The culture of reservations does not go hand in hand with the constant delays
This article should have been released promptly, to keep up with the hot topics of the frivolous world of the internet but has been postponed. Basically, perfect for the topic these days.
It is now known how many products expected for the first half of the year have slipped, of little, or occasionally as appropriate. Some of the great games of these months to come are in turn products of 2019 capitulated in 2020. I am no longer able to move the slightest complaint about these behaviors.
Here my work background enters arrogantly, inserted in a context of industrial production of products, including the composition of software. It's not quite the same as producing a video game because I'm dealing with finished physical products in a different business model, but the workflow is much more similar than you might think. So I will use this space to investigate why delays occur. Because until something is produced, even just for passion, together with other people, it is not often possible to understand the process dynamics.
Let's go fast, immediately removing tactical-commercial references. The marketing team, committed to estimating sales projections based on the release moments and the unknown moves of the competitors, sets certain exit periods for the products. Sometimes, these turn out to be wrong: they become too crowded, the closing / opening of the fiscal year is upon us and the convenience or otherwise of having collections can vary. These cross-references are either short-term or fairly telegraphed in terms of the relocation period. The economic part also affects referrals for technical reasons. If a game needs 2 more months to complete and this would bring the launch window in the middle of summer, much better to wait for September.
Time is money reads a famous maxim. The exact calculation and considerations that lead to the decision of how to manage a videogame project vary from entity to entity, but the logic is always the following. I can afford a € 100.000 budget: if a developer costs € 1000 a month, I can get 100 to work for 1 month. Or 20 for 5 months. If the job is not completed within this period, I overshoot the budget. Thinking of operating without time and budget limits is crazy. There is a limit to how much you can cash in on a given product: Rockstar can afford extremely extensive development cycles because their games sell a lot and over time. All those who live in the launch period cannot pull it long.
Aim high, without exaggerating
And then imagine doing everything without time limits or with extremely extended times. Nothing would work anymore. Of course it is bad when they tell you from the upper floors: "time out, you go out, in any state, we have no more money to keep you standing". In this regard, to organize the work of many people, there are numerous IT tools and specialists who must keep track of the progress of the project. The video below, from a beautiful cross-section of one of the many ways to organize workflows.
Oh yes, the video is from Star Citizen. The game that "will never come out" according to many. In constant beta, with a thousand promises, many still to be respected. With a freely accessible roadmap. Why then, despite all these tools, despite the watchful eye are there still referrals? Why do you keep making mistakes? The development of a video game is a very complicated process, especially in the AAA industry, where a very large number of people with many different skills must collaborate to arrive at a finished product. An element that perhaps eludes us, it's like it's a unique process, not standard in most cases. This introduces a significant number of unknowns, which collides enormously with the human factor, which is difficult to control.
The first critical point is identify what you want to do. In English we use the term "scope", we would be more appropriate to use the size and ambition of the project. It happens most of the time to aim high. For wanting to do something new, different, to elevate the franchise. Something that does not exist before, that must be created from 0. Something beyond the reach of the study, which therefore must grow, study, train before being able to get where you want. And entire production phases cannot start unless the base is first built. At other times the purpose of the project changes constantly and a lot of work is done for nothing. Star Citizen is somewhat the symbol of this problem.
The continuous monetary foraging has led to an expansion of the objectives, changing numerous times during construction. The ambitions are so high, that entire technologies have been developed to prove inadequate and have to be redone from 0. It is not only a problem of SC, it exists in every production chain: the problem is when you are wrong, not like. Making a lot of mistakes at the beginning of the production chain costs little, in terms of time and workforce. Making a mistake towards the end can mean having to do it all over again. Getting almost to the bottom and realizing that that level really sucks, that the story doesn't work, that that beautiful idea on the gameplay card once made is terrible leads to enormous imbalances in the rhythm of the work. For this reason, the more refined the iterative part of the initial test is, the less problems there will be.
In a process that leads to a physical good, errors are weighed more because each iteration must be made physically. New revisions of printed circuits, new molds to have new geometries for plastic and metal. The advent of CAD systems and 3D printing has also allowed these supply chains to make mistakes ever more quickly, but it is nothing compared to the highly iterative nature of video games. The complexity of one's game is felt even when "everything went well".
