off-topic 2023-03-07 | Slack Archive

Thomas Moerman10:03:44

Does anyone know or use a naming convention (involving ascii art and\or ligatures) to represent an event? Just for having a nice visual cue.

Thomas Moerman10:03:23

something resembling an envelope 📧 perhaps? 😅

p-himik10:03:35

> ascii art

$ cowsay event
 _______
< event >
 -------
        \   ^__^
         \  (oo)\_______
            (__)\       )\/\
                ||----w |
                ||     ||

:D Jokes aside, I'd say it depends on the context. The only context-independent thing that immediately makes me think "event" is the word "event" itself.

Thomas Moerman11:03:52

i kinda like this

Thomas Moerman11:03:33

but yeah, it might be a case of obsession with details on my part 😉

p-himik11:03:17

Oh, if it's for code itself I myself would settle on -> or something like that. With re-frame, instead of #(dispatch [:id ...]) I use #(>evt [:id ...]). (It's definitely a personal preference but I absolutely abhor ligatures.)

Ben Sless12:03:19

!event?

👍 2

Nate13:03:46

local-main=> (def |v| "##--@@-->>  Message for you sir!")
#'local-main/|v|

Nate13:03:02

(p.s. just because I wrote it, doesn’t mean I don’t hate it)

respatialized14:03:22

https://github.com/josephg/schemaboi Interesting concepts here, seems to overlap a fair bit with the motivating ideas for EDN & themes in Rich Hickey’s “Spec-ulation” talk, but embeds the schema in the serialized representation of the data for self-description.

phronmophobic22:03:31

> The big new idea in schemaboi is that the schema for any data set is included inline with the packed binary data. Is this different from avro?

respatialized14:03:47

I think the interesting point of difference is potentially in the compositional way schemas are defined: “given two data models we can define the union (merger) of those data models.” I’m not familiar enough with Avro to know whether this is possible in Avro, but making schemas compositional in this way is an interesting idea

mauricio.szabo21:03:27

So... I've been using CirrusCI for the last 6 or more months on the Pulsar project. I have to say - I'm not impressed. In fact, if it wasn't for the almost unlimited free-tier for Mac and Windows builds, I would probably not use it... it's shocking how much obvious bugs appear on their UI, some groundbreaking. Like, for example, in this very moment artifacts can't be downloaded by their UI...

borkdude22:03:45

I always upload artifacts to Github releases via a script since I couldn't programmatically download them. UI-wise I'm not impressed, but at least they offer mac aarch64 unlike others which is what I'm using them for. And it's way faster than Github Actions. The UI of Github Actions also sucks since often you can't even see what long running job is doing.

mauricio.szabo22:03:01

Yes, I agree with all of these, but comparing Github Actions with anything is not a good idea with me, because I feel it's the second worst CI I ever used (the first being Gitlab's) We actually interface with the Cirrus API to download artifacts directly from their site. The issue is the "alpha" builds - builds that we make to test features, these are not uploaded anywhere (and they shouldn't be)...

2023-03-07

Channels