What does software development have in common with board game development?

What does software development have in common with board game development? Sergey Milfgard Abdulmanov who studied in the specialty "mathematician-system programmer" and owned an IT-company knows this best of all, and is now known to Habr's readers in posts on the Mosigra blog .

He will complete the Joker conference with his keynote “How we expanded the bottleneck of development”, and on the eve of this conference we asked him a number of questions about what the two industries look like - and how they differ.

Sergey : In short, it would be more correct to announce the thesis “how badly we fucked up”. This is so that you do not think that we have invented a new Ajile or something else. And we have testers almost crammed developers face. In private. But this is a separate song.

- There is a well-known thesis from the book “The Mythical Man-Month”: “if you add more people to a software project that is lagging behind the schedule, it will only slow it down”. Does the same effect appear in the development of board games or not?

Sergey : It depends on what kind of people. There is a horizontal scaling, this is when one team labs one game for 5 months. You take and make 10 such teams, the speed grows 8 times. And there is vertical - this is when you break everything into microservices and expect that each section will go faster. We tried the one and the other, and the third. I think so - if you add people, you can grow in speed. But it must be professionals. The second question is where to get them or how to train them.

In general, my practice shows that it is cheaper to work with professionals. Even if they are more expensive.

- Often, programmers compared with other employees are considered to be very difficult to manage (“it's like grazing cats”). You have seen the management of both programmers and non-programmers - do you think this view is true?

Sergey : Listen, man, what did you get to the bottom of me? I am an introvert. I do not know, I am equally difficult with all. But at least I understand the developers.

But seriously - no, there are no such problems. There are others.

- Usually, development is said to be a rapidly changing industry, where you constantly need to retrain and adapt to the new. And the development of board games in comparison with this - a quiet backwater, or is there actually also all the boils and just have time to relearn?

Sergey : Our key skills remain longer, but our market itself is changing and developing. And the company, you know, not the most calm. There is a motto - “if you do not go forward, then you go on ...”, well, in general, you don’t go there.

The coolest thing an employee can do is make decisions for himself and take responsibility for them. And it's always scary. This is what we need to learn.

Here it is also necessary to take into account that we, with our 9 years of experience (and someone has two times less), are now trying to get out to Western markets. And it turns out. And there the competition is much stronger. That's where everything really boils. We are plush for them from the point of view of development - we know nothing in the tastes of their people, and they are for us from the retail point of view. In the end, somehow unusually well with the Dutch. They work there, judging by the games, in an altered state of consciousness - and they understand us very well.

- In software development, it often turns out this way: a successful programmer is assigned to the managing role, but in this role he turns out to be far less successful, and in the end no one is happy, but it is not clear who to replace him with. And in the case of board games, do such difficulties arise?

Sergey : This is you about Uncle Edison, right? We have another problem - the development watches hundreds of games a year from the authors. Therefore, it is forbidden to work on their projects. In general, our CIO recently became the head of retail. That it was really cool feint ears.

- You recommended on Habré the book “The mental hospital in the hands of patients”, which describes the stereotypes of thinking of programmers / designers and the difficulties that this causes to users. And to what extent is its content applicable to board games - do such difficulties arise from the differences between the “typical game creator” and the “typical game consumer”?

Sergey : Oh, to the fullest. The commercial success of the game consists of two things: the beauty of the art and the interface. The second is a purely engineering topic. How to optimize the thinking time, how to reduce the required memory capacity of the human subsystem, how enchanting to generate a random number in the desired range, what timer to use, where the symbol is needed on the card and what icon it is, how to formulate the rules more correctly - all this should be understood.

And the typical difficulties are slightly different. Do not know, for example, how the white box behaves on the shelves (gets dirty from touching and looks dirty almost immediately), do not know how to design the box so that the components do not wake up, write such rules that you cannot understand with half a liter. All this does not affect very strongly, but it does add up. And it shows in sales. There is no better metric.

- Development of software is always a combination of “what was originally planned” and “what turned out in the process,” and even with a clear initial TK, everything can noticeably go away. And how do board games relate to “input data” and “what happened?” - are there many input games in general, and how often in the process do you have to give up on your original plans?

