I once used Zoom's web interface and the video quality was horrible. When I switched to the downloaded app, it was much better. Not sure if this was a one-off problem or an underinvestment in their web interface.
replies(2):
Beyond the privacy concerns it's a security risk. Their desktop client had a remote code execution vulnerability last year:
https://www.zdnet.com/article/researcher-says-zoom-web-serve...
On the other hand tampermonkey/greasemonkey are content scripts that get injected into loaded pages. An implementation here would look like: the user navigates to a Zoom meeting, they load the entire page, and then a script gets injected to perform a `window.location` redirect. This will be slower and depending on the timing of events you may even still get the Zoom file download prompt.
So I don't think tampermonkey/greasemonkey is a good fit here.
I refuse to use or install their product.
They also limit the resolution to 480p for the web app, probably because of performances. Browsers and zoom both use h264, but browsers usually use vp8 instead.
There is no reason webrtc cannot offer the same quality (or better) than zoom, at the same bit rate, but it all comes down to the actual implementation of browsers and web apps. webrtc-based apps work well, these days.
zoom seriously needs to die. no friggin way I'd ever engage in a responsible disclosure with this company - no matter who gets thrown under the bus.
this isn't the first time zoom got caught red-handed[1]. Last year they were called out for installing a local web server in order to disable security controls to get around the deprecated NPAPI[2] ... this is _literally_ what malware does. Seriously fuck zoom!
[0] https://news.ycombinator.com/item?id=22658173
[1] https://medium.com/bugbountywriteup/zoom-zero-day-4-million-...
I expect it is more to do with the fact that you don't need the Zoom app and why should you have to download it when pretty much everything works perfectly through the web client?
If they would be a Chinese company they'd be banned and probably even sanctioned. Stop using this shit and stop justifying its use just because your employer makes you use it. Grow some balls (or eggs) and speak up naming it for what this is (malware) - so that we can all have nice things and not be forced to engage in endlessly justifying ourselves because "team or company XYZ is using it too and it works great for them ..."
You're commenting on a post that is about a link that helps people use a web version of Zoom, which by its definition doesn't have the malware issues that people talk about (unless they are breaking sandboxing in the browser which would be pretty major).
What I was replying to was the "no grey area allowed" black and white dying on a hill response to the existence of the tool at all. This is why non technical people roll their eyes at technical folks and ignore us, because so many of us live in this world where we aren't willing to negotiate or hold more than a single thought in our heads at once.
I don't want to use Zoom, I bring up alternatives at my org all the time, and meetings that I control do not use it, and I do not install their binaries on my own devices, instead opting to use the web client when required. But the reality is that I don't get to make that call all the time, and if it's a choice between using Zoom on the web and not communicating at all, then the choice seems pretty clear to me.
if we don't speak up now and give them FIRE, then the covid19 crisis will have been the reason why another surveillance technology gets normalized. working under tracking a la "upwork.com" - where marketeers decide how to screen capture and key-log all input is somehow normal.
note: I'm not attacking your point and didn't think you agree to Zoom's way of doing things. I just feel really strongly about not giving them any benefit of the doubt because they have already got a history of abusing trust.
my comment in the sibling thread mentions why this literally can't be fixed with a browser add-on: https://news.ycombinator.com/item?id=22662212
again: not an attack on your comment, not attacking OP's work either. and we probably agree on more than we disagree here by what I can tell
By "actively bypass the security" do you mean "it's a program that you need to install on your computer"?
Can you elaborate why Zoom is malware in ways that VS Code, VLC Media Player or Photoshop aren't?
EDIT: I mean the question honestly, as a question. I might have missed something. I mean, I saw yesterday's HN topic on a tweet that claims it sends info about all active programs to a server. But I saw nothing to substantiate that other than an "attention tracking" feature which is way less invasive than what's described in that tweet and off by default.
Did I miss the evidence, or some other damning privacy invading misfeature?
So far, the response as been zero (it's been an hour) but let's see. Maybe I can make some coworkers and clients be more wary of Zoom.
For example: https://www.theverge.com/2019/7/8/20687014/zoom-security-fla... By design, instead of using a URL handler, they run a HTTP server on your machine to bypass the "open with" dialog. There are good reasons not to trust the binaries they ask you to run.
Here, it turns out they offer a web client after all, which is nice and sandboxed, but they default to trying to run a binary on your machine where you have less control over what it does.
[0] https://support.zoom.us/hc/en-us/categories/200110033-H-323-... [1] https://www.gnugk.org/h323-endpoint.html
What we're seeing is the constant balance between easy-of-use VS privacy/security and people making the choice without really understanding the consequences.
