HN user

ValleZ

49 karma
Posts1
Comments69
View on HN

If you split all the changes for a feature this way not only you hide the way all changes interact with each other but also make the development at least 10x longer because an average approval time is often more than a day.

Actually after reading it more carefully I probably see why it didn't work for me, but the notes in the page are bizarre:

* Avoid special characters such as * in M*A*S*H, use MASH instead.

Since when a common ASCII character is a special one? What about more common unicode characters I use?

* Do not abbreviate the Season folder with S01 or SE01 or alike.

I.e. if I put anything not in the folder named "Season XX" it won't work? Ugh... really?

* Season folders shouldn't contain the series name, otherwise Jellyfin can in certain cases (Stargate SG-1 due to the dash and one, for instance) misdetect your episodes and put them all under the same season.

Well, how about to fix it?

* Episode numbering for specials may vary from metadata provider to metadata provider.

Very helpful, so the "Series XX" required above won't always work.

And even if everything above fails why not to sort by name? It should not be hard for any engineer, right?

I have the series in their own folders. I tried to do a more nested structure to no avail. After a day of attempts to fix it I switched to Plex and it despite having its own quirks just worked fine.

Awhile ago. XEP-0286 (mobile connection improvements) dated 2018-01-25. File sharing has many different XEPs and I have no idea which works, but I know that early XEPs usually didn't work because of NAT. XEP-0384 (encryption) dated 2022-01-18.

XMPP was not a mobile friendly protocol for a while, had efficiently no security out of the box and lacked simplest things like image or other file transfers.

It's closed source for now, but it's based on llama.cpp and I'm planning to improve the included android example in the llama.cpp

One way to run LLM on android phone is to run ggerganov/llama.cpp in console like termux and it works quite well but is not super convenient. I was working on the app to make running on-device LLMs more user friendly, please try it out!