This is most likely a bulkhead acquisition to anticipate the growth of Discord and to try and keep those same streamers in the fold.
Twitch gets the massive community reach of Curse, plus they also get all the current momentum Curse has to be a Discord competitor.
It's a fantastic acquisition, if a little defensive.
Only thing I can think of.
Discord has more easily manageable chat roles, eg: Moderator, Helper, Regular, Supporter, etc. Twitch chat is nothing more than simple IRC, and relies almost entirely on bots for the more complicated stuff. Discord also has voice chat built in, whereas Twitch has none for their chatters.
Discord also has a better documented API and libraries in a lot of languages. Overall, Discord (despite using an electron app...) is simply a better chatting product.
Discord is a really well-put-together app, has text and voice chat that do their jobs, and get out of the way. The only complaint I have is that there's little to no support for searching history, but for an app focusing on live "community" chats, that seems a reasonable compromise.
The Discord developers are working on this. It's not just web components but also native dependencies as well, so it's understandable I think if the Linux version comes in a little bit behind macOS and Windows versions where the majority of the gamer market is.
Next up, I'm starting to like when companies use desktop HTML5 apps. HTML and JavaScript does enable developers to move quickly with their interfaces, and I find them to be generally much more well-kept than native counterparts, where there's more inertia in making interface changes. A good comparison gamers will be familiar with is Steam. The Steam UI is written in a valve-proprietary, CSS-like language that I imagine is less performant than CSS. It's also REALLY buggy, so much so that I'm surprised that Valve has not already moved to an Electron app for Steam. If they did they'd be able to iterate quickly on interface improvements which they certainly don't do now.
Discord works in a browser or with the native apps. Just visit the website and click 'Open Discord in your browser'
We have been completely happy with the switch, in special the ability to hop into voice chat with remote coworkers with one mouse click.
It looks like we will be able to use Discord when you add search. Seriously, tell the management to think about it being an alternative to Slack. We are a paying Slack customer and while they are better than hipchat/IRC, it is far from perfect.
Our top "wanted" features right now are video chat/screenshare, so that we can drop Hangouts entirely, and search. We wish we could pay you guys. Seriously.
Better documented APIs and libraries than IRC? That’s not just quite a stretch, it’s literally a lie.
There’s IRC libs in every language you can imagine, usually dozens per language, with many perfectly documented ones, for every possible purpose.
And about all of them work for Twitch chat.
Curse I'm admittedly more than slightly biased on, since I used it basically from its inception back in the World of Warcraft days, because that's all it was initially
Curse was the least awful way to manage a bucket load of WoW addons, almost like how there are package managers for Linux today. It simply provided some central place to manage and keep mods updated, especially ones that would break whenever Blizzard patched it
Curse at least, slowly evolved into this hydra-like monster, trying to split itself off into as many different things as it could. The most insidious one I remember was when I reinstalled it (for trying out the newest World of Warcraft expansion at the time) it'd evolved far beyond what I expected
Upon relaunching my system it prompted me to add literally every friend I had across Skype, Steam and League of Legends. While also overlaying its voice chat system when playing certain games (League and TF2 come to mind)
Keep in mind I never used any of these features, it was simply new garbage that Curse added to try and stay "relevant"
If Discord becomes a critical component to the content creators of the whole gaming ecosystem, there's going to be plenty of revenue opportunities.
Discord promises it will always be free with "no gotchas". It specifically notes no "slot limitations" as an example of that, which indicates premium tiers are unlikely. They promise to never charge for core functionality. Presumably they may be able to do some sponsored promotion of some sort, but banner ads are probably out.
The revenue plan they mention is here: "In the future there will be optional cosmetics like themes, sticker packs, and sound packs available for purchase." And I have not ever seen an emoji pack for a chat service as an appealing sales item that's going to fund a startup. I'm super skeptical of this, and it's why I don't use it.
