Hacker Newsnew | past | comments | ask | show | jobs | submit | perchlorate's commentslogin

This is pure FUD. They're still trying to block it, the latest three attempts happened this week. Two of them were done in the middle of the night as training exercise, maybe for 3-4 hours each, and the last one then happened in the middle of the day. All three broke large parts of the internet and were quickly reverted.

When something newsworthy happens in some region, all messengers get blocked in that region for days, Telegram included. They don't care about collateral damage to other websites then.


I'm curious how the people attempting news blackouts reason about it.

I doubt they explicitly say to themselves, "Today I do evil for fun and profit.". I wonder what their rationalization is.


You're the one who is spreading FUD. Telegram was officially (legally) unblocked in 2020. There is 0 evidence there is an active force trying to block Telegram in Russia. Which is very busy blocking every non-russian platform btw. As you yourself pointed out, most likely the reason why TG was down is because of attempts to block other platforms.


[flagged]


And? I use TG everyday. I know about the times when TG is down. as YOUR OWN links show, usually it's not just telegram who is down, so it's clear it's mass block.


What is an example of something that is newsworthy?


Here is a couple of typical examples when blocking is limited to a single region:

https://storage.googleapis.com/gsc-link/cbe9d20e.html

https://t.me/agentstvonews/4973

https://t.me/meduzalive/94295


Just don't do that. Some of us (hello) live in countries that perform or tried to perform HTTPS MITM on a massive scale, and only had to roll back because so much well behaving shit broke.

If software suddenly started accepting invalid certificates, they would have no incentive of rolling it back. HTTPS would make zero sense then.


curl does accept if you enable the option to do so. It is optional


This doesn't make it a good idea to break HTTPS by default. Defaults matter, if everything ignored HTTPS errors by default, I would be talking to you over a MITMed connection right now. Because so much software stopped working, they had to roll back that braindead idea in less than a day.


You likely already know that, but to anyone else interested: a good way to prevent these kinds of situations is to run 'nosey parker' on your git repo before pushing it to a remote. It will dig through your code and configs, looking at files and through all the git history, and highlight anything that looks like tokens, passwords, keys, etc. You can set it as a pre-commit hook to block the offending code from even being committed.

https://github.com/praetorian-inc/noseyparker


https://github.com/trufflesecurity/trufflehog is a similar tool but checks for far more secrets, so I think it'd be a better choice.


Github has a similar feature that’s free for public repositories IIRC.


Someone linked to a Calibre recipe that assembles everything into one ebook. Here is the result if anyone else needs it:

https://www.mediafire.com/folder/b8fdqx7eqcpdl/linker

or

https://0x0.st/Xycy.azw3

https://0x0.st/Xyct.epub

https://0x0.st/Xycv.mobi

https://0x0.st/Xycw.pdf


Thanks. It looks like this includes comments which blows up the page count dramatically. A little unfortunate.


It's pretty easy to remove them by adding one line to the recipe and re-fetching:

https://www.mediafire.com/folder/u84s3art26lni/no-comments


Great, thanks!



A PDF? Oh! There it is. Thanks.


Plain Waydroid might work depending on the exact application you're using. Try that first because it's pretty seamless and has good performance (as Android processes run directly on the OS).

I used it for three years before giving up and getting a used phone (for other reasons). My banking application detected all other emulators, but not Waydroid.


thank you! i will try.


I don't like Namecheap, but this particular price increase is caused by the registry. I just checked Cloudflare, which sells domains at cost, and they're raising prices too:

https://paste.debian.net/plainh/2ae79637

.com is listed separately at $10.44

I don't think they have a public link for this.


Verisign hiked prices, once again: https://www.nasdaq.com/articles/verisign-vrsn-q4-domain-name...

> VeriSign announced that from Sep 1, 2024, it will increase the annual registry-level wholesale price for .net domain names by 67 cents to $10.26 from $9.59.


Just set mitigations=off if you want this, it's a meta-parameter that enables all the others. I used to run it on old hardware (Haswell) because of significant impact of all those mitigations on IO performance. It's not a problem anymore on my newer Zen 4.

