janinge

caniuse doesn't differentiate between markets, but it appears that even here within the EU we're now back to being allowed to use most basic features.

However, previously you got an exemption on the 7 day TTL on Web Storage, by being installed as a PWA. This must have been too generous. I guess they figured that if you combine lack of stable local storage, with no support for the Background Synchronization API, nobody should be able to implement anything of value on top of these APIs.

And just in case this isn't enough, let's just hide the install option even deeper than before. On iOS the path is now something along the lines of

1. Tap site's domain name 2. Meatballs menu 3. Share 4. More (Meatballs again) 5. Scroll until you find "Add to Home Screen" 6. Leave "Open as Web App" switched on 7. Add (upper right corner)

There is no obvious install prompt, and no clear browser level affordance. Nothing in the flow even tells you if PWA functionality is supported by the website. You might end up with just a weird shortcut that opens a URL full screen.

OpenWRT also got it packaged. So a solid and easy way to run it is to just throw the client on some old routers with OpenWRT support, connected to $20 USB DACs with S/PDIF.

https://www.behringer.com/product.html?modelCode=P0484

Yes, you wouldn't get this running on an existing smart speaker (without first rooting it and some serious hacking).

If you'r in the Apple ecosystem and are using AirPlay with your smart speaker(s), it's however possible to also play synchronized audio across to your own DIY speaker setup, using another open source project.

https://github.com/mikebrady/shairport-sync

Or you could of course choose to only use your old dumb speakers with this, and they will pop up as easily selectable sound output devices on all Apple devices connected to your network.

Or combine it (and librespot[2], owntone[3]...) with Snapcast to create a virtual speaker for your whole house that shows up everywhere.

[2] https://github.com/librespot-org/librespot

[3] https://github.com/owntone/owntone-server

Yes. If you’ve bought a consumer router, and not an “access point”, it will likely have a Ethernet port labeled WAN. Due to this label, the OpenWRT profile for this model will most likely also set up this port as a WAN port for you by default (where it requests DHCP from somewhere and applies NAT and some basic firewall rules). Just delete this interface, and make the LAN network also span the WAN port. Then disable DHCP and IPv6 RA on the LAN interface. Your router is now a dumb “access point”.

You could also make the OpenWRT router connect trough your phone’s hotspot feature temporarily, while you set up your primary connection. Three clicks in the GUI is all that’s needed to join a wireless network for WAN connectivity.

Zoom also use the Selective Forwarding Unit (SFU) architecture. Or a central server that kind of "mix" streams and send them to all participants, if you will.

Jitsi Videobridge does this by receiving simulcast streams from each participant. This server individually picks out streams and qualities that will fit in each recipient's downstream pipe based on measured bandwidth and configured priorities (e.g. you could choose to give more bandwidth to those who are actively speaking, or shut down all but the last N speakers video streams).

Yes, the available network bandwith on a Jitsi Videobridge could be a bottleneck. Each meeting need to fit on a single bridge instance. However, using common servers connected at 1 Gbps or 10 Gbps, it shouldn't be any problem to have meetings with substantially more than 2-4 people. Say a HD stream from most webcameras outputs 3 Mbps at full blast, with 50% overhead due to the simulcasting. That's over 2000 participants on a single server connected at 10 Gbps, all receiving all video streams in full quality.

For meetings with 2 participants the streams run peer to peer, so here you get the best possible latency and quality.

We've set up Jitsi on a VPS instance at a nearby provider, and have not seen any problems with meetings of 10-15 people. Stability and performance also outperforms all the other solutions I've tried. There's currently a bug in Firefox's simulcast implementation, so if you have any participant using this browser, this feature gets disabled for everyone. Even with this issue, I haven't noticed anything but excellent performance with this few number of participants.

The following article might also be interesting.

https://bloggeek.me/webrtc-vs-zoom-video-quality/

I was referring to the padding attack. Did they patch this?

And are there any property of MTProto that makes it infeasible to replace AES IGE in a later revision of the protocol?

It is not, or was not? Did this issue get rectified by Telegram?

I really like this feature. A page redraw generally takes less time than the time needed for my eyes and brain to re-parse the page. When I have to stare at a blank page for a varying amount of time after having decided to go back, the entire navigation process feels very sluggish and slow. Using other browsers I find myself opening a lot more temporary tabs to account for this. Swipe and hold is also nice when you just need to go back to reread a few sentences, etc.

