Sender | Message | Time |
---|---|---|
26 Sep 2023 | ||
17:10:07 | ||
jz: I would like to contribute on the development of Decred on chain side | 17:13:52 | |
Can you share which ticket would be good to start with? | 17:14:05 | |
There are 18 issue tickets open on the daemon repo https://github.com/decred/dcrd/issues | 17:14:54 | |
I would like to start contributing from small and would like to join this project full time | 17:15:21 | |
#dev:decred.org would be more appropriate place | 17:19:45 | |
Okay thanks for the reference ay-p | 17:20:17 | |
Forwarded from #dcr:decred.org: Decred Journal – August 2023 Decred’s governance was on full display this month as the project upgraded with two consensus changes. Cryptopower and Bison Relay saw new mainnet releases and DCRDEX got full node wallet support for Dash and Firo. https://www.cypherpunktimes.com/decred-journal-august-2023/ | 17:48:57 | |
New post in Decred Reddit: Decred Journal – August 2023 | 18:18:47 | |
27 Sep 2023 | ||
https://twitter.com/dreacot/status/1706825064216457691?t=Fhv5DkZDV7UJG6lT-auGEw&s=19 | 00:24:04 | |
* I'd be going live with phoenixgreen @phoenixgreen:matrix.org: and Exitus: this Sunday at 2PM UTC | 00:25:44 | |
* I'd be going live with phoenixgreen @phoenixgreen:matrix.org: and Exitus: this Sunday at 2PM UTC https://twitter.com/dreacot/status/1706825064216457691?t=Fhv5DkZDV7UJG6lT-auGEw&s=19 | 00:25:57 | |
* I'd be going live with phoenixgreen : and Exitus : this Sunday at 2PM UTC https://twitter.com/dreacot/status/1706825064216457691?t=Fhv5DkZDV7UJG6lT-auGEw&s=19 | 00:26:33 | |
* I'd be going live with phoenixgreen : and Exitus : this Sunday at 2PM UTC https://twitter.com/dreacot/status/1706825064216457691 | 00:27:35 | |
05:21:46 | ||
Really looking forward to this one, If anyone has any questions for the show, let me know and I'll get them added. | 06:37:00 | |
What risks and unknowns are in Go/Gio/gomobile builds, how Gio development (that we are betting on) and adoption has been doing (latest advancements), how Gio performs in areas where all new gui toolkits struggle vs old mature toolkits (keyboard navigation, scaling, OS integration, etc.), will we get on official F-Droid or our custom F-Droid, the state of reproducible builds (some work was put but it needs more afaik) | 15:06:18 | |
Yeah, would love to go deep into Gio itself and hear if that's an actively developed framework, and what kind of stuff it's adding/unlocking besides the fact that "it can all be written in Go". Also, it a dev-friendly framework to work with? | 15:15:47 | |
Everyone has their preferences. I don't like developing with Gio personally, mainly because I do not like immediate mode UIs. I find them extremely annoying to code for. | 15:21:14 | |
That isn't the fault of Gio though. It seems perfectly fine to work with if you're into that. | 15:22:07 | |
davecgh beyond the code complexity with immediate mode UIs, are there other drawbacks you’ve noticed? | 15:31:56 | |
I put a lot of emphasis on writing efficient and robust code that doesn't require a lot of revisiting because debugging later almost always ends up taking far more time in the long run. I also very much prefer to work with native widgets that work the way everything else in the OS does. For example, widgets that have all of the typical little things like ctrl+z undo in text widgets, being able to type the first few characters in dropdowns to skip directly to that part of the list, accelerator keys, keyboard navigation, accessibility support, etc. Immediate mode GUIs are the antithesis of those things. Then there are also a bunch of underlying things like font kerning and sub-pixel layout correction which immediate mode GUIs basically never handle right. I also find that once you get beyond the basic widgets, the complexity blows up very quickly and you basically end up implementing a bunch of things that boil down to retained mode facilities anyway. I do understand why some people like them too though. For example, games are a primary example of something that almost exclusively uses them. They give you a lot of flexibility to do non-standard things. They're generally pretty easy to get started with and typically have a gentler learning curve when it comes to avoiding a lot of the pitfalls endemic to retained mode like widgets that don't update because the code blocks the main thread, showing the wrong information due to not properly propagating updates to variables, etc. | 15:55:21 | |
I’ve noticed that too. | 16:03:01 | |
It does have its advantages and disadvantages. The code complexity is one major disadvantage. | 16:03:54 | |
Redacted or Malformed Event | 16:22:24 | |
Very useful thanks for sharing davecgh ! | 19:38:33 | |
Something to consider if the app hits a dead end with one of those issues. It would not be an end of the world for Cryptopower though. If the architectural layer separation is good enough switching to a retained mode GUI or even to a native widgets UI library would be a painful but not a full rewrite. I guess. | 19:42:16 | |
We should not get attached to a specific technology and rather keep investigating and reevaluating. | 19:42:22 | |
21:00:53 | ||
28 Sep 2023 | ||
New post in Decred Reddit: [Weekly] Random + Trader Talk | 12:03:36 |