In the end, privacy/security is so invisible and hard to understand for "normal" people that easy-of-use and performance always seems to win in our current system.
> Stop using this shit and stop justifying it's use just because your employer makes you use it.
This seems a bit simplistic. We're all humans and navigating what's almost a political situation regarding Zoom and it's issues can hurt you individually, while not gaining a lot globally.
I'm not saying you're wrong, ideally we should all stand up for what we believe in. But sometimes the contexts and environments prevent us from doing so, and not all of us are ready to die on the hill that is privacy/security.
As I understand issue linked above, they want to support it but currently does not, in official releases. What's supported is catching all requests in a page, but that's after the contentscript has been applied to the page, which means what arkadiyt wrote would still be accurate.
Whereby (formerly appear.in) https://whereby.com/ has a really nice and simple system. No more jumping through a dozen hoops, no more installing software with glaring security holes and borderline malware behavior (looking at you, Zoom).
Seems like they don't, and haven't since July.
at some point in the last 100 or 200 years technology has become so complex that society has agreed to compartmentalize problem domains into subject matter expertise and we install specialists to work on these problems.
you're spot-on saying that the majority can't tell and probably doesn't care. but the majority also isn't as deep in this as most people here. if it's not our job as engineers or as an industry to raise raise alarms when it's justified what chances do we have - or what chances do those have who aren't skilled to ask or answer these questions?
Things have become so complex that our reaction is now to no longer question things and instead point to team-XYZ who claim that they are using it successfully (but have they really investigated what they're using or are they just so desperate to turn a blind eye to what's happening?)
I'm not willing to die on that hill but am prepared to fight this for long enough until people wake up to the problem. The point is to stall the nomalization of this behavior for long enough - until a sizable portion of specialists/subject matter experts is aware and can no longer be ignored. The #DeleteFacebook movement and people inside Google and Microsoft fighting ICE contracts are an example that pressure maybe doesn't solve the problem but it still is a very effective "spanner in the works" of Surveillance Capitalism.
from a technical pov I still wonder if running jitsi (or another similar solution) on dedicated hardware which is better tailored to a GPU intensive operation. This could then be easily deployed in-house (with all the benefits: full control and eliminating a lot of attack vectors). Seems like a cool problem to solve while in corona quarantine.
You're assuming ill intent where there is none. At the worst, it's incompetence. And they fixed the local http server flaw.
I'd much rather we reserve the term "malware" for actual malware and not dilute it to mean "any program made by a programmer who's either not very good at security or doesn't have the exact same opinion about it as me".
https://www.cnbc.com/2019/03/26/zoom-key-profit-driver-ahead...
> they actively bypass the security control on the host-system where it is installed - this is literally what malware does
it is not equal to malware. It is what malware does, which is an important distinction. If you're OK with a product disabling the host-system security controls and still happy to trust the product with this taken in consideration then fair enough: use it, defend it, and evangelize it as much as you want. As somebody who has "security" in the job title it is a problem for me.
> And they fixed the local http server flaw.
it wasn't a flaw or silly design bug, it was a conscious design decision to gain market share which other players felt too risky. please read the NPAPI spec and why it was deprecated. A company doing this has no place in an enterprise network!
> it is not equal to malware. It is what malware does, which is an important distinction.
That's not an important distinction at all. It's like saying "Ooh George talked to Mary when they were alone in the elevator. That's what rapists do", and then later defending it by saying "well, I did not say that George is a rapist".
I agree with the remainder of your comment, fair point. I think your initial comment would've been stronger if you had used the "no place in an enterprise network" argument instead of the malware comparison.
Installing an HTTP server on your client to bypass security control is not talking to Mary in an elevator. It's following Mary home, and making a copy of her house key.
As stated elsewhere, it doesn't show the same thumbnail/gallery view of non-speaking participants that you can get with the native client. I was able to share an application window, so there is some decent functionality. I don't know if that might vary with graphics stack, i.e. Intel vs NVIDIA and Wayland vs Xorg...
EDIT: I meant to include, WebRTC itself does not prohibit such things, so the point was for small stuff it works w/ minimal setup, dumb client, and for bigger stuff it would still work, but would need more robust supporting code.
For <=2 people, p2p WebRTC is used (though a STUN server might be needed to traverse NATs for each person [0]).
For >2 people, the Jitsi Videobridge is used instead of p2p [1]: it takes in all media streams from all clients (possibly even in different resolutions) and selectively forwards them to clients based on bandwidth [2]
[0]: https://en.wikipedia.org/wiki/STUN [1]: https://jitsi.org/jitsi-videobridge/ [2]: https://github.com/jitsi/jitsi-videobridge/blob/master/doc/s...