[torbital]

> but it is also not >$10B of software engineering

Really? How did you come to this conclusion?

Discord is, without a doubt, 10x better than Slack, Teams, or Zoom combined. They've solved the problem of a communication in the non-business space. That's huge.

For comparison, look at Coinbase, who has a 100B valuation. What they're doing is relatively trivial. Is that >10x the value of Discord?

I don't even think Coinbase is overvalued. 10B for Discord is just a steal.

[ALittleLight]

Suppose you have seven dev teams of ten, plus a manager, tester, and a pm for each, plus an executive team of 9 people. 100 total people on the team. They each make an average of 200k a year. How many years will it take this team to recreate Discord?

If it would take them less than 500 years, then, in some sense, Discord isn't 10 billion dollars worth of software.

I think the point is that almost all of the value of Discord comes from the network effect of people locked in to using it. On the one hand, that's kind of a trivial observation, of course the users are value comes from. But, if you don't realize this, it's important to know. It means you that if even if you build something identical to Discord you're not going to be worth anything like as much - unless you also happen to get users. Or, put another way, you could build something technically worse, but if people adopted it for some reason, you'd be worth more.

[tomnipotent]

Complexity of software is not necessarily the only important dimension of determining value.

The game industry is larger than both music and movies combined. Discord is behind almost every gaming community I've run into the last five years, from Eve Online to Valheim and Smash Bros. Looking for a Discord community is one of the first things I do when playing a new multiplayer game. Same with hundreds of millions of other gamers. You can't program your way into that kind of market adoption.

[kortilla]

Right, so it has nothing to do with the software and its value is entirely the network momentum. Reddit is a perfect example of how software can be made worse and worse while maintaining user growth.

[bluecalm]

Discord is 5 years old. The reason more and more people and projects use it over competition is that it offers better experience. It's not about some magic momentum they luckily got. I use it and moved my project to it. Last time I played a video game was about 20 years ago. Quoting one other commenter here: it's much better than Zoom, Slack and Teams combined and it's not a small margin. They delivered while everyone else failed. That's why people love it.

[kortilla]

Nope. I only finally signed up because my friends are on it for gaming. Don’t give a shit about it otherwise.

Your use case of “putting a project” on it is a minority. Most users are there for gaming communities.

[Jetrel]

Yeah, the open-source project I'm on moved all of our dev communications (†all of them) to discord, several years ago, and gladly left IRC without the dignity of a burial.

† Email and mailing lists were something we forbade from day 1. There are a variety of sociological reasons why they're vile, but I think a lot of it has to do with bikeshedding, via "low-cost involvement" - people who care more about potstirring than actually getting work done have a cheap and easy way to keep tabs on big announcement and pick fights. I've experienced a huge number of people who've earned a 'stake in the discussion' by making some small contribution, and then jumped in on every discussion trying to exercise some informal kind of https://en.wikipedia.org/wiki/Liberum_veto (insofar as such a thing arises naturally from the basic human decency of giving people the benefit of the doubt). What I like about realtime comms - especially ones with nice, logged discussions in multiple, visible rooms, is you don't have to be plugged-in 24/7 to keep up with stuff like you would on IRC (without a bouncer), but at least you have to have some real "skin in the game" of paying attention and being involved - which generally biases towards the folks doing the real work. Decisions get made without the armchair generals; occasionally they'll show up after some big decision got executed on and something was built, with an angry assertion of "Why wasn't I consulted?!?"††, but usually the existence of the actual finished work, fait accompli, tends to to stifle that nonsense real fast.

After multiple decades of dealing with that petty squabbling on email, I don't miss it at all. It was a huge source of burnout.

†† https://www.ftrain.com/wwic

[tomnipotent]

> so it has nothing to do with the software

Of course it does, because software is part of the overall customer experience that built that network momentum and is inexorably linked. Users would not have tolerated, promoted, and evangelized otherwise. It may not be the primary ingredient, but in Discord's case I think it helped.

[charcircuit]

That's his point. The main value is the userbase and not the software itself.

[xmprt]

That's about as profound as saying the value in Coke is the people buying it and not their secret formula and bottling machines.

[thaumasiotes]

You say that like that's not an interesting observation. On the contrary -- that is a common observation that people frequently make about Coke in particular. It doesn't taste good, but people buy it anyway because it's Coke.

