Fifty percent of Facebook Messenger’s total voice traffic comes from Cambodia
restofworld.org
restofworld.org
The problem is as the article mentions. Difficulty searching back, not being able to use Google translate (I'm still learning Spanish), and the time it takes to listen to them. Sending a voice message is faster than typing for the sender but much slower than reading for the recipient which makes it especially annoying in group chats where you end up forcing this extra time on all participants. For this reason I don't think it's very social.
And then there's the issue of listening to them in public, having to dig around for headphones if you don't want to bother your surroundings.
So in general if someone audio messages me I just ignore it until they type their message.
In most languages swipe typing is not a lot slower than talking anyway. And if people really want to talk they can use speech recognition.
I run all my chats through my own matrix server (with bridges) and I was actually thinking of making something that automatically tries to transcribe them. The problem though is that I speak 3 languages :)
I'd be very interested in something like this! Also deal with 3 different languages on a daily basis in different group chats, and many people use voice messages. If you get started I'd for sure contribute to your effort, as having to change from "reading" to "listening" so often really sucks and would save me a lot of time if I could have it transcribed.
Part of the problem is that I'd like to use a service that isn't too privacy invasive. Unfortunately I doubt there is one that's good and doesn't do that :)
all apps with voice messages I happened to use support playback through voice speaker, if you put the phone to the ear as if you were just speaking with someone
but I don't really remember how I discovered that and it seems that majority of users don't even know about this feature
With a lot of (cheaper) phones the speaker is also terrible enough that the different modes don't make much of a difference.
Not really, I disagree with "not a lot slower". It is much slower and much more frustrating than talking when you have to redraw a word for the fifth time and the keyboard insists on "Chegamiknit" when you want to write "Champagne".
Anyway, it's not an all or nothing thing. Sometimes it's going to be faster, more convenient and appropriate to send an audio message for both parties because tone and articulation convey meaning that text does not. edit: and when you know the other party can act on it. Eg: they can listen and can't talk back but can write back, or the inverse and you have to send them text messages because they can't listen (rushing to catch a metro or to their car) but they can read and they can talk back but can't write fast enough (because rushing to catch a metro or to their car) (yeah, it happened :). So, use when appropriate ?
I also suspect voice messages have different usage across cultures and subcultures and group of peers and the context.
It's also a different way to be with the other person, sharing an audio space. Of course there are limits.
Dont use autocompletion then. You can abreviate or shorten some words too. Champgn and so on.
That just negates the swipe typing argument from parent for text not being much slower than talking.
This should be the norm for everyone.
I just refuse to listen to any and don't receive them any more.
Yes, there are valid use-cases for one-way voice messages. There just aren't many of those.
This is exactly what I wanted to say in my post, but perfectly phrased and to the point. I'll remember this one if it comes up again, thanks!!
Given that many people (including GP and me) prefer document content (not "text" because you can put interactive tools and diagrams in documents, but not text) and it has all of those advantages, it seems like it's pretty clearly the superior medium.
Plus, human brains are notoriously bad at introspection, so I wouldn't be surprised if you don't actually learn faster from videos, and just prefer them because they're e.g. more engaging.
My work has been trying to promote e-learning with ted talk-like videos and I absolutely hate it. It's not just about the speed.
I can learn so much quicker with text. I can skim through the parts I already know and spend more time on the parts I need to carefully consider. With video I need to skip around and it's hard to keep track of what's being discussed then.
I think it's really the younger people in the organisation that ask for it because they're used to youtubes. I rarely watch youtube, probably once a month or so. Makes sense but they try to push it on everyone by setting a goal of so many videos to do.
However even when I was young I thought that classroom teaching was inefficient and I could learn much better myself. One of the problems (also with video) is that you have to slow to the speed of the slowest participant.
It correlates with the wealth in a country. An-alphabets will prefer voice chat.
It would seem obvious for the transcription and the sound file to be available to both sender and receiver of the message.
I think so far the sticking point is that message transcription is still hard to do with any accuracy on-device, and if you use a server then you can't claim e2e encryption.
Voice messages are also useful when writing is not physically viable (ie. while cooking). I even tend to prefer it when I want to develop an idea while delivering the message and have the message development process itself registered as part of what I'm sending.
It's definitely a much worse experience for the recipient, I think new social norms of asking first are probably the best solution. I wouldn't ring someone who wasn't expecting a call without asking them if they were free to call first for example, and I think the same should apply to voice messages.
If I can't privately listen right now, I don't, and if they get annoyed, I'll tell them why.
If something would require me to replay it multiple times to take notes, depending on the situation I'll tell the sender to type it up.
It isn't a matter of trying to punish, it is negotiating preferences. Send me something I have to interact with, and it may well take me some time to get to it. This is no different than emailing me a PDF.
True, the latter is definitely a good change. I noticed it in work too, people don't call without asking first. It's really a great norm because you can say "just give me 5 minutes to complete this" when you're concentrating.
