help-circle
rss

There is an official foss telegram client in the main f-droid repo. Packages on fdroid are built from source by the fdroid build server, not developers uploading their apks like on play store.

https://f-droid.org/packages/org.telegram.messenger/

There is a really small chance it’s compromised and could got unnoticed in the source code.


Very useful video. I miss that you don’t list the Chromium browsers. A lot of people, the target audience of this video don’t know that edge, opera, vivaldi, brave are all affected some way.


I’m running official Lineage, and it saves this data and I found no button for it, so I guess it’s not just Samsung only…

You can automate stuff like this on android, Easer and Automation are well known FOSS apps for these kind of tasks. IIRC in Easer you can call terminal commands


There’s no rational reason

https://xkcd.com/1172/

I don’t share screenshot frequently, and I store them for a long time, in 10 years in the future this data can be useful, I won’t remember what phone and rom I had, so it can be useful for some people. I also store gps data in my photos’ exifs, but again I never share them on the public internet.

But a button to switch it onoff would be useful, that’s true.


Yeah, wiki says it’s 6000 pages. But that’s not that long compared to other similar file standards, and it also contains pptx and xlsx.

For comparison PDF standard is about 1000 pages, HTML (without CSS, just pure HTML) is 1500 pages.


Docx is not a proprietary format, it’s a standard, it’s called Office Open XML: https://en.wikipedia.org/wiki/Office_Open_XML

And M$ published its specifications, so Libreoffice devs could support it. But here comes the funny part: M$ (deliberately?) doesn’t follow the specification it published. So the formatting problems of LibreOffice come from M$, because they don’t follow their specs, but M$ can just do whatever they want because of its market share.

I read this story a long time ago, and I’m paraphrasing, but on this wiki page you can read a lot of controversies related to this format: https://en.wikipedia.org/wiki/Standardization_of_Office_Open_XML


THe good thing with linux, that it’s totally transparent, you can easily find what’s wrong, if you look at the right place.

First I would check journalctl, dmesg, running processes with htop, if I can see something wrong, unexpected. Check hard drive smart and run short selft test with gsmartcontrol.

Here is a general troubleshooting guide from arch wiki: https://wiki.archlinux.org/title/General_troubleshooting I know you use ubuntu, but most of the commands here are the same on all distros, basically anything not related to pacman should be the same on your system

Ubuntu wiki has a similar guide, but shorter, guide, maybe start here: https://help.ubuntu.com/community/TroubleShootingGuide


Yes, this doesn’t seem official, it just started by some random forum user, it requires a registration on that forum, so the data will be very much skewed and limited. Signal has 100 million plus downloads on play store just for comparison, I don’t know what you can get from the vote of some hundred people.


Ah, ok, I misunderstood, because you started with the piped bot, and that’s totally different from this services. So you are not writing about imgur frontends, but for uploaders to upload their images elsewhere. Sorry for the misunderstanding, I just don’t like that bot.


But for embedding images, your privacy scan is not relevant, it’s just the home page of imgur you scanned. If you share the direct link to the image, no tracker downloaded, only the image. Just rightclick on the image and ‘Copy image link’. If you paste this link to this privacy inspector it says no trackers, as it’s not a webpage, just an image. They will know your ip and useragent, but that’s all, no extra tracking:

Of course it’s better to support more privacy respecting services, and thank you for the recommendations, but it’s not as big issue as it sounds. It doesn’t affect the viewers of the images just the uploaders.


I think it shouldn’t be managed on lemmy side but on the users’ side, e.g. on your device/browser. Libredirect can automatically redirect to those sites, not on just lemmy, but everywhere on the internet: https://libredirect.github.io/

Another problem is these alternative frontends relatively frequently disappear. If you post a link to a random instance, it’s quite possible that 5 years later the instance will be down, and the link won’t work at all. Libredirect addon updates the urls of working instances, so it will work later. There is even a button in the addon to switch to another instance, so you can find the best available site.

Edit: I misunderstood what are these sites, they are not imgur frontends but separate image host websites, so this comment is just about the first paragraph.


It seems you are right: https://www.magicearth.com/faq/#traffic-and-events-help

Where do you get the traffic and road closure info from?

We get it from a third-party provider; it is not created by us.

It’s limited to mostly first world countries: https://www.magicearth.com/feature-availablity/#hd_traffic

But somehow, someone tracking it’s users, I can’t imagine other way to get data like this. Maybe some carsharing service, company fleet management or something like that.


OrganicMaps only updates the maps monthly, so you will see your changes only on the next month. Also map updates are tied to app updates, so you have to update the app in your store first, then a button in the app will let you download the new map files. Related issues, more info about this:

In OsmAnd you can set up the map update frequency and you can even enable online maps, so you can set it the way you want.

I never used Magic Earth, so I don’t know how it works.


There are websites which scrape the play store, and let you download apks without login. The problem is you cannot be sure that you get the same apk as you would get from play store. (But actually you cannot be sure about anything on play store as well: the developers build the apks and upload it, e.g. an attacker impersonating the developer can publish a fake apk to play store.)

With these things in mind, you can download apks from apkpure without login to anything: https://apkpure.com/magic-earth-navigation-maps/com.generalmagic.magicearth Afaik they never had an incident where their apk was different from the one on play, but you cannot be sure when they change their mind.


Aha, I see, you can consider it whatever you want, maybe the “not fully free software” would be a better term, but “not open source” is too harsh, because source is open, as you can see it, but doesn’t fit the definition of Free Software as defined by FSF. If you use requirements by FSF, please use their terminology as well, it’s confusing.

Also please contact FSF, because they recommend this non-free app on their website: https://directory.fsf.org/wiki/Collection:Replicant-expanded#Navigation


Usually an Any app that knows traffic situation, knows it from your (and other users) location, so it obviously doesn’t know about traffic.

Edit: Magic Earth gets this data from a third party: https://lemm.ee/comment/1993667

In openstreetmap it’s not recommended to map temporary things, and the map only updates once a month in OrganicMaps, so that’s also expected.

https://wiki.openstreetmap.org/wiki/Good_practice#Don’t_map_temporary_events_and_temporary_features


Some icons of the undergrounds have different license. Read your first link carefully. And you link the source of the ui, or you don’t consider png files as “source”?

If it wouldn’t be foss, it couldn’t be built by the f-droid build system, it can only build foss projects

Edit: i was wrong