The actual form of the saying is something like "if every Coke factory burned down, Coke would be back in business the next day. If you hit every Coke customer on the head, Coke would be out of business the next day". (The idea being that if Coke's customers all forgot they were Coke customers, Coca-Cola wouldn't be able to grow a new customer base on the strength of their product.)

[Jetrel]

Coke was a business success because they're probably the best-tasting cola when you've already drank at least half of a can.

Lots of other soda makers, especially the bargain-bin "generic store brand" stuff, taste pretty gross after a can or two, by comparison. Pepsi's alright in that department, but this is a huge market differentiator for Coke.

[ALittleLight]

I never said it was profound, in fact, I said it was obvious. I called it a "trivial observation" - but it matters if you don't realize it. The Discord software isn't being bought and isn't worth 10 billion, probably not even 1 billion, but rather the totality of what Discord is (brand, users, data, etc).

[usrusr]

That's called brand and in one way or the other, in postindustrialized societies most people seem to work in that field (e.g. everything that is funded by ads is effectively part of the supply chain).

"The value is in the user base" is just a particularly sticky form of brand, nothing else, set least as long as there are no long term contracts involved. It's not in knowing their logs, it's not in having their authentication hashes, it's not even in exclusivity (I believe that most users of anything in the wildly overlapping spectrum from chat to instance messenger to screen sharing conference call are active on quite a lot of offerings)

[tshaddox]

It’s not profound, but it’s important to point out in a conversation where people are actively denying it.

[numpad0]

As a Dr Pepper drinker, I have no doubt that that is actually literally unironically the case.

[gsich]

You don't get a userbase if you have shitty software.

[Supermancho]

SVN, CVS, Perforce, etc still exist, alongside git. Software features and quality are not the only reasons people adopt software.

[gsich]

Yes and compared to git they probably have a smaller userbase. Or a used for historic reasons.

[bluecalm]

Well, Microsoft has more engineers, money and can hire top talent. They also had a huge headstart in user base and adoption. What did they manage to come up with in a chat/collaboration space? Right - a total mess.

Discord didn't "happen to get users". It made a platform people started using despite huge network/entrenchment effect of other players because it's so much better than everything else. I've learnt about Discord following open source projects and moved mine to it as well. It's fantastic, much much better than anything else on the market. Zoom is valued at 100b. Discord for 10 is just a steal.

Coming back to your question: even if you multiply it by 10 and add infinite acquisition resources the team would most likely never be able to recreate Discord.

[ZephyrBlu]

> If it would take them less than 500 years, then, in some sense, Discord isn't 10 billion dollars worth of software

And what about Slack, Zoom or Coinbase?

Are they $27B, $100B and $100B worth of software?

What company is $x billions of dollars worth of software where x is their valuation? Even Google almost certainly isn't $1.3T (!) worth of software.

[ALittleLight]

Yeah, what I'm saying is that companies are valuable or not based on their users (or what they earn) and not their software. Purchasing companies aren't paying X billion dollars for the code and assets. They are paying for the users. (Which, I also think is kind of obvious).

[bluecalm]

As proven by among others Zoom, Whatsapp and now Discord if you build good software people will switch to it. I think you're undervaluing how good the software is. They are 5 years old, started from 0 user base and now they are a huge player and getting bigger by the day. You would have a point if there was something remotely close in quality on the market but not as popular but there just isn't. Discord is the first communication platform which doesn't suck in some obvious way. If Microsoft manage to buy it for 10B and not mess it up it will be one of their best acquisitions.

[meheleventyone]

The users without the software have no value though. If you could theoretically sell just the user base would it be worth as much?

It’s more like the user base and network effects are a multiplier of the software value.

[sjg007]

Well it can be users and/or features. A feature might give you an advantage in your bundle that you can't otherwise build.

[ZephyrBlu]

Right, in that case I completely agree with you.

[matkoniecz]

> If it would take them less than 500 years, then, in some sense, Discord isn't 10 billion dollars worth of software.

If someone wants software now, rather than in 450 years it may be worth paying a premium.

[Aeolun]

> If it would take them less than 500 years, then, in some sense, Discord isn't 10 billion dollars worth of software.

I think the problem is that this random group of people could never create Discord.

They could create a superficially similar chat application, but it’s just as likely to be worse as better.

[hu3]

You seem to think Discord can be recreated by throwing people and money at the problem.

History shows it cannot. Discord is on another league in terms of technical and usability excellence when compared to competition.