In a very obvious way, the more lines of code and the interactions between them, the more likely there are bugs. And maybe we don't realize the numbers at stake. Even when they are older, QA teams cannot compete with the release of the game to the public: a team of 10 testers, even if they played 24 hours a day for 1 month, would only accumulate 7200 hours of play. One million copies sold at launch, are millions of hours of play, which by the law of large numbers, are able to find more than 7200 hours only bugs. I would say that Cyberpunk 2077 is the case. They defined their goal in the right way, they showed it to us. Except that to clean it to an acceptable level, it still takes months. And here are some publishers who prefer to put it off than to put out a pitiful game. I would finally say.
This is what happens on a technical, technological or concept level. Bugs, need to develop new technologies or have shot too high. But then there are all the mistakes and difficulties in the process. These involve the human sphere, as on average we are very bad organizers and there are numerous psychological aspects that go all out to blow up a project.
Those who work are still human
The first of all is perhaps what brings us to overestimate our abilities and to always consider the best possible boundary conditions. This is mainly because we tend to shift the causes of our failures and delays to external elements rather than to ourselves. It is a very difficult mechanism to avoid. This applies both in the definition of the scope of the project as discussed above, and in ordinary works. "How long does it take you to make the models of the three characters?" - "Ah, but go easy, easy, it takes me 5 days!" And instead 2 weeks later, they are still not ready. Why? "Eh, stomach ache, coffee breaks, my foot hurts, my desk neighbor bothers me, those idiots in the IT department have messed with the database, the tools I have are not adequate" ...
Even if you manage to finish your duties earlier, however, another mechanism comes into play, known as the Parkinson's law. Its statement is as follows: "The work expands to occupy all the time available; the longer the time, the more the work seems important and demanding". Basically, if the deadline for a job is one week, the job will be completed when time runs out. This takes three different forms.
In the first one, my work flow slows down until it is optimally spread to complete the request in the estimated time. Another possibility is to procrastinate until the last useful minute and take exactly one minute. Last option instead sees the work being completed first, and the extra time is used to embellish it, to add outside the request. Also because going to tell your department manager that you have finished earlier, could have the consequences of putting you in too much light and therefore expect more. Better not to be loaded with more work.
And in the midst of this speech is the crunch. Spending an extremely high number of hours in the office to finish a job. Which is a more complex behavior than how you often paint it: let's face it, if your bosses force you to do huge overtime to meet deadlines, and if they are not respected you risk dismissal, they don't deserve to be bosses. If there are no protections due to lack of unions, things need to be adjusted. But it must be said that many of us suffer from student syndrome or are extremely tied to their work / company.
When there is a known time pressure, we push to the maximum. Study just before the cannon exam for example. Write your thesis on time, doing sleepless nights and weekends indoors. This behavior does not change when you enter the world of work, with final last minute runs or overtime days and days because, in the first place, you are the one who wants to close the matter on time and well. Nobody orders you, you don't have any external pressure, it all starts from your own person.
This is why I honestly don't think it's something that can be fought on an absolute level. Simply letting people free to work as much as they want, helping if these hours become too many and harmful, that is when they become harmful instead of useful, pay everyone adequately for the effort and never give external pressure from the upper floors. I don't see a real solution that is not repressive even of one's will. After all, it would be enough to really implement laws such as our collective agreements. Metalworkers, for example, leave 8 hours of overtime weekly, up to 200 hours per year. Which is equivalent to working 13 months out of 12 staff. If it were respected even for those who work in the gaming sector, especially abroad, it would not be so disastrous.
At this point I would say that we have seen, broadly what are some of the mechanisms involved in product referrals. Unfortunately, as the industry becomes more complex, these problems only intensify. The moving parts are more and more and knowing how to combine them without exceeding becomes an increasingly difficult job. At every postponement I do not think of myself that I will have to wait more for something I want with all of myself, I think of all the people who are working hard to bring their personal masterpiece to light and who knows how many hitches they met along the way. It is part of the nature of the industry and in my opinion altering it would change its soul too much. It would mean being satisfied with the homework and not wanting to aim higher.