Sergey : If this is a job of localizing a western license, then usually the result is the same as the plan. Although, for example, we made “ Trains ” (there will be a mega-twisted “ Express ” game, I'm really looking forward to serial copies in order to drag off a lot of presents at once) - this is a Japanese bestseller. They are just crazy on the trains. So, she has a very delicate balance, tied to the playing field. And the playing field is a map of the Japanese railways. But Russia needs it, because our player will not be very interested to play in Japan. We made visits for almost half a year in order to maintain the initial balance - and it worked out. Then another railroad showed, and just carried it out, that he only corrected some terms, and did not get to the bottom of anything else. I was not directly involved in this process, but I was interested. When in Australia in Pain Creek met the caretaker of the train station, there was something to talk about. He thought I was a Russian railwayman.

And if the development is pure, then it has never gone according to the original plan. In the introduction, usually just a market segment and the need for it. How to solve is already a matter for development. From the undecided - I really want to play the game on “ Elite ” for several years now, and as soon as it comes to just about what a commercial product should look like. Or even a dream - a game of This War of Mine . Crazy license, but potential opponents have done it, and have made it not very playable, alas. Many projects of previous years are lying, waiting in the wings - when we understand what and how to do with them. Experience accumulates gradually.

- Developers often talk about how important testing is and how disreputable it can cause disastrous results. But at the same time, the software has hotfixes and the ability to roll back to the previous version. And in your case, when you prepare a physical product, the price of the error “sent the wrong thing to the production” is much higher, or not? How does testing in the case of board games, and how it is meticulous?

Sergey : Circulation recall is a fat polar chanterelle. And it happens after all. We now counted the losses on the alterations within the process - even before the release, purely on beta, and realized that we were doing some kind of unproductive stuff. But this is a system-related thing, such a development framework. It is necessary to change everything, but we can not. Therefore, now another option is trying one process.

Well, it is right to compare with the business requirements in development. This is when you saw-saw software, and then during the play the business requirements of the internal customer change. Here rollback will not save. In a good sense of the word.

What is even sadder, testing gives far from everything. Here you can’t cover a unit, so many things are numerically immeasurable. It is necessary to release the circulation and see whether the project is successful or not. Much solves intuition and understanding of the market. Not always it can be formalized.

- Software stands out for “iteration”: while many other products are created “once and for all”, there are gradual changes, work does not end with 1.0, and conservatives in general only start to use something from version 1.1. Do games usually come out immediately in the final version, or do they also have a second draw different from the first?

Sergey : We have hotfixes - these are new versions of the rules or errats. Tournaments are usually made with separate rules with the tournament regulations, they differ greatly from the usual - for example, otherwise the situation of permissible draws is negotiated.

The circulation itself usually differs little, and you will not release 1.1 - 2.0 goes right away in six months or a year. Of course, we collect all opinions, we look at the reviews, sometimes we change something at the level of mechanics. Now the most frequent request - it is difficult to play color blind. We put special patterns on the cards, we make additional indications often. More from circulation to circulation, knowledge about production is growing and new lines are coming into operation - we can change components, for example, wood for glass - it turns out to be more beautiful and more environmentally friendly (although how to look) and cheaper.

- It is known that many IT people love board games. Do you observe any more specific correlations in this regard? For example, among developers, some genres are more popular, and among system administrators - others.

Sergey : The general correlate is a game where you can show the skill. That is usually deep tactics. A vivid example is the Hive . The second correlate is simple games like Djenga , to pobohat-socialize. They are taken to offices most often.
Although no, it's all nonsense. This is me about the generation that saw 80386 live.

- You wrote about the computer game Factorio, which is a “peculiar programming language”. And among the desktops, can you name one that is most relevant for you to programming?

Sergey : Of course, Neyroshima-6 , this is directly embodied debugging and ready language. Excellent tactical game of Mikhail Orach. We had regular tournaments for her, and children came in, watched how big bearded uncles played, quickly learned logic and won them. One such Ender just a couple of times in tournaments serious players vomited, but, alas, did not rise above the second place.

I recommend to Shimu to download first - there is a free version in the app stores, it helps to understand whether it is yours or not. And then play, if inserted. Post here .

It was a good game about programming " Roboralli ", very children for learning. Her Richard Garfield wrote, it’s almost like Lenin. But in Russia, it has long been gone, only if someone has stale.

There will be a miracle how good Trains are . There is generally a construction of the deck inside the game, such a constant recursion. They can still be obtained in English now, but very soon there will be Russians.