All the integrations, and workflows that Slack has enabled for our company make us more productive.
I would hate to go back to what came before it.
IRC is complete trash and I don't know why anyone would use it in the same sentence as Slack. UX matters. It matters A LOT.
The nice thing about IRC is that everyone can use whatever client they like and it will just work. I preferred command-line UX (irssi) because most of my job is based around the terminal, others prefer desktop clients, while others prefer web clients, and all were available for our IRC server.
Our problem was that management wanted everyone on the same thing, and neither side (IRC vs Google chat) wanted to switch (they liked integration with web Gmail, we liked extensibility, and had already built useful plugins, like "run a build"). In the end, they didn't like our "less professional" plugins (Chuck Norris facts, gifs, etc), and we ended up compromising on Slack (we rebuilt some of our custom integrations).
We went through a year or so of adjustment, and now we're reasonably productive with it. However, morale is a bit lower, and I think we're a bit less productive than before, but that's really hard to judge. Honestly, I just wish there was some nicer looking IRC client. I don't like trusting Slack and would prefer the tried and true protocol that we can tweak and just pay for a flashier client for those who like such things. However, it's better than the previous situation where teams just didn't communicate in text because they couldn't agree on a medium, which led to more interruptions than the annoying Slack notifications (non developers tend to always @ the tech lead instead of letting it be answered by whomever happens to be looking at the channel).
I think it's wrong to blame IRC here. The UX of IRC is fine, it just doesn't have a flashy client (pidgin is really easy to use and available everywhere, but it's kind of ugly).
What's wrong with the UX of IRC?
* Poor abuse tools, e.g. accidentally IP-banning web IRC gateways.
* No searchable history
* No scrollback / visibility of messages that arrived while you were offline
* No support for uploading images or files.
* No sharing of a single identity / session across multiple clients (e.g. phone+pc+laptop)
As a veteran IRC user I know you can mitigate these problems by getting a shell account on an always-on unix server and running Irssi in Screen over SSH with logging, automated NickServ login, a bot to provide !seen and !tell, sharing images via Imgur, sharing files by uploading them to my personal website, and so on.
Slack has simply looked at the big collection of bodges I just described and made them native features. And you don't even need to know to use Ctrl+Alt+2 to change between channels like you do with ssh+screen+irssi.
One sentence later:
> ...we scrapped it because only our software developer used it and other people in the company preferred something else (they used a mix of email, Google chat/hangouts, and text).
Clearly something was wrong with the UX if only one person was willing to use it?
This place is a parody of itself sometimes.
And the only company out there that was able to stitch together more or less usable UX for a group messenger is Slack. IRC has no such killer app (sorry IRC fans!..)
I can search my logs easily https://i.imgur.com/nTpoYs0.png (soon with even better integration, the prototype is a separate web app), and I can just open the web client or app on any device on this planet and instantly have all my logs and channels, and just chat: https://i.k8r.eu/54BPMQ.png
Of course, I use self-hosted quassel and quasseldroid.
Many other projects such as weechat and IRCCloud are also working on making IRC better than slack at everything it does — IRCCloud has a web client, mobile client, hosted team servers, a slack bridge, reactions, threads, emoji, attachments, and all you'd want from slack, natively.
Other projects such as IRC.com are also working on bringing the feq advantages Slack has back to IRC to make it more attractive.
The one single real advantage of Slack is marketing. They can call companies and spend a lot on money on getting more customers, due to the VC funding.
Respectfully, NO.
Providing a very useful turnkey set of features out of the box that are technically possible but fiddly and hard to use in IRC is in and of itself a huge advantage. This is why they're successful. This is why people use it. This is why people use it who barely know what a command line IS.
IRCCloud can even connect to Slack workspaces.
The only advantage Slack has over IRCCloud is marketing.
It's relatively ironic that I have to end up praising IRCCloud here, considering I'm contributing to a competing IRC client.
That makes it much closer to the open ideal than Slack is.
That sounds exactly like the infamous first comments on the HN Dropbox post. I'd rather pay someone to use their app than go through the pain of self-hosting and administering yet another service.
Quassel on the other hand is to Slack what Seafile or NextCloud are to Dropbox — definitely not as comfortable, as you need some knowledge to self host it, but it's easily doable for everyone on this site, and once it's set up, it's just as comfortable.
Of course, if one wants to use a hosted service, there are solutions like that — IRCCloud is pretty much identical to Slack, except based on IRC, and in addition to paid team workspaces, you can also connect to any IRC network and use the same features there.
So, you see, while it has some hints of the same attitude as the original post — and definitely, neither of these products is as polished as Slack, after all we can't make 380k$ losses a day, while Slack can — they fulfil pretty much the same purpose with pretty much the same ease of use
And of course work on single-click deployments and hosted services for more such products has already started
And that's where you lost sight of the big picture. Slack is _not_ for people on this site it's specifically for people out there: people who don't know and don't need to know what a protocol is, who don't care about netsplits, who couldn't be bothered to install some random stupid stuff on their machine and spend an afternoon configuring it just to be able to talk to their co-worker. We're not the target here. And yet it has convinced some of us because it ticks so many boxes by default.
I realize I'm feeling (objectively?) "defensive" as an everyday professional IRC user for the last 10-ish years. That said, having tried many of the recent obese chat tools (which do have some nice properties), I still maintain the stance that IRC's clutter-free nature brings a certain tranquility and clarity.