Of course it was never a norm before because text messaging didn't exist but it makes sense.
It would be great if people started doing that too for voice messages.
- Text messages require brevity.
- Depending on who you're talking to, Voice messages ramble on and on, without getting to the point. As a result, it can take minutes to understand a main point, rather than seconds as in the case of text message.
Just learned via another comment that playback can apparently be sped up to 1.5x or 2x. That's good, I'll have to look into that.
What I still highly dislike are voice messages without prior context or prompt. Eg. just receiving a message out of nowhere, not having any idea what it is about, having to dig around for headphones just to find out, that it says: "hello! how's it going?"
On the other hand, if I ask a person a question that can't be answered in just a sentence, I'm perfectly fine with the other person quickly recording a 1 min voice message instead of having to type it out.
Also, WhatsApp and Signal allow you to play the voice note at faster speeds, like 1.5x.
There is no real standard romanisation similar to Pinyin for Chinese. The UN tried creating one in the 90s, and there are a couple of homegrown alternatives, but as mentioned in the article, most people just kind of make up the spelling and hope the recipient can figure it out. Until quite recently lot of people still had 'dumb' phones sending character limited sms, so spelling has been 'creative' (I stopped carrying a dumb phone as a daily driver probably around 2017-18, so really quite recent).
Basically as the article says, its a huge pain in the ass to type in Khmer, especially on a phone. If that crack team of overpaid consultants had jumped on the phone to anyone in Cambodia, they could have explained why voice messaging is so prevalent.
I’m immediately skeptical of this for two reasons:
- My kids’ school announcement voice messages in the last year started combing from phone numbers in Cambodia. I live in the US. I just assume there is some extra cheap voice service there, and a strong advertising campaign.
- Pure speculation, but even if Cambodia’s voice message use is 100x higher than the rest of the planet, it would still be hard for Cambodians to reach 50% of the traffic for this global app. I could be dead wrong, but this feels unlikely to me, especially in combination with the above. (Population of Cambodia: ~16.7 million. Population of earth: ~7.75 billion.)
That's highly person-dependent. Some kind of auditory processing disorder runs in my family, and I definitely have it. Under a lot of conditions, I need to spend extra time to digest and understand what has been said.
But not everyone is like this. For a lot of people, hearing is digesting, is understanding.
And like many things, people presume that their own experience is the universal experience- you presume everyone needs to digest, and the person sending you the audio message presumes everyone can process it instantly.
>You can read fast enough to churn out information, but you need to wait for every second to digest the information in a voice call. This can happen, but I can reread the message or just parts of it fairly easy. With voice messages it's getting annoying when you need to rewind to a specific part of it, especially if the message is 2-3 minutes long.
And there is also the audio quality and background noise, but it's a different topic.
I agree you can read faster than listen, but I just don’t buy that typing is faster than talking though, not for conversation. When you talk, the speaking and the listening are overlapping. Sentences are generally understood before the speaker even finishes saying them. That’s doesn’t happen with text chat. When you type into chat, the communication is serialized. There are reasons a group sitting together might use devices instead of talking, but I find it hard to believe that efficiency of speech is really a primary reason, unless you’re talking about sharing media like sources, images & movies, quotes, references, code, etc. for a group project.
Also, as a language learner I can write 我是美国人。 But can I speak it properly ? Being able to send voice messages is a great option
26 to 74 is just double, and most likely some are more common in written communication than others.
It's closer to triple, isn't it?
It depends on how you define "character", I guess. The hirigana or katakana "alphabets" have 48 characters. These alphabets map best to the western use of characters.
Kanji is probably what you are thinking of, where each word in Japanese gets its own "character". There are ~50,000 different kanji symbols.
Wonder why that did not happened in Cambodia.
Also older people dont always know english characters very well. Some do, or the might know some french so they can figure it out, but not everyone
Also it sounds so strange that facebook does not have any Khmer speakers who could help them understand what is going on in the country. How do they translate the site or sell ads there? Or they simply dont, since the market is too small?
Historically, FB used to let users translate the site (presumably they don't do this any more).
And yeah, Cambodia is not a large ad market (not enough rich people) so it's unlikely that they have a sales team based there.
Geez, if I had a social network and no morals and no fears of getting caught (or the fines are cheaper than the possible profits), I'd quietly try to sell "eyeball access" to high net-worth invidivuals, who I can easily find in the database. Although who'd want this access, aren't Facebook ads just mostly scams nowadays?
(I just realize I get a lot of ads on Instagram, and most of them look legit enough)
Sell your stocks. that's a dead, bloated company.
A plane ticket and asking people on the street would probably cost less than their "study"..
Also it sounds so strange that facebook does not have any Khmer speakers who could help them understand what is going on in the country. How do they translate the site or sell ads there? Or they simply dont, since the market is too small?