As a Liquid Glass hater, I hope this will help them roll back some of the terrible garbage that should never have been greenlit in the first place.
That said: He also oversaw design languages that I liked (everything before and perhaps including Big Sur), so I guess it’s not like he messes up everything he touches.
LLM providers know a lot about their customers. I expect that, with their post-privacy approach, they will not have qualms about scanning a customers’ future chats for mentions of the advertised product (oh, the analysis potential!).
Of course you’re deep enough in your domain that to you “editing” means “video editing” and “studio” means “video editing workspace”. But when I clicked the link, I expected a fully featured general application for power use of LLMs.
Is this be the long awaited LibreOffice modernization?
I think it’s remarkable how almost every popular application makes the step of being a web-app at some point. It makes total sense but is a bit of a funny double shoe horning situation here.
The birth and death of javascript remains accurate.
One thing I really care about is extensibility. Every now and then one of the big consumer apps adds a feature I really like and the self-hosted solution should have some way to integrate that.
The main thing I really care about is voice mode, as that's my far preferred way of interacting with LLMs for longer backs and forths (most apps I've seen disable a lot of other functionality during it, which I hate, btw).
Two other things I would like to see are canvas mode and scheduled actions (with decision making capability - e.g. "send a notification if X happens").
I assume such features are going to continue being invented, so I find extensibility to be a huge deal. So much so that one thing I could imagine going really well would be a UI on top of Langchain, which already has most of the facilities for that!
Totally makes sense! We've defined simple `Connector`, `Tool` and (soon) `Agent` interfaces to make it easy to plug in your own implementations/apps. If you wanted, you could just use Langchain under our Agent class to build arbitrary flows.
Additionally, the main chat is created from a series of `Renderer` components, and it should be easy to build your own.
Do you think that's in-line with what you're thinking of, or do you want to build outside of those confines?
> canvas mode and scheduled actions
Yes, writing and recurring are two big areas we want to tackle next year.
> The main thing I really care about is voice mode
> Do you think that's in-line with what you're thinking of
It definitely sounds good. Of course it's hard to anticipate what interfaces new features will require, it probably won't always be possible to integrate new features with a generic plugin (voice mode is most likely such a feature), but if the code is well architected, this should be okay.
> Interesting. Why do you think that is?
Writing has a lot of friction to me. It's much more comfortable for me to provide context through verbal rambling, which LLMs are great at processing. I like to research stuff and throw around ideas while pacing, doing chores or just lying with my eyes closed.
Unfortunately I just discovered that I won't be able to run Onyx on my low powered home server anyway (https://docs.onyx.app/deployment/getting_started/resourcing#...). I understand that a Vector database requires significant resources to run, but I wish there was a version without it.
It’s more like “why does Netflix kill my battery within an hour when I used to be able to play for 20”
reply