Yesterday I attended the first external Android meetup from colleagues in Kazakhstan telecom. For the first time it turned out quite large-scale, the guys have a nice office in the city center, there's a place to hold such things, many people came, accordingly we have something to envy. What kind of fashion has started to make meetups on a weekday, I don't quite understand this phenomenon.
As for the talks, the topics are quite beaten for a sophisticated viewer. Design system on Compose - I caught flashbacks from my own talk at the spring BeeTech and the past Podlodka. All the same basic concepts of design systems and all the same attempt to automate Figma, optimistic ideas and my, already somewhat pessimistic, view on them. TLDR: Design system is good, generating code for tokens is also good.
The talk about the Kotea library - here are the vibes of any UDF-architectural talk, during which the prevailing thought is that we don't seem to need it, we already live quite well without a dozen new entities with their own names. The talk was very memorable for its design, and now I seem to pay attention to this even more than to the content itself. TLDR: Shitcode is bad, architecture is good, it can save from shitcode, but it's not certain.
Unconditional like for the pizza break in the middle. But I would stretch the duration, this is where deals are made.
Well, and the talk about testing, in particular about screenshot testing, concluded the evening. It was also about Compose, telling about its advantages in '24 already seems like bad manners, and in two talks in one evening. I have probably even more skepticism about screenshot tests than about any others, but the use case with covering design system components is quite understandable. From trying it, honestly, I'm stopped by the unresolved question in my head of where to store the samples, because not in the repo (you won't convince me). The question is solvable, of course, but somehow it seems lazy to invest time in this communication with devops for the sake of what doesn't really shoot yet anyway. Well, and in the talk there was a bit of a lack of going beyond Google tools, there's usually not much trust in them. TLDR: Writing tests is good.
As you understood, the general message of all talks was read as "do it properly, it will be proper" without any revelations. But I'm glad I went, background communication with colleagues from other companies about minor things always leaves some useful marks in the head. And I would like there to be a little more time for networking at such events, otherwise from bell to bell talks, hello everyone, goodbye everyone, thanks for coming, and you don't have time to chat. It's generally quite cool to watch how the Almaty community is spinning up, a bunch of conferences, meetups, and familiar faces already, and there's literally more and more of this in front of our eyes.
PS. There are already a hundred of you here, greetings to all the newbies, my respect to the oldies