But, it'd make a solid acquisition for a company like Twitch that has a shoddy/simple chat function and needs something better in the gaming space.
If Discord can keep up the growth, things like stickers, emojis, and sound packs could pull in good revenue. Also, the gaming community is already used to paying for these items on sites like Twitch
Everyone likes to keep coming up with new names for the word "emote".
But I have a "global twitch emote" plugin on chrome so I presume kappa will already show up for me if used in discord?
Granted, a lot of people will run the electron-based desktop app (which packages the web site), but IIRC, there's a plugin for twitch emotes for it too.
You underestimate gamers as a market. Valve's built a thriving business on virtual hats. Twitch sells access to premium emotes and a wider range of username colors as a subscription upgrade. "Cosmetic purchases" are huge.
Discord could also do what Twitch does and let streamers offer recurring donation options to their fanbases, in exchange for perks like being able to talk during restricted times (eg. during large streams) and custom community-specific stickers.
My thinking is that hats aren't that profitable without these kinds of tricks, and Steam effectively sponsored everything that happened at Valve before they implemented "gambling".
And if you refer to crates of random items as gambling, they've also filled in that gap with the Steam Marketplace where you can buy individual items for market-dictated prices from other Steam users, where Valve takes a portion of the money the items are sold for.
This business is extremely profitable. Looking at their game Dota 2 for example where they recently hosted the world's largest e-sports tournament, where the pricepool were completely funded by the players of Dota 2 through in-game virtual items. This pricepool were over $20,000,000 USD, funded through in-game purchases where 25% of the items price would go to the pricepool, and the rest would go to Valve. That is $60,000,000 USD that Valve profited in the span of two months, for one of their games, in one of their tournaments. Valve hosts 4 tournaments per year for this game.
Looking at the pricepool of their previous tournaments ( http://www.esportsearnings.com/tournaments ), we can see that this amount just keeps increasing year after year.
# Crates
Crates or doubling, tripling, etc. item's cost. Say I want a certain hat, on average, I'll need to pay multiple times (crate contains 5 other items, but successive crates do not yield duplicates in most recent Dota 2 versions, so worst case you pay 6 times a price).
# Steam Marketplace
Adds at least 30% to item's cost, since you buy with real-world dollars, but sell for "Valve Dollars" (only usable in Steam, no withdrawals), this is effectively paying more for a single item (at least 30% more expensive which is Valve's margin for Steam purchases of non-Valve titles).
In addition, Valve makes purchases non-tradable via Marketplace for a period of time (multiple months, usually) since original purchases for "preventing fraud". I see it as "incentivising gambling with crates", though.
# Third-party site
These were probably shut down because they involved real money, so Vavle could be potentially helping organise casino/bookmaking businesses from legal standpoint. It's all fine with Steam Money from marketplace.
So, yes, it is extremely profitable, probably orders of magnitude more profitable than "honest" selling where transaction is "pay money, receive item".
# Regarding TI
It is incredible how much money it made. It is as incredible as how Valve decided to split matchmaking pool for months with International Ranked to boost prize pool, so you had to pay $10 to go into faster queue.
In the beginning of Dota 2, you were able to buy item directly, now you either need to go to Commumity Market, or buy crates from Valve. Check out dota2.com/store — there is either half a year old sets, or crates for new ones. Even older items are not in the shop and I believe w/ever you unlock from crates is not tradable for quite some time.
Correct me if I'm wrong, but there seems to be no way to buy particular items directly, at least not newer ones. (Exceptions are $30+ arcana quality sets which bring a lot of money as they are, so no need to increase price with "gambling".)
Amazon is another company with opportunities to make profit, but they traditionally haven't done much of that, eschewing that for just growing value.
In the years when Amazon was not turning profit, would you consider it a non-functional business?
Edit: I really hope their isn't a consolidation of gaming communities like Twitch is trying to do because it's only going to pander to the lowest common denominator, which in gaming tends to be absolutely cringeworthy.