Probably a bad idea in the general case if you work with anything remotely private (ssh keys, banking) and browse the internet routinely. I visit few weird sites and block JS almost everywhere, so it both made sense to me and was relatively safe.


Advertisers are going to use it as just another data point, though. Why give them any more information than they already have? You know some of the best technical people we have waste their lives working for advertisers, and they _will_ find a way to extract more information from these technologies regardless of how supposedly private they are.

I don't think there can be a technical solution to this problem unless advertisers are forced by government regulations to behave, with very heavy fines for non-compliance.

I personally really hoped for Mozilla to take a strong stance against advertisers and introduce an aggressive ad blocker, but it was pretty obvious they're not going to do anything like that because of the conflict of interest (Google's money). Now we get another conflict of interest on top that makes it even less likely.


> I don't think there can be a technical solution to this problem unless advertisers are forced by government regulations to behave, with very heavy fines for non-compliance.

Indeed, but such regulation is much more likely to happen if there already exist privacy-preserving solutions like PPA.

"There's a good way to do it and a bad way, and we're making it illegal to do it the bad way" vs "There only way to do it is bad, and we're making that illegal, so sucks to be you"


This isn't an alternative, that's the argument


AV1 is already faster and has been for around three years — if you're not using the reference encoder, which is dog slow, and that will probably not change. Have a gander at SVT-AV1.


This is very outdated information. AV1 encoding with SVT-AV1, which is a high performance encoder written by Netflix and Intel, works faster than x265 (and other h.265 encoders), and is much closer to x264 IIRC. With lower quality presets it worked at 3-4× on my old Haswell i5. On higher ones that still make sense by their own recommendations, the speed was around 1×. That CPU is over 10 years old.

https://gitlab.com/AOMediaCodec/SVT-AV1


Encoding video on CPU is a no-go, AV1 is only supported on Alchemist IIRC. Certainly no hardware I possess has accelerated AV1 encode. I can do HEVC at 8x, SVT-AV1 manages 0.5x for the same file. There's just no comparison.


On which preset for SVT-AV1? One of the strengths of SVT-AV1 is the presets offer a very wide trade off between quality and complexity. See the chart here:

https://engineering.fb.com/2023/02/21/video-engineering/av1-...

If you try a preset like 8 you may find you get good enough quality and a fast enough encode at a good enough bitrate. Some encoding guides:

https://gitlab.com/AOMediaCodec/SVT-AV1/-/blob/master/Docs/F...

https://trac.ffmpeg.org/wiki/Encode/AV1

You'll have to play around with the parameters to get the results you want. Try 5 minute clips until you're happy with it.

And use SVT-AV1 2.2 since that has more performance improvements:

https://www.phoronix.com/news/SVT-AV1-2.2-Released


Even using -preset 12 and it's still 8x slower than hevc_nvenc. Encoding on CPU just isn't viable, not to mention the enormous perf impact on everything else running on my computer vs next to no perf impact when I use nvenc.


It's always going to be slower than a hardware encoder.

I have an AMD Ryzen 7 7840U laptop which has hardware AV1 encoding. If I use preset 8 with SVT-AV1 on a 1080p encode I get 30 frames per second:

ffmpeg.exe -i "input.mp4" -c:v libsvtav1 -pix_fmt yuv420p10le -crf 40 -preset 8 -g 240 -svtav1-params tune=0 -c:a libopus -b:a 128k av1.mp4

With preset 12 I get 143 fps. With preset 10 I get 84 fps.

If I use the hardware encoder I get anywhere from 15 fps to 589 fps depending on the settings I use. I get 103 fps with this:

ffmpeg.exe -i "input.mp4" -c:v av1_amf -pix_fmt yuv420p -quality high_quality -preencode true -b:v 4M -c:a libopus -b:a 128k av1.mp4

Hardware encoder parameters: https://github.com/GPUOpen-LibrariesAndSDKs/AMF/wiki/AMF%20E...


You can get any encoder to be extremely fast if you don't care about quality and turn off all the tools.

Generally speaking x265 was never know to be fast. Even Netflix moved their HEVC encoder away from x265.

And My question is where is AV2?


If you are doing CPU encodes, SVT-AV1 M12 is faster than x264 veryfast and produces higher quality output than x264 veryslow.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: