I must be in the minority (based on the comments that I have read), but I hate it.
The other day the PHB announces to the office: We are going to use Slack and move away from IM, Skype, etc.
So I signed up.
What I don't like:
-you need to keep a a tab open all the time,
-you need to keep an eye on that tab in case something comes up,
-the 'notifications' don't work all the time (Archlinux + Firefox)
It's like somebody took all the bad qualities of IRC, and shoehorned it into a web-page and all the horror that brings.
The features that I don't understand:
search-able logging of messages. Email and/or Pidgin already does that.
Group messages: Email already does that.
Transferring of files: Email and/or corporate LAN shares already do that.
But it does add the necessity of stopping my workflow every 5-10 minutes so I can check to see if there are any messages that _might_ apply to me.
The quicker it can be killed with fire the happier I will be. Or am I missing the point?
curmudgeonly - check
beard - check
Unix admin - check
Perhaps there is no hope for me. Next thing you know people will want to take pictures with their cellphones! =)
That's a bummer. I use Slack on Mac+iOS with the "native" app (which really just frames the web app but helps integrate notifications, etc.)
For the most part I experience none of the issues you mention. I get pinged when my name is @'d or on some emergency channels where I have it set to ping me on any message. I find it massively more useful and less distracting than IM+Email.
The one notification challenge I do have with it is that channels can move quickly enough that if I get @'d more than once while I'm afk for a bit it's easy for me to to respond to the latest ping I got but not catch earlier ones. I'd really like it to have a separate view that summarizes your mentions.
As for the features you don't understand:
- Email doesn't give you search for conversations you weren't a part of to begin with. That's the huge upside of a transparent-by-default tool like Slack when it comes to search. I can search for "Solr latency" and find conversations I wasn't a part of, from a time I may not have even been an employee. That's huge.
- File transfers - same, see above for email. Corp Lan shares rarely get search/indexing right.
Oh and FWIW I'm at least 70% curmudgeonly neckbeard too :)
Honestly, there is now way on earth I'd allow my company to use Slack given their Termination clause. It's a nice idea, but total loss of control of your data is just not worth it.
<snip>
We also reserve the right to terminate your account (or the access privileges of any end user) and this TOS at any time for any reason, or no reason, with or without notice. Upon termination of your account, we will have no obligation to maintain or provide Your Data, and will delete or destroy all copies of Your Data in our possession or control, in a reasonably expedient way, unless legally prohibited.
</snip>
"We can terminate your account for any reason" is a legal term which should be in every SaaS's (default) terms of service. Your lawyer will be happy to explain why. If you'd like a less formal explanation, consider what happens when you get a phone call at 4 AM in the morning which begins "Hello, is this the owner of $COMPANY? Great. This is Sgt. Stevens with the $CITY police department. I have a lawyer named John Smith in my office here. Mr. Smith alleges that you're assisting in the violation of a temporary restraining order."
Data retention is a separate issue, but I can envision reasons why I'd want to reserve a maximally "We don't owe it to you" clause, as a SaaS operator. (Slack, for example, allows arbitrary file uploads. This is a high risk feature, for a lot of reasons, data security, copyright compliance, and explosive reputational risk being only three of them.)
As a separate matter: if these clauses discomfit you, speak to enterprise sales. For $10,000+ you can negotiate better ones. If you do not wish to pay $10,000+, that's fine, but you don't get custom legal language.
> "We can terminate your account for any reason" is a legal term which should be in every SaaS's (default) terms of service. Your lawyer will be happy to explain why.
For a free account this might well work.
For a paid account, a unilateral-termination right, if not worded properly, could kill all the legal protections of the terms of service by turning the TOS into an "illusory contract." [1] The SaaS provider could lose its limitations of liability, choice-of-law and choice-of-forum clause, arbitration provision, etc.
A better approach might be to provide that the SaaS provider can temporarily suspend the account for good reason, and perhaps enumerate some example reasons. That could be coupled with a termination for cause clause (with termination following notice and an opportunity to cure except in egregious cases).
Usual disclaimer: I'm not your lawyer, this isn't legal advice, YMMV, small differences in fact can make big differences in outcome, check with your own lawyer before making decisions, etc., etc.
I guess I don't get it. When you terminate the account you can stop accepting payments and auto-trigger a pro-rated refund. Then the consideration is "the service, while you used it" or "the money, while we accepted it", depending on which party you are. Can you expand on that idea?
> When you terminate the account you can stop accepting payments and auto-trigger a pro-rated refund. Then the consideration is "the service, while you used it" or "the money, while we accepted it", depending on which party you are.
That sounds right --- the key difference is the refund, which wasn't mentioned in Patrick's original post.
If termination is for cause, you might not have to give a refund (although it'd look better to outsiders, and thus be more defensible in court, if you did give a refund).
Sure, as a provider, you want to have absolutely no binding obligation to the person you are providing service to.
Of course, conversely, as the person receiving service, you absolutely want the provider to have binding obligations -- and "we can cancel this at any time for any reason or no reason with no obligation to give you anything, including your data, afterwards" isn't what you want from a service you are relying on in any kind of business use.
> As a separate matter: if these clauses discomfit you, speak to enterprise sales.
Or just don't use the service that offers them. The reason services use boilerplate like this isn't that its essential, its that the perceived cost/benefit ratio warrants it because most people don't read TOS and don't change behavior based on them.
I think your comment conflates purpose and efficiency. This might be because you've only used slack for a few days now, and haven't realized its full power.
While email can be used for group communication, it's not ideal for the typical, constant interactions of a small team. Let's consider one use case. In email, file sharing with an individual is a multi-step process (new message, select recipient, insert subject, add attachment, write comments, send). In slack, you click upload, select file, and it's sent. No formality or friction in creating a new thread, selecting recipients, ad infinitum. Not to mention, it's hard to discover all the files shared between you and a group of people via email. This is important in an organization. A similar argument could be made for search and group messaging.
This snap judgement is the equivalent of suggesting that you'd forgo wearing gloves in the winter because wrapping warm pieces of fabric around your hands serves the same purpose. Purpose ≠ efficiency.
p.s. the desktop app > keeping a tab constantly open :)
> While email can be used for group communication, it's not ideal for the typical and constant interactions of a small team. Let's consider one use case. In email, file sharing with an individual is a multi-step process (new message, select recipient, insert subject, add attachment, write comments, send).
That's not so much a feature of "using email" as "the UX of a typical general-purpose email client". But, heck, the UX of sharing-via-email from even a typical Android app is somewhat more streamlined than what you present, and the same is true of many desktop apps that have email distribution as a feature. The recieving side is still a problem, though the kind of progress in email clients we've seen with -- just to look at gmail as an example -- schema-based actions and Inbox's workflow -- suggests that we may not be too far from the time when the recieve workflow for document-sharing-via-email is improved, after a fairly long period where email client workflows were fairly static.
> Do you realize you can connect w/ any IRC client?
Isn't that an horrific end-run around corporate security? or does irc.slack.com provide IP address whitelisting to restrict connections to those from within the company? Client SSL cert checking?
It just seems like a remarkable focal point that could give access to a lot of sensitive corporate data.
Look at all the sibling replies.... You just have to get a Mac - works great there! It's like Windows of 15 years ago, redux.
I've had the same experience as you with Slack and Hipchat. The front end guys with the shiny tools love it, and I live with it.
To the extent you're missing the point... it lets people that might not be able to set it up otherwise have secure messaging across desktop and mobile, and a unified place to have all the features that you mentioned vs getting familiar with nc/scp/ftp, grepping chat logs, etc.
That said, they must have one hell of a demo deck (or secret master plan) to get a $1B+ valuation.
I can't help but feel that HipChat have somewhat got the rug pulled from under them. Yes, they have done ok in their own right, but they've missed a few tricks as well.
One of the most important for our team is the ability to be signed into multiple organisations/accounts at the same time. Slack handles this perfectly, and HipChat not at all. I don't think I've seen a uservoice request with more votes than HipChat has for this http://help.hipchat.com/forums/138883-suggestions-ideas/sugg...
At least with hipchat I managed to get desktop notifications working in KDE + Firefox. I tried connecting to hipchat via the XMMP interface but it was horrible and didn't work very well. <rant> I fail to see what any of these "new-fangled" chat systems bring over IRC with a bot that tracks all the convos for history searching. </rant>
* > <rant> I fail to see what any of these "new-fangled" chat systems bring over IRC with a bot that tracks all the convos for history searching. </rant>
Easy: our entire agency (70+ people, half designers/developers, half management, client and accounts, etc) is using it, no matter how tech-savvy they are.
I love IRC, and I ran an IRC server for us developers for a good 6 months, but got us to switch over to Slack for the sole reason that accounts management and PMs are happy to use it, and we get transparent searching across every chat :)
Yup, I evaluated Slack and HipChat about a year ago and went with HipChat mostly because of Slack's lame Windows user experience. (IMHO, the Slack client also had too many bells and whistles. I would've like a default "simple" mode that just is an IM client with group chat.)
The funny thing is, I'm pretty sure the HipChat client is (or at least was) mostly just a shim around a web browser with some notification stuff bolted on. This doesn't need to be a huge undertaking.
Maybe it's changed since I last looked, but they didn't have a client at all. You could download Chrome and then install an "app" through the Chrome store. Well, some of my users prefer Firefox... I don't know, it just wasn't a great onboarding experience.
In retrospect, though, I might've gone too far the other way -- hipchat is pretty bare bones.
Slack's lack of a native Windows app sunk it for our org - then the rest adopted Hipchat and I think it is too late for us. I would love to use Slack, alas.
> people that might not be able to set it up otherwise
And also people who want it, are able to set it up, but would prefer to pay someone else to do it for them. Lots of useful services simply deliver something that people want, could do themselves, but prefer not to. Like and email providers and restaurants.
The valuation "sounds high", but they've shown that they can deliver a product that many users love and many individuals businesses will pay actual money for. I was shocked when I read the amount, but at least they sell a product!
I follow a public company that reported earnings recently, is in a fragmented multi-10-billion segment, has fundamental patents in it's field, and has 9-figure revenue. They have a valuation 1/10 of Slack's. I have no beef with Slack, but the valuation seems completely detached from fundamentals, or reality for that matter. I'm not hating, I hope all the founders & early contributors do great and never have to work again... but it is a high valuation.