Although I'm slightly relieved there is a way out of Googles verification system, it's still pretty wild if you compare this to installing software on a Windows pc. I'm sure Microsoft is heading in the same direction with Windows, but today its still "only" a few confirmations to install anything.
This will sadly still put a major damper on adoption of open source apps, while giving a false sense of security that apps from the Play store are safe.
Years down the road, the low usage of apps installed from outside the Play store will be used as an argument for removing the functionality completely.
This looks pretty interesting from a laymans perspective. Can anyone with more in depth knowledge relate this to the current state-of-the-art auto regressive models? are diffusion models closing the gap in a meaningful way, or are they still in the experimental stage?
The court subtracted a fee for the artwork and for the display of the work, which he does not have to give back - so the artist did in fact get paid for his empty frame artwork. Now he just has to give back the money he lent for the originally agreed upon artwork that he did not create.
Seems like a fair deal, since he also got a ton of publicity out of it.
I wonder if this could be used to finally get universal audio streaming from the device, kind of like airplay. Perhaps using the received audio stream to create an internet radio station, which could then be played by most multiroom speakers, without needing specific support for each player app on the device.
Wow, this sounds like a really important decision. I wonder what Metas next line of defense will be, if any. The article says Meta can appeal, so I guess they will try that.
It will be interesting to see how many users will decline to opt-in to Personalized ads. Meta will probably still try a range of dark patterns to keep users from making that choice.
To avoid syncing my calendar and contacts to the cloud, I'm using wireguard as a always on VPN on my phone, to connect to a locally running CalDav/CardDav server. Wireguard on Android can be setup to only affect specific apps, so my DavX app is the only one using the VPN.
You might be interested in https://radicale.org/v3.html.
Runs on my odroid board with 26mb memory. The documentation is particularly good. I've used it as a replacement for the Synology CalDav and CardDav services.
It's very easy to install and does not require a DB. As a bonus, it stores everything as files which can be read and edited manually. It does require python.
Thunderbird mobile will be open source, so they, or anybody else can fork it, and continue development independently of Mozilla, using a different brand. It will not be "Mozilla property".
I'm happy to support Thunderbird; it's great to see a healthy project supported by donations.
It would be pretty great if Firefox development could be supported by direct donations too. Although maintaining a browser is probably orders of magnutide more expensive than Thunderbird.
This will be very dependent on the specific apps you need, but I have done almost the exact same process with some success. I switched to CalyxOS which has microG to "replace" Google Play Services. For banking, goverment ID, payment apps etc., this mostly seems to work. I use Aurora store to install and update the apps that are only in playstore, which sadly is still a lot of them.
I Use f-droid as much as possible. The odd app that doesn't work, can sometimes be replaced by using the website of the service instead, even though the experience is likely worse.
Some banks or government services allows you to buy a physical token to generate TOTP, and that comes in handy if the app doesn't work.
Dowsing has in fact not been proven to work. There is instead overwhelming evidence that it does not work better than chance.
Wikipedia sums it up nicely, with citations:
--
The motion of such dowsing devices is generally attributed to the ideomotor phenomenon,[7][8][9] a psychological response where a subject makes motions unconsciously. Put simply, dowsing rods respond to the user's accidental or involuntary movements.
The scientific evidence is that dowsing is no more effective than random chance.[10][11] It is therefore regarded as a pseudoscience.
--
https://en.m.wikipedia.org/wiki/Dowsing
For an entertaining time, check out the late James Randi and his debunking of dowsing
I highly recommend The Skeptics Guide To The Universe:
How to Know What’s Really Real in a World Increasingly Full of Fake
https://www.theskepticsguide.org/our-book
Its somewhat of a reference book and very accessible
A really hard problem that often arises when doing E2E tests, is creating and managing test data.
If you have one or more integrations to external systems, where you cannot control your test data, it becomes much harder to write stable E2E tests.
Some don't have test environments, some have too few. Most don't allow you to setup data easily either way.
You can, of course, mock the external systems, but if they play a large enough part, your tests start looking more like integration tests again, but with the added overhead of something like browser automation.
To me, this seem like a fine initiative, all things considered.
There's a bit of a false dichotomy in some comments, where Mozilla Rally is preventing or hampering Mozilla from improving Firefox. They can and should do both at the same time, especially if both are positive things.
It doesn't mean all their initiatives are equally good, and that we shouldn't still criticize the bad ones. But this seems benign at worst, and could possibly do some good. it's completely opt in, transparent, and with very strong user control over the data.
Instead of chastising them for all their other shortcomings yet again, I prefer to be happy that this might benefit ethical research.
But it really seems like an own goal if true.