I’m honestly blown away by whomever finds this surprising. This is Microsoft we’re talking about. Everything they touch turns into this. Taking what is not theirs, using it for profit, and not even giving credit where credit is due.
Hi, quick question. I already have some apps installed via play store, thats also available on F-Droid. Is there any advantage to uninstalling them, and installing them back, from F-Droid?
Most privacy tools from the PlayStore are light versions “decaffeinated” by Google. Google hate all apps which avoid his control. Recommended app for Android is InviZible Pro (FOSS)
What about the time they fired their artists and then immediately wrote a blog post congratulating themselves for making AI art from a model trained on the ex-employees’ art. Inspiring.
This looks interesting, but I don’t understand what it’s for. I read through the readme, but came out none the wiser. What exactly is a compose sequence?
A compose key (sometimes called multi key) is a key on a computer keyboard that indicates that the following (usually 2 or more) keystrokes trigger the insertion of an alternate character, typically a precomposed character or a symbol.
It’s a method to combine several characters on your keyboard and use it to create a special character which is not on the keyboard. For example “ and e produces ë. This tool allows you to configure those combinations.
But thanks for the feedback. I’ll update the readme to add some more context.
Important note: app developers don’t publish their apps on the official F-Droid repository. Other people (maintainers) download source code and compiling these apps. Therefore, updates are delayed by a week. You cannot update the app from other source because F-Droid version signed by a different key, so you must reinstall the app, deleting all the data.
I started using Obtainium to get updates directly from GitHub. It also has support for F-Droid and many other sources. I use F-Droid website mostly to discover apps.
Depends on whether you’re going to install apps from the official F-Droid repository or not. Third party F-droid repos (like IzzyOnDroid) are not affected by this.
Suppose you have some app (a hypothetical Lemmy app) installed from the official F-Droid repo. You logged in an account, changed some settings. Then the developer announces an update: new features, bug and security fixes. It is published on GitHub and Google Play. F-Droid version will come after a few days, when the maintainer builds the app from source and publishes that update.
You may don’t want to wait till update comes to F-droid. But you can’t install it from GitHub or Google Play, because it is signed by a different key. You’ll have to reinstall the app, which will erase your settings and require logging in again.
This is the hassle you probably may encounter in the future. If you want to avoid it, install official packages from the developers (from GitHub or Google Play). Obtainium can check for updates on GitHub, official and third-party F-Droid repos, and more.
For some it’s a trust issue since they F-droid is a middleman however with reproducible builds I don’t have that worry. Feels like a good thing to have the code verified twice. I have obtainium as well but had issues with some beta apps I wanted to use like organic maps. Obtainium only works if it is the way it’s setup. Organic maps did not follow obtainiums way
Also to use obtainium you actually need to know the app exist, I have a lot of stuff from F-droid I would not have found without it
Your right, now I did remove FFUpdater, I’ve got Mull updated adding DivestOS Official repo to Droidify (included, but you need to enable), and Cromite from GitHub.
Because Mull from default repo is 1 version and 2 weeks behind.
I feel very dumb but I just installed it and I don’t get the logic: do I have to enter every app into Obtainium to make it happen or is there some kind of auto discovery / import that I can use?
Thanks for your answer that’s what I was afraid of, that’s too much of a chore to hunt for every source and put it in the app, I’ll pass for now and keep getting (more or less) delayed updates as usual
That specific repository has no releases so it won’t work AFAIK. You need a repository with releases, that have apk attached. (Typically the developer would set up a CI workflow to build and attach apk for every release)
Edit: For example AuroraStore has releases with apks. So you can just enter gitlab repo for AuroraStore into Obtainium and it will install it and keep it updated.
Any benefit over just downloading the APK directly without managing it through Obtanium? I ask cause I’m guessing that for Mull, since there are no releases on its GitLab repo, it’ll just have to be downloaded directly.
Just to get it out there… I checked this out about a year ago. It’s not completely open source. The project consists of many executables and “pre complied dependencies” that don’t appear to share matching checksums which may indicate modifications of some sort. Looks like a great tool, but I’m extremely skeptical of what’s going on under the hood.
Hopefully they do truly open source it and prove me wrong, I’d love to give it a try some day.
You should check out Droid-ify! It’s a much more friendlier alternative to F-Droid, and also has more applications by default (gets some apps directly from Github).
There’s no debate. LLMs are plagiarism with extra steps. They take data (usually illegally) wholesale and then launder it.
A lot of people have been doing research into the ethics of these systems and that’s more or less what they found. The reason why they’re black boxes is precisely the reason we all suspected; they were made that way because if they weren’t we’d all see them for what they are.
The reason they are blackboxes is because they are function approximators with billions of parameters. Theory has not caught up with practical results. This is why you tune hyperparameters (learning rate, number of layers, number of neurons ina layer, etc.) and have multiple iterations of training to get an approximation of the distribution of the inputs. Training is also sensitive to the order of inputs to the network. A network trained on the same training set but in a different order might converge to an entirely different function. This is why you train on the same inputs in random order over multiple episodes to hopefully average out such variations. They are blackboxes simply because you can’t yet prove theoretically the function it has approximated or converged to given the input.
opensource
Hot
This magazine is from a federated server and may be incomplete. Browse more on the original instance.