Redesign is hard because there always users that like the current/old design. For me the new sidebar looks nice and cleaner (I use it on gitlab.com for 3 weeks now).

It was communicated transparently: it was abused too often.

It's not a secret that a no-log policy also attracts abuse.

https://mullvad.net/de/blog/2023/5/29/removing-the-support-f...

Cool, I have 8 months prepaid for a service I can no longer use because they have a months notice they're removing a feature I need. And they refuse to refund crypto, the payment method they supposedly prefer.

What I get for trusting mullvad I guess.

What's the feature that's being removed that's making this unusable for you?

Funny enough, in most cases it's the other way around.

Set up and forget is a often a bad idea. Keyword: Security Updates

> via Bluetooth too

Wouldn't recommend, bluetooth has the disadvantage of being very slow.

i wrongly assumed BT5 would be ok sharing a 3G connection recently... oh how wrong i was. BT is pure garbage.

after seeing the abysmal bandwidth i understood why every headphone becomes garbage when you start using the microphone channel.

there's no hope for BT and it must go on favor of anything else from this century.

> i understood why every headphone becomes garbage when you start using the microphone channel

AFAIK, this is a codec issue, with most codecs not supporting bidirectional audio. IIRC, newer Bluetooth revisions tackle this.

USB2 speeds are better, but not that much.

USB2 has a cap of 480mbps (includes overhead) while Bluetooth is stuck at several megabits per second (1.7 for Bluetooth 4, should be around double for Bluetooth 5).

WiFi can easily beat USB2 but Bluetooth certainly can't.

USB2 can do 480mbit (connection at least) and has the advantage of working properly even in a noisy wifi area.

We're in 2023 though. I get about 1.6Gbps over my phone's CDC-NCM interface in iperf over USB 3.0.

> without drining my phone battery for a hotspot

I faced a similar problem. I have now invested my money in an external battery instead to charge my phone if necessary. For me, this seemed like a simpler solution (also, I don't have to worry about a second SIM this way).

For iPhones, there are now batteries that do not even need cables.

https://www.anker.com/products/a1611

https://www.apple.com/shop/product/MJWY3AM/A/magsafe-battery...

I was disappointed with how little charge the Anker 622 holds.

They advertise the 633 to hold 1.3 charges for an iPhone 13.

The 622, at half the capacity would only extend battery life by ~60% (not even one charge which is a big bummer)

Interestingly the 633 is priced at $79.99 in the US, and €79.99 in EU. The EU price is therefore equivalent to about $88.30.

On top of that, they currently have a $10 discount voucher in the US. https://www.anker.com/products/a1641?ref=select_your_product...

Whereas the EU does not get any discount currently https://www.anker.com/eu-en/products/a1641?variant=415101967...

So that’s $69.99 for the US vs $88.30 for the EU, for the same product.

The difference is more than explained by VAT.

> They advertise the 633 to hold 1.3 charges for an iPhone 13.

I think you have a typo there. The product page for 633 says 1.8 charges for the iPhone 13.

I was confused why 60% extra would seem low for half capacity of 1.3 charges (130%). Dividing 130% by two gives 65%.

But since it’s actually 180%, I agree that 60% extra seems low. Should have been about 90% extra one would think.

You’re right, it’s 180% and I guess extra 90% for the 622, which makes Apple’s offering a bit ridiculous IMO (60% additional charge for double the price of the 622)

Thought the same thing. Finding the root cause is easier when you don't have to debug dozens of abstraction layers (KVM, Containerd/Docker, Kubernetes, etc).

Operating at this scale you can assume that all these things you mentioned do exist :)

It's all public: https://wikitech.wikimedia.org/wiki/Wikimedia_infrastructure

You can even get access to devops/SREs tickets (and IRC messages)

Docker? You’re probably right. K8s? Wouldn’t be so sure. Would like to hear from the horse’s mouth though.

Edit: looks like it’s used. Wonder if it’s more tooling than production traffic. No time to dig further.

Check whether an alternative firmware already exists (e.g. OpenWrt or DD-WRT).

https://openwrt.org/toh/start

> because I'd hit some limit (..) 10 days max is pretty bad

The free(!) plan includes 300,000 queries per month. That is transparent everywhere. On the website, pricing page and in the dashboard. You can even see how much you have already used. That's pretty fair for a free service for which you pay nothing. For me, that's enough for a whole month.

> (..) which doesn't arbitrarily stop working

1. NextDNS sends an email before the limit is reached

2. Once the limit is reached, no more ads are blocked, but dns continues to work

Hey if it works for you, great. It didn't work for me. 10 days casual use for 1 person and then "pay me" is not a positive experience, so I found another solution.

How many days should a paid service give you free casual use for a positive experience?

You are using a consumable (CPU & data transfer, along with your desired amount of analytics data storage) that they pay for, and, shockingly, they are not ad supported.

What? It’s a paid for service.

I currently use a pi-hole. In a home of 2, we have 74k queries per day. That would effectively be a 4 day trial... so I can understand the other person's frustration.

GitLab.com scheduled pipelines. You get 400 free minutes per month and can do whatever you want.

https://docs.gitlab.com/ee/ci/pipelines/schedules.html