[Supermancho]

History shows that Large Companies tend to fail when recreating products because of internal politics and initiatives (outside of the core goal), not because they physically can't.

[lostmsu]

I am not sure what exactly makes Discord so attractive. Let's for the sake of the argument say it is a low-latency noise cancellation technology.

No amount of dev + test + management can get that technology in 500 or 5000 years. You need researchers, and they need to be smart or lucky enough to give you a competitive result.

[kortilla]

>I am not sure what exactly makes Discord so attractive. Let's for the sake of the argument say it is a low-latency noise cancellation technology.

Discord is not popular because the technology is advanced. It got popular earlier on because the UIs for alternatives were garbage and they nailed the ease of multiple servers/channels/etc.

The entire point is that discord could be easily reproduced from a software perspective. That won’t matter thought because the users won’t be on the new network and will have no incentive to switch.

[b0tzzzzzzman]

If Microsoft buys Discord I think several people would be looking to switch.

[xmprt]

If you've never used Discord before, then it just feels like another chat application with a "gamer" theme. In reality, it's so much more and I think startups could actually do a lot more using Discord instead of Slack/Zoom.

The UI is really simple to navigate. If you set up roles properly, it's really easy to automate certain things. There's a massive bot ecosystem. The voice, video chat, and screen share features are so much simpler to use than any other product I've used before. The thing that I find insane is that it's 100% free to use.

[wildpeaks]

I did exactly this: initially I was going to go with Slack because the bots API was nicer and they had slash commands, but in the end Discord is just so much more pleasant to use, and it even started adding slash commands now.

My one wish would be an easy way to handle multiple accounts.

[k12sosse]

Can't deny Microsoft finally nailed the single identity. It shows in teams b2b use.

[lostmsu]

I thought they collect a lot of data with desktop clients, including what apps you use and when.

[charcircuit]

There is an optional feature that let's you share what game you are currently playing to other people. It works by listing the current processes and matching against a preexisting list of games. If it finds a match it will set that as your current game assuming the setting for that is enabled.

[charcircuit]

What was attractive is that it had extremely low friction to onboard new members into a community. Click a link, type your name, and you are now a part of whatever community.

Compare that to skype where you had to download a whole piece of software, register a new account, go to your email to activate your account, click the link for the community group chat, and then you can finally start talking in a single channel.

[Bancakes]

The next best thing IMO - Element - takes a few seconds to send and deliver a message and up to 10 seconds to initiate a voice call. Has no voice sub-rooms with auto voice detection, either.

[numpad0]

It’s a free beer carbon copy of Slack except it uses single account for all instances. And they have the voice channel feature, but that’s kind of an aside.

[pizza234]

I wouldn't call it carbon copy. Slack's performance is abysmal (while Discord is relatively snappy), and this is, interestingly, toxic to Electron, as Slack is frequently used as reference to assert that Electron applications perform poorly.

[kortilla]

Discord hasn’t solved it anymore than slack, iMessage, teams, etc. The experience is more polished but it doesn’t do anything fundamentally special.

If discord shutdown tomorrow my friends would be annoyed and would just move to some other similar group chat + audio.

The value is entirely in the user base. I wouldn’t be on it if nobody was there.

> Really? How did you come to this conclusion?

Because that’s how every social network works. Here’s a thought exercise. Do you think Microsoft would offer $10B to buy Discord if they weren’t allowed to transfer over any of the existing accounts?

If it’s $10B worth of just software engineering, then that’s an easy “yes”. The realistic answer though is “no” because Discord is useless without the user-base.

[saalweachter]

Do chat apps lock in users for long?

My online circle has been on Discord for a couple of years, but they've migrated from IRC to Campfire to Hangouts (or whatever it was at the time) to Slack to Discord with maybe a few stops in between. Maybe Discord is finally perfect and we'll never move again ... but moving a community from one chat venue to another has been pretty easy so far.

[bluecalm]

Try following some projects on Discord. Maybe a small one like Leela Chess Zero. It's a completely different experience than Slack. Maybe you and your friends could be happy there but for projects with hundred/thousands of fans/developers/contributors it's in league of its own. It got it users because of it. It didn't luckily stumble into a userbase. If it disappeared today people would be sad and switch to something else if it would reappear in 2 years everyone would be switching back ASAP.

[popinman322]

Discord nailed reliability, quality, and usability. Its tech is still not worth $10B, though, even if the tech needs to be as good as it is to keep attracting users.