Most pages shouldn't need a refresh when you go back after a few seconds or minutes. What annoys me are websites that set stupid expiry times or Cache-Control values.

Having a modernized pause/break key on your keyboard can also be handy if you ever have to use software that acts stupid on purpose, like these "Safe Exam LockDown Browsers" that some schools and universities around here have started forcing upon their students. Software like this typically just run a timer or similar to make sure that only their windows are running in the foreground, but sending them a SIGSTOP stops this annoyance.

This problem is mentioned in the article.

"Mastering engineer Bob Ludwig created ultra-loud master of Led Zeppelin II, but his version was pulled when it skipped on a record player owned by Atlantic boss Ahmet Ertegün’s daughter"

For that to work the networks themselves would have to securely distribute a list of locations, or it would have to be configurable on the devices. Many business and educational networks (like eduroam) span multiple locations. Even my "home" network is available multiple places (home, cottage, boat...).

Smartphones mostly use wireless networks and cell towers to determine their approximate location, which can be easily spoofed, except for the current active cell (which could be miles away). If devices had to acquire GPS fix every time they reconnected to a network, batteries would drain much faster. And satellite navigation doesn't work properly indoors. Civilian GPS can also be spoofed.

Manufacturers would probably prioritize usability over rectifying such a "problem" which never had bothered anyone before, except maybe if there was PR involved. I think there's still no way to list all configured wireless networks on iOS devices? Fixing this would probably improve privacy more (if people cared) than this randomized MAC feature.

For iOS you can use iPhone Configuration Utility or similar to add profiles for WiFi-networks, and set their SSIDs to be always broadcasting. That option should make it so that those names aren't included in the active scans, if it is to make any sense.

> The HummingBoard uses a Freescale i.MX6 quad core...

Current i.MX 6 silicon aren't able squeeze out more than 470 Mbps from their Ethernet MAC.

http://cache.freescale.com/files/32bit/doc/errata/IMX6DQCE.p...

ERR004512

No, it worked mostly on my desktop last time I tried it (didn't seem robust vs. 500 errors from Google, and crashed a few times). However, it's not small when you include all the Ocaml libraries it depends on, so it's not an option if you want to run it from built-in flash. And cross compiling with Ocaml isn't quite straightforward, so you'll also need to maintain a build environment on ARM, MIPS or whatever you have (e.g. using QEMU).

I stopped using Dropbox when their OS X client repeatedly got stuck syncing on most of my machines, and started pegging the CPU and draining my batteries. Tried reinstalling a few times months apart and promptly hit the same problem each time.

Google Drive could maybe be useful if there existed some working FUSE implementation that would fit on cheap single board computers like the BeagleBoards and Raspberry Pi.

Range with this setup and a Phantom isn't much more than 10 meters, from experience. The wifi module on the GoPro runs a web server where segments for HLS are published, so every packet and frame must be acked by the the phone and segments downloaded fully, which might not be the best when you want a bit range on the thing.

Hopefully they reject this mail, not simply discard it.

I was also happy with Hetzner when I used them a few years back. But I didn't need their support other than for replacing hard drives, which they always handled quickly and without problems.

One of my servers were once used in a amplification attack (DNSSEC...) for a few days before I noticed. I guess Hetzner didn't detect this because just the uplink got saturated. Had to manually request a null route so I could SSH to another IP alias on the box. I wouldn't mind if they automatically did this for me since the offending IP would be unavailable either way. At least they don't charge you for DDoS traffic, like my current European budget provider does.

If you move to something like Cloudflare, make sure to at least firewall off everything but their IP addresses. Otherwise it will be trivial for the attacker to connect to all the port 80's in the IP allocation to the provider they know you were using, and compare the responses to what they get from Cloudflare, to obtain your service's origin.

http://www.youtube.com/watch?v=bmzHIB18XT8

They have multiple encodes in various qualities for each title too, though.

Same (opposite as OP) experience here in one of the Scandinavian countries (on TeliaSonera's network). Buffering and seeking always happens almost instantly with Vimeo, and they deliver better video quality.

YouTube often fails to load videos here, and I have to switch between bandwidth settings (up/down, doesn't matter) to get things going.