I've tried one extra layer of complexity - "Remind me to water the plants when I get home" or something like that. It has no idea what I'm asking it to do despite knowing the "remind me" phrase and knowing where my home is.
Simple timers, controlling the lights, background music, reminders, and occasionally thing like "how many tablespoons in a pint".
I don't ask it for the weather because I don't want a 30 second lecture on the weather, and it won't give me "cold now, hot later, might rain between".
Given that Gemini seem to be unable to give short answers, and it can't control stuff, it's missing most of what I require.
Odd I seem to be able to control all my IOT devices no problem so far from Gemini with voice commands over the phone. This includes all smart switches, lights, thermostats, and so on in my house
It seems to function a little quicker in response with the voice commands IMO.
Really? I use the Google Homes and they say "today it'll be cloudy, with a high of x and a low of y".
Sometimes it also starts going off about "by the way, if you want a summary of your day, just say 'good morning'" or some shit like that which is annoying but not quite annoying enough to figure out how to turn off.
I do a job where my hands are potentially full and I cannot touch my phone for extended periods of time. I use the assistant for as much as it will let me...
BTW if anyone has an alternative for Pocket Casts that is free and multiplatform I'm all ears lol (so far Spotify seems like a good alternative, but I don't like it as a Podcast replacement app).
I use Antennapod on Android, it syncs with Gpodder (which requires self hosting on Nextcloud) and then there's plenty of compatible clients over on desktop
There's a lot of functionality that a lot of people could like using, but you have to go out of your way to learn about it all.
Gemini is slower to respond to tasks (for now, at least), can't do all the things assistant can do, and can't do anything at all without being on data.
After being on Graphene for the past 2 years,I decided to reinstall with no google at all,since my bank app now works without any google services dependencies.
hooray for me,I'm finally free and can profit from multiuser installed sandboxed services in another user profile if needed.
What do you mean exactly by “no google at all”? How do you get rid of it of not using Graphene (I know google pay doesn’t work with that, which is a big showstopper for me). Looking for alternatives :)
Since GrapheneOS added eSIM support withput google I'm tempted to try to remove Play Services from my installation. How do you find out which apps require Play Services?
I need a degoogle wiki to all me through all the things I can do to degoogle and what conveniences I might loose and potential workarounds for them. E.g. I spent a few hours getting rid of chrome, but then found out that if I want to use maps from my home screen, I can't use the search bar and I need another button on my homescreen. Also, apparently I use the images tab on Google often, and ddg doesn't have an images filter.
So anyway, without good replacements for my typical workflow, I end up just adding inconvenience and still falling back on the old workflows when I can't figure out how to get what I need degoogled.
It's really good at showing what you can safely disable, and is easy to re-enable stuff if you have problems. I've only ever had issues from disabling stuff listed as problematic.
I don't know of a good list like you're describing - your chrome issue was likely because you disabled the rendering engine, and not just chrome. Android needs a web rendering engine (the default is chrome), which can be replaced.
I've already replaced It with Home Assistant. Wake words are still shaky, but everything else is at least as good if you're willing to configure it. At least for my use cases.
I knew they were going to end up going full gemini, but this seems too soon to pull such shenanigans, considering it's still slow to respond, can't do several things that assistantcan do, and is 100% useless when not on data.
I tried it, then uninstalled pretty quickly. If I say, "Play music" then 10% of the time it would play music, and the other 90% it would tell me it can't. Same with many other assistant commands like controlling lights.
What even is the option here? When Google got rid of adding things to lists, I started my official transition away from them by moving to Proton, self hosting more stuff, etc. But for a voice assistant it seems like open source just isn't there yet, it doesn't have the hardware, and my only remaining option is to switch to Amazon (no.) Or Apple.
Is this gonna be one of those tech bandwagon things that Google fails at so consistently? You know, like Google phones, Google Plus, Google Pay, Google Stadia, Google Your Poodle, etc.
Google phones are some of the most popular in the marketplace (and their os is the most popular in the world) and Google pay works absolutely fine - I use it everyday, albeit with the new name of wallet.
Nobody's violating your rights by publishing an update to the software they provide. If you don't like it android gives you an option to freeze apps at a particular version. You can let go of your pearls now.
Make yourself a favor, delete this crap and instead use a real open source (means apache2) AI that performs even better than chatgpt4 or Google ai. Get Mistral 7B.
It doesn't seem to be able to actually take actions.
Things I used to do with assistant that do not work with Gemini
play something
Play "song title"
Play artist
I don't know about other things like seeing timers or running routines, I shut off Gemini as soon as I realized it couldn't handle my most used requests.
And I still try to figure out how to enable it. I just want to try it out so I can decide whether I like it or not, but so far it hasn't shown up for me...
On my samsung N20U when I use assistant for the past week it will sometimes pop up on screen asking if I'd like to try swapping to gemini. It only stays up a few seconds before going away.
It's not only the default: you can choose to use the old assistant on Gemini's config, but the next time you open Gemini it's as if you just installed it and it sets itself as the default again.
It looks like you can switch the assistant to the old one, and then turn that one off.
But just like Microsoft, Google is going to use this technology everywhere. If in the future (or now, if it is already available to you), you use features to describe images, summarize data, create texts, you probably will be using some form of Gemini.
The article claims that the default assistant for a new phone is Gemini, but it seems people who responded here haven't seen it. I already have the option to switch to Gemini, which I haven't.
I loved being able to open up Assistant to quickly summarize an article for me, but Gemini requires me to also attach the URL to its context and then type/say "Summarize" and then hit the submit button only to wait longer for the summary.
The only pro I can think about is that I can continue to ask for details about or related to the article.