active Record are colocation of domain logic and data storage/retrieval, making them easier to understand and simpler to implement for those developing these sytems.

@ICooper What if persistence was transparent; just happens? Why are transactional boundaries a problem? Is the desire to persist multiple records in one transaction?

Do you like the Active Record pattern? If not, please explain in reply.

looking forward to getting back to this book and finishing it up maybe in August :)

It was a pleasure to meet all you brilliant people at #DDDEU! ♥️ This year I decided that hands-on sessions will be my focus. The talks will be (eventually) released, but the hands-on sessions are unique in so many ways. To me, they are the highlight of the conference! Good bye!

@mathiasverraes I will need to wait for the next training budget cycle 😩, otherwise I'd have challenged the number 1 spot for fastest ticket sale in the history of #DDDEU!

We announced DDD Europe 2025 before Eric's closing keynote, and we already sold a ticket before the keynote was over! I guess that's feedback 😍

@einarwh Is being squared better or worse than being cubed? 🤔

On the train Berlin ->was Amsterdam for #DDDEU - choot choot! Really looking forward to meet everyone there. This conference is always a highlight for me, and this year my lovely wife is accompanying me, so it will be double the fun. Be there, or be squared!

@einarwh Sad to not see you there. Have a safe trip to Budapest. I try to note down as much as possible and release my Miro board afterwards, so you might be able to trick your brain into thinking that you actually had been there in the first place! 😁

My excitement for #DDDEU '24 is growing day by day. Are you going and have not chosen your hands-on lab yet? If so, then do it quickly (check your email inbox for your private link), because they are one of the best parts of the conference! And some of them are full very quickly.

We're looking for a technical+organisational coach in #Belgium, with DDD & architecture skills and a Continuous Delivery mindset, to help our clients deliver incremental value faster. contact@aardling.eu for more info.

@mathiasverraes This step would not be necessary in an Ensemble environment. mob.sh e.g. disables pipeline runs by default via "skip-ci" commit messages. But this only works if there is parity between the local & CI server test environments (which IMO should be the goal).

Commit and push the test. Yes, this will break the build. I think this is perfectly fine: the code was already broken, it was just invisible.

Here's roughly how I fixed bugs early on in my career: 1/ Browse around in the code. 2/ Try stuff. 3/ See if it works. Here's my preferred way of doing it since about 2012 🧵

@Grady_Booch Having ISO standards and the reality within organizations are two very different pair of shoes. Who actually has read the complete standard nowadays? And no, I am not joking, I know of no one in my org at least.

@Grady_Booch The stakeholder might be the wrong person to provide insights into that usefulness. The only reliable source of information are the users/customers of those software systems, that probably span many business capabilities to make the organization realize it's value proposition.

@CFDevelop @luke_pighetti I'd even go as far as saying iteration based agile is dead. Long live the flow.

@luke_pighetti That's how the industry was 10+ years ago. Scrum has ruined that.

In all seriousness, are there any jobs left where you just build the software in peace?

@vlad_mihalcea I believe that is partly due to the underlying JVM itself gives a higher flexibility in terms of supported technologies. You could run Java, Scala, Kotlin, and more (even within the same application) side-by-side. And more options means less risks.

When I started learning Java in 2002, devs were saying that Java was slow and could never compete with C or C++. 20 years later, many High-Frequency Trading companies use Java instead of C/C++. efinancialcareers.co.uk/news/2020/11/l…

@quii @VaughnVernon I like the "Functional core, Imperative shell" principle, which combines the best of both worlds. This often leads to leaner systems. Interfaces & modules describe operations, and the implementations benefit from the FP paradigm (testability, guarantees, compos ability, ...).

@VaughnVernon Sadly, the latter is what a lot of FP looks like in the wild. Obviously not setting stuff, but it seems a lot of FP advocates will excuse this style because it's "data oriented" and it's usually immutable.

The world would be a much better place if every programmer used the Tell-Don't-Ask principle. That means to do this: quote.extendExpirationBy(days); Not this: LocalDateTime dt = quote.getExpirationDateTime(); dt = dt.plusDays(daysToAdd); quote.setExpirationDateTime(dt);

@VaughnVernon And even knowledge and experience are only helpful in the way that they teach you to ask the right questions to uncover as much hidden information as possible. And also to understand when enough information is gathered to run the first code experiments.

What's a "best practice"? It's a procedure known to yield the best results for a given situation and desired outcome. There are no general best practices because no practice is always best. E.g. Best practices don't exist for unknown unknowns. Knowledge and experience required.