Veli Tasalı @tasali@mastodon.technology

I made a shitty website that generates indie band names: https://indiebands.velitasali.com/

Cursed kotlin code

Unlike Flutter, one would expect Compose would draw native views, but it doesn't even use the native view system, which is odd. That explains why it took them so long to come up with. Still, I am happy that stateful and stateless widgets have become a thing. In the old design, it was the developer's job to explain to views what changed. Now, what contributes to the drawing of views is treated as "State"s. DiffUtil.ItemCallback and other similar classes were probably the result of the old design.

This is my favorite font btw.

My Cocteau Twins scrobbles are disappearing from last.fm and I am not okay with that.

One thing most developers don't understand about IO streams is flushing being an option rather than a necessity. If a write operation necessitated a flush op every time the former was invoked, the existence of the latter would be pointless. What should actually happen is invoking flush only to ensure the data is written to the target. Mindlessly invoking it will kill the performance and the optimization made for the target.

Ah, yes, good ol' private fields exposed by public fields!

It is time to move to Jetpack Compose baby and API 21, which also means getting rid of data binding and XML resources. Btw, I am feeling cute, might use spaces instead of tabs.

This demo app is created using Kotlin Native bindings for GTK. I am glad somebody already has started working on it. I don't know much about the ongoing limitations of Kotlin Native, as it is mostly in experimental stage, but I think this could be the best thing that has ever happened to Desktop Linux since Kotlin is a very popular language with consistent set of APIs (unlike JS).

Ah, yes, I can finally connect my graphics card over a wireless network!

I have been kinda working on an Android TV app. I didn't have to write a lot of code but the UI stuff. Also moved the business logic into a separate module which also reduced the build time. I am going to release the app when "it just works". I am not planning to do any crazy stuff since it is an Android TV app. Lastly, I still think it was a good decision to create uprotocol. It is somehow faster than SFTP or SMB combined and reaches the speed iperf reports.

If I recall correctly, FOSS communities never needed the help of Microsoft. In fact, Microsoft was one of the few causing harm with its patents and slowing their development. Microsoft might have come out open-source friendly recently, but I don't believe its stance has changed. The only thing new is that it now knows is it can make money out of FOSS if it doesn't hate it publicly. I wonder what projects/organizations other than Linux and Ubuntu (or canonical in general) it has supported so far.

I am testing a different approach to my usual design. The work in the screenshot isn't finished, but it shows where I am going with it. The purple and cyan colors are Android defaults. Purple will probably stay, but I will replace cyan with a dark pinkish color.

My biggest issue is the backwards compatibility. I have already spent a lot of time keeping the app Android 4 compatible, so I don't want to increase the min SDK yet.

Also, I haven't made my mind about the font. Rubik may be a fit.

You may not like a FOSS project, but shitting on it is something next level. What now, you want a refund? It gets dumber when the development of the said project doesn't affect another project (like Linux Kernel does).

"It shouldn't exist."? There you have it, it no longer does!

I just removed the Weblate page and hid the repos, so TrebleShot project is officially dead now.

Let's celebrate it with a relatable song: https://youtu.be/l3G_Y3aIlis

2021 has been disappointing so far. Maybe a good alt-J album can change things?

https://www.youtube.com/watch?v=gFKhxi8b9SQ

For my next project, I am going to use Flutter.

The only thing missing is a good guide on MVVM in Flutter. So, I thought I might create one.

I haven't started yet and I don't even know how it is going to turn out.

I guess if you have a good product, say an album, but no budget for exposure, what you will probably get is a cult following.

Similarly, if you have a lesser good product with a budget for marketing, you can get a large consumerist crowd that discovered you because others were talking about you and they were too lazy to look for a better and cheaper alternative?

I don't know. I am just sad hearing some good-ass albums from 2010~ with almost no listeners compared to so-called good albums.