Can You See Me Now? A measurement study of Zoom, Webex, and Meet
arxiv.org
arxiv.org
When you're hopping from one call to the other, having the ability to quickly drop and join the next one [EDIT: with audio from my phone] is a big win
That, the global Alt+A / Alt+V for the audio / video mute toggle respectively, and the "hold spacebar to talk" hotkeys make it the clear winner from an UX perspective IMHO, even if that X isn't really perfect
At the end of the day, most calls are virtually lag free so I care more about my ability to efficiently interface with the app than I do about a perceived streaming quality [EDIT: except, of course, for those participants who don't join the audio via their phone, as their voices inevitably suffer from some lag and distortion]
This sounds inconvenient? For Meet, I click the invite link from my calendar, email, or meeting list for an upcoming meeting list, or from the slack or other message a coworker sent me for the adhoc one, and I'm in, with video and audio going through the one device.
Dialing in from your phone increases the audio quality for everyone by a factor of approximately ∞
After all, the listener is receiving over VoIP, so you've got phone network issues layered on voip issues.
Unless there’s an enterprise bandwidth constraint.
Maybe it was user error, but it felt much slower to get everything up and running
If you don't have the app, joining via phone is required, finding the number is a few seconds including passcodes and then sometimes passwords which often elude me in terms of their location.
Huge whereby/appear.in fan, used to run our family wknd catch-ups on there.
Former webexer here. That depends on many criteria, sessions can go through multiple locations depending on where the customer is located, the features being used, the type of customer, etc. Since Webex also does transcoding (useful when a 4k wall system and a cell phone are on a same meeting), that can influence where the traffic flows.
Webex, Zoom and Google also all have their own backbones to reduce the number of hops over the public internet.
In this case the researchers appear to be based in New Jersey so they would be more likely to enter the internal network via the eastern POPs.
> We find that the existing videoconferencing systems vary in terms of geographic scope, which in turns determines streaming lag experienced by users. We also observe that streaming rate can change under different conditions (e.g., number of users in a session, mobile device status, etc), which affects user-perceived streaming quality. Beyond these findings, our measurement methodology can enable reproducible benchmark analysis for any types of comparative or longitudinal study on available videoconferencing systems.
Seems like a very useful study at first glance!
Surprise, the somewhat primitive Webex offered the best experience in that scenario and vendors who wanted to meet were inclined to accommodate us.
Let's say you get a new server. It needs several teams to set it up - storage to provision/rescan/configure disk, infosec to do domain/antivirus/dlp work, db team to get the database going. They're going to share an rdp session over webex and hand it off to each other to type commands, with everyone observing and catching mistakes. Yes, it's possible to give desktop control to someone on zoom, however that is a secondary afterthought of zoom, while for webex it's the primary function. Right tool for the job - don't transport a piano in a sports car then complain the sports car is.. not great.
1. The reaction buttons don’t indicate that anything is actually happening for you when you click them, you only can see others reactions. They’re also quite hidden when they come, make no noise so easy to miss, and don’t tie to the video.
2. The “people joined” ding is very non-intelligent. It used to be when 30+ people were queued for something to start, once it did it just dinged like crazy. Now it seems work has turned that off in a way I cannot turn back on, which means when people are late for a 1:1 they’re staring at me and I have no idea since I now have people blocking. A heuristic that would be aware of # participants, when people are joining, and batching up dings I cannot imagine would be groundbreaking research when this is your product you’re building.
3. The actual controls for video on mobile to switch camera is quite bad.
4. Screen sharing works _ok_, except it has no integration with the resolutions on both sides, Apple’s accessibility-based zoom, I’m given a red dot for a cursor but it doesn’t appear that way on the other side, and cursor tracking is extremely laggy and jumpy on the remote side despite my red cursor that’s giving me the impression of a laser pointer equivalent.
5. When someone is screen sharing, the controls for what to do with the video of others are not very good. Order is non-deterministic (afaik), not easily changed, provides few layouts, and not easily hidden.
6. When screen sharing I’m given a large list of all my windows to choose from. They’re in effectively a random order (not alphabetical, not by z-index on my Mac, etc), and the mini photo of them are usually unintelligible and difficult to pin point the one I want.
These are just some of the bad UX that I encounter daily in work usage. I could probably come up with a list of 10+ more issues without even thinking much about it, and we haven’t even touched the “interface” level issues (eg actual design).
I had to un-mute and then re-mute my audio to continue.
I know I shouldn't have relied on a software switch when I have a perfectly good hardware disable button, but WebEx embarrassed me in front of dozens of my peers, and I will keep that in mind whenever I get to choose what platform a meeting goes on.
Teams, by contrast, is buggy and inconsistent and weird and has the awful arbitrary rectangle subdividing of the chat window instead of just showing me the other people's full cameras.
But it’s not a surprise in this scenario since WebEx is the only one not hobbled, no?
I’ve used all 3 and WebEx is second to Zoom but better than Skype.