For sure he notices; the author runs their own email server and
founded a direct competitor to WhatsApp. The author is making the
point that what each of those have done - build proprietary software
around federated protocols - is a financially lucrative business
model. I'm sad to agree.
FWIW my opinion is that Signal's actions against these clients is
petty and just shit. Thankfully, elsewhere we can see things
happening differently: the interaction between Tailscale, Headscale
and Wireguard gives me hope. Sourcehut is a cool project too.
I guess so - I've actually never used Teams. There are lots of potential mitigations, but sandboxing is not really a solution to buggy code. For some better engineering discussion on the topic, there's the series of articles Transparent Telemetry, in particular The Design of Transparent Telemetry.
I wonder whether that many people actually do this. From my quick look, I saw jobs flooded with offers for well below the minimum wage of my country (Australia).
For many jobs the ones doing the hiring are thinking of their domain, so more experience in the domain means a better worker. But a software developer who has developed CRUD apps 50 times on-budget and on-time over 20 years is almost certainly going to be a fantastic candidate alongside the dev who specialised in the health insurance (or whatever) domain for the entirety of their 5-year career.
Now I'm aiming for more software-focused companies and consultancies since I think I'm more likely to meet people who appreciate that broader experience.
I know where you're coming from. Right now the way many of podcasts' audio files are served is via HTTP CDNs. The podcast client fetches the RSS feed, then fetches the linked-to audio file.
The VPN, as you say, just changes the source address of that request.
What we could work on is reducing the number of requests to those CDNs.
One idea: A service which serves a mirror of the podcast feed and audio files.
Users would need to manually enter podcast feed URLs into their client,
rather than select the podcast from a convenient in-app search.
You'd have to trust the service operator isn't collecting and sharing its usage data.
Going further,
we could use Bittorrent to distribute episodes between mirrors.
Mirrors could subscribe to a RSS feed of torrents for particular shows.
I could imagine some community-run effort in this space.
Definitely creepy that it phones home in the first place.
But it's not necessarily creepy that it keeps trying; it could just be sloppy programming. Hanlon's Razor comes to mind. Microsoft Teams behaved in a similar way apparently. If you blocked it phoning home at the network level it would buffer gigabytes of data on disk until the disk was full.
Nice! I actually didn't know they had RSS feeds at all,
so I'm going to help my news junky friend get set up with them.
Thanks for the tip on the ABC RSS! :)
Right now I've got NetNewsWire configured to use my iCloud account
(which uses CloudKit under the hood),
so that syncs my iPhone and iPad.
This still involves a 3rd party -
Apple's magic hosted database service thing -
but I'm ok with this for now.
Ideally I would set up something like FreshRSS
and host it on one of my own servers.
FreshRSS exposes the old Google Reader API which NetNewsWire and Reeder can use to sync feeds.
Personally I'm not interested in the FreshRSS web interface
and I'm not too keen on running a PHP app either.
So I'm considering writing my own service which serves the bare minimum required to be able to point
apps like NetNewsWire and Reeder at.
But whether usage data is transmitted back to the service needs a quick check too.
For example, compare accessing Gmail via its web app and via a boring IMAP mail client.
Or in the old days, Twitter's web app and an app like Tweetbot.
For news in particular, RSS feeds are a great option if you can find them. No ads, no Javascript, purely chronological sorting of articles.
Here's some feed links for SBS, for example: https://www.sbs.com.au/news/article/feeds/nbv1rs3kw
I highly recommend the open source app NetNewsWire for iOS.
In this case, is using the web version of the app (which is often an option) more private?
I’m assuming mobile Safari with privacy relay, plus some extensions to stop trackers etc.
Long story short: probably.
The crucial bit in this example is that the extensions explicitly blocks code being executed on your device used to track you.
They aren't a silver bullet, though.
Officially on iOS it is forbidden to track users' activity on between apps and websites unless the user explicitly grants it via the AppTrackingTransparency framework (https://developer.apple.com/app-store/user-privacy-and-data-use/).
Not a silver bullet either.
Whether there is a significant difference how usage data is used and abused from accessing content via a website versus mobile app is a tricky question to answer definitively.
We can measure things like network requests, blocked scripts.
We can read policies which promise stuff and can be enforced through courts etc.
But things happen behind closed doors like selling data through legal loopholes, grey areas, and data breaches.
It's a big business.
If you've got any specific examples we could dig a bit deeper.
Assuming you're Australian: I just checked first news provider that came to mind and found that https://abc.net.au can be read easily without Javascript enabled at all.
That's hard to beat.
Nice digging. Someone savvy may be able to extract the ANZ Shield APK using apktool,
then maybe some decompilation from there to find any Symantec VIP libraries...
and log files eating up storage space was a common culprit.
Another classic symptom of poorly maintained software.
Constant announcements of trivial nonsense, like [INFO]: Sum(1, 1) - got result 2! filling up disks.
I don't know if the systems you're talking about are like this, but it wouldn't surprise me!
Interesting. It sort of does save them - but for how long? Depending on who is overseeing what projects the brand loyalty could last long enough for the particular people responsible. It’s a terrible strategy but that takes a lot less work than coming up with a new one!
Revenue rose 16.4 percent to a record €61.8 billion while operating profit climbed nearly 40 percent to an all-time high of €7.6 billion.
The brand is strong relative to Chinese competitors but I don't think it will stay this way forever.
I think they want to move fast, but they simply aren't able to do so.
They want to move as fast as they can maintain their profits.
I think major shareholders would ideally like to see more tangible results from their R&D division. But it was clear at the time that it didn't matter enough for real action.
Middle management I interacted with were actively hostile to me when I spoke about, for example, making source code visible between teams. There was constant calculated behaviour to keep things the way they were and delay completion to maintain funding.
For sure he notices; the author runs their own email server and founded a direct competitor to WhatsApp. The author is making the point that what each of those have done - build proprietary software around federated protocols - is a financially lucrative business model. I'm sad to agree.
FWIW my opinion is that Signal's actions against these clients is petty and just shit. Thankfully, elsewhere we can see things happening differently: the interaction between Tailscale, Headscale and Wireguard gives me hope. Sourcehut is a cool project too.