Tangential: I really do fear for them if they get acquired, though. Their tech is built on the actor model, which almost nobody else uses or understands enough to build on. It feels like it's likely that if they merge with Microsoft they'll get rebuilt internally (probably to use Azure) or have their infrastructure replaced with Teams' infrastructure.

[tybit]

Microsoft is surprisingly open to the actor model. They’ve had various implementations of virtual actors after publishing the research, e.g Project Orleans for .NET and Durable ~Functions~ Entities on Azure.

[verst]

Don't forget about DAPR :)

https://docs.dapr.io/developing-applications/building-blocks...

[tomnipotent]

> almost nobody else uses or understands enough to build on

Tools like Akka are very much alive and well.

> t feels like it's likely that if they merge with Microsoft they'll get rebuilt internally

I doubt that, unless they're using a lot of proprietary GCP tech. A Microsoft-owned business using Azure to (most likely) save money is kind of an obvious move, and I don't consider it a bad thing.

> their infrastructure replaced with Teams' infrastructure.

Or maybe even the other way around? Discord's tech is no slouch.

[torbital]

Is there an open source example of the actor model in use, particularly in Go or Java?

It sounds pretty cool but I can't build the mental model of it without an example.

[reducesuffering]

TIL https://en.wikipedia.org/wiki/Actor_model

[EdwinLarkin]

Does not MS employ people like Sylvan Clebsch? I am sure they employ a lot of folks that understand the actor model.

[tshaddox]

Imagine how much Microsoft would pay Discord for just a hosted version of the Discord service, or even an on-prem version that includes maintenance. Do you think they would still pay $10B for that, with none of the accounts, account data, brand reputation, etc.? Do you think they’d pay even 1% of $10B?

[bluecalm]

Well, imagine how much someone would pay for NVidia if the only thing you could do with their GPUs was to watch videos. Surely not 300B. Yet you can do so much more with the GPUs and you can do so much more with Discord then keeping it as in house tool.

[framecowbird]

Could you expand on how it's better? Genuinely curious. I love Slack and tried Discord but found the UI pretty clunky, and the lack of threads annoying, but didn't probe it much further so I'm wondering if I'm missing something...

[omnimus]

Well it has started has voice chat for gamers. Their latency and sound quality is top notch. You can have voice call with 5000 people. You can finetune sound of each individual. They have mass moderation tools... it surprises me with something in a good ways.

It feels like theyve nailed the technical part.

[IceWreck]

> You can have voice call with 5000 people

Discord with a voice call of > 50 people really starts cracking about and it just isn't usable. audio quality goes way down.

Not sure how you got the 5000 number.

[whatbutwhy]

I've personally been in voice calls on discord with thousands of people. Obviously it's impossible to pick out anything discernable with that many people but discord handles it well.

[torbital]

FWIW, I probably hate Slack more than anyone else on Hackernews (because everyone somehow seems to like it). Keep in mind I also don't use the Slack desktop app but I do use the Discord desktop app, so right away this isn't a fair comparison but I'll continue with my rant anyway.

I mainly just find the user-experience to be terrible. I hate how each server is separated as if it is it's own isolated website or something. On Discord, I can switch from server to server in a click easily.

Playing GTA? I can join the GTA Discord server. Playing League? I can join that one. It's all right there.

It's also the weird SSO Slack has going on that just seems broken. When I open up Discord, I have one account, I log in and have access to all other things in the account. Basic software behavior we're all used to.

When I open up Slack, it does or does not want me to login again based on if I've been on this server before. It may or may not want me to make a new account depending on if I've made one on this server before. I try all of my emails with all of my potential passwords to see if login works, if not new account time I guess? I'm never sure weather to use my work account or not. Sometimes the @domain matters, so this creates even more of a mess.

Then don't even get me started with a lack of usernames and having intentionally hard-to-find ID. Again, why not do what Discord does and do Username#6789? Instead I have to search my friend 'John Doe' and hope it's the John I'm thinking of and not some random other John that I just invited to see all my business's internal discussions.

[quantified]

It’s not only better than Teams or Slack, but it has a great gamer community. Connect that with XBox, maybe it’s a monetary win.

[ewindal]

Discord will forever be behind Slack in usability as long as it doesn’t have threads. Nothing quite like having three different conversations going on at once in the same channel.

[amelius]

The value is in product design, branding and userbase, not so much in software engineering.