Sure, IRC absolutely has its problems. However, as noted elsewhere[1], despite its flaws, IRC's strengths still shine with great luminence, when it comes to plain text-based communication.
[1] https://news.ycombinator.com/item?id=20119407
- - -
It reminds of something I recently read in an essay called Coon Tree (written in 1956) by the inimitable E.B. White:
"Many of the commonest assumptions, it seems to me, are arbitrary ones: that the new is better than the old, the untried superior to the tried, the complex more advantageous than the simple, the fast quicker than the slow, the big greater than the small, and the world is remodeled by Man the Architect functionally sounder and more agreeable than the world as it was before he changed everything to suit his vogues and his conniptions."
I didn't quote the surrounding context (which itself is quite enjoyable) for brevity. But do read the full essay if you can.
</digression>
UX isn't just graphical design, it's also feeling responsive, fast, and accessible. Something that Slack, in my opinion, gets horribly wrong.
Honestly, I can't find fault with it. It's such a basic and uncomplicated piece of software, I'm not sure how they could get anything wrong.
A lot of discussion from this post from yesterday: https://news.ycombinator.com/item?id=20227115
Slack is certainly no email replacement, and nor will I ever use it for such but in my experience and for the purposes I use it for, it is far, far more featured than IRC, unless you're using some uber IRC client that I'm unaware of.
I'm genuinely trying to understand out how anyone would think IRC is better than Slack and I speak as someone who used IRC from the early 90s up until only a few years ago when the communities I'm part of upped sticks and moved over to Slack.
However, for me personally, a major attraction of IRC is that I can easily use it from within Emacs, and automatically have the same Emacs key bindings available that I also use for navigating and editing files.
With IRC, I can therefore quickly copy and paste text between chats and programs, search the history etc., all in the existing Emacs session, using the same key bindings, with direct access to built-in features such as autocompletion, dynamic abbreviation and spell checking, and — importantly! — without switching applications.
We all have our own ways of doing things and I personally prefer to have applications run independently so I like my Slack client and my IDE and terminal etc. etc. separate.
I've always found if one tool can do it all, it's generally a worse experience than dedicated software applications and doing a Cmd-TAB (I'm on a Mac) is just as simple as switching to another window in a single application
Personally: responsiveness. Slack's native app on linux is slow. Switching channels has a noticeable delay, switching Workspaces takes seconds if you haven't used the workspace for a while. Starting the program takes ~8s on my machine (three workspaces, i7 with SSDs and 32G RAM).
For most IRC clients or Telegram, startup times are about as fast as Slack changes a channel, and there are no noticable delays when switching channels/IMs etc.
On both, it takes seconds to boot, and that's just to load the local cache. Syncing with the main server takes a few more seconds.
It has some good UI patterns. But overall, it's heavy and slow. It takes longer to load than the previous chat clients I used.
We'll talk about it 10 years from now when Slack is gonna be "old cruft" while IRC still gonna be there rocking.
IRC has probably 95% of the features slack has, is completely free and has been around since 1988. Seems like a logical comparison to make.
And we are comparing a service provider/client/protocol to something that is simply a protocol which is a bit apples to oranges.
Assuming the same kind of provider exists in your company (IRC+BNC+searchable logging) and you standardised on something like weechat, then perhaps a comparison could be made.
Slack Pros:
- Voice/Video chat
- slackbot "remind" notifications
- Number of highly polished integrations
- Rich integrations (button prompts etc)
- Webhooks for bot integrations instead of needing to subscribe to RTM
- User tags ("out of office" markers, "in meeting" markers)
IRC Pros:
- Your data doesn't have to leave your site
- You can force connections to be on your intra-net
- Simple, well understood protocol
- Which is open, and not going away
- More potential to integrate authentication with your normal authentication provider.
- Rich moderation story
- Flexibility of client, even if we take weechat as the standard it's still possible to run another.
Is this fair? Maybe we should weight how much they're worth.
But for sure if IRC had the investment slack had then it wouldn't be "trash", there's work to bring it into 2019 too[0]
[0]: https://ircv3.net
One massive one being keeping history. Another is not being limited to text only, something people absolutely expect in this day and age.
To me, that sounds exactly like someone defending AppleTalk, AOL, ICQ, or Minitel. Focus on usability, not interoperability. That's what people want, right?
When it comes to communications, what matters even more than user experience, in the long run, is being compatible with everyone, even your competitors. There will always be customers who don't use your system, so lacking compatibility or federation will be a thorn in your side for as long as you exist. I've run out of fingers to count the incompatible proprietary chat systems I've seen die off.
Even Google has killed off a few of their own, and Hangouts is next. Just being a young smart agile tech company does not make you immune.
In 10 years, I bet people will be communicating over email, IRC, POTS/SMS, plus 4 or 5 new tech startups created by people who are in elementary school right now. Nobody's arguing that $(proprietary_system) doesn't make you more productive today (that's the point!), but they always have a sell-by date.
Yes. That's exactly what they want.
debatable. It is just a chat app after all. Don't get me wrong, it is alright, but I am not sure I would ever call a chatapp "great"
> The valuation makes sense
16billion? To me it doesn't. It is just a chat app. People and businesses can move off it when the next chat app comes along with a new UI