From what I can tell, streamers have no issue streaming on both twitch and youtube, and it mostly comes down to content restrictions as to which they use as their primary (twitch is fairly strict about streamers doing something 'creative', I think to avoid cam girl like activity).
Are you able to say why you think Curse is terrible?
That's a bit of an overstatement. It's very slow and, like any ClickOnce application, is a pain to get working under Wine.
Searching around there were apparently better tools, but Curse shut down their access. Imagine if GitHub were to stuff their desktop client full of ads and then require you to use that instead of the Git CLI.
And if they do merge it with Curse Voice (as they claim to plan to) then I expect it will stop working entirely...
For syncing/backups you can just symlink Interface/Addons and WTF to somewhere in Dropbox/Owncloud, no need for a specific application (and subscription) for that.
Full disclosure: A few years ago I published an add-on on Curse and apparently I still have free premium ~80% of the year after that. I still use their client because they're the only game in town these days, sadly.
I use curse mostly to handle addons for games I play a lot (world of warcraft primarily). Every now and then, when I bump into some terrible interface on their client or bad practice by their business managers, I have wondered how much work it would be to move to an open addon publication system. If you get all the addon authors to move to a different storage platform for their actual addons like github, how much extra work would it be to build a client on top of that? The addons themselves are pretty much FOSS by definition - you can't distribute for-pay addons for WoW at least.
I imagine there'd be some scale issues, and github isn't really a distro network, but I'd bet the technical work could be done in a weekend.
It's not quite as automated as publishing a NPM or Composer package, but it's close and works well.
The precursor system to Curse was actually originally a shared SVN server a group of addon authors (called Wowace; still around somewhat) were using to collaborate on addon development. SVN checkouts were so convenient for distributing addons that the authors started using it to keep their own addons up to date, and then usage started to rapidly spread among "ordinary users".
Unfortunately, the traffic every time a new patch came out absolutely smashed the server and generated enormous bandwidth bills. Various attempts to fund it via donations or sponsorship went nowhere.
Curse had been knocking around in various forms for a while (I think they were a guild at one point, and already had a failed pivot to a "gaming community" behind them), but they rose to prominence by, basically, taking over the SVN server and finding ways to keep the lights on.
Switching to github would be coming full circle in many ways. And it would have some advantages, but I'm not sure how happy github would be to act as a distribution network for addons.
Note that 1.1 million clones and fetches in a week for the CocoaPods repo caused major strain: http://thenextweb.com/dd/2016/03/08/cocoapods-popular-basica...
Using github as the backend for a Curse-clone would probably exceed that handily.
As for Discord... my guild leader switched our raid team from Teamspeak and there was some acrimony because a guild member refused to install Discord as he felt strongly that it was a security threat and malware.
I did try to substantiate these claims but couldn't find proof one way or the other.
As a precaution I've resized my hard drive and installed a separate Windows for gaming and encrypted each partion separately to isolate Discord from my personal files.
Several weeks into using Discord I think it is better than Teamspeak for overall features but in a raid situation with 20 on simultaneously the quality isn't as good as paying for a private Teamspeak server. Having said that though Discord is free and allows for a much richer experience so if they can sort out voice reliability and quality it will be a compelling proposition.
I still have concerns as to the security of Discord as their website pretty much brushes over the topic but they do offer two factor for what it's worth... but I can't raid with my guild without it so I'm stuck using it.
(I assume you don't do this for every application you install)
a) vendors I trust b) confident in their security, and c) couldn't write the software myself.
In the case of the Curse client I just wrote my own batch scripts and used cygwin. I could have just as easily done it in Python or AutoIt.
As I wrote the software I know there is virtually no chance for malware.
I can't be as certain about the Curse client. As far as I know it isn't open source.
Maybe I'm more paranoid than most about security but it could be because I know how easily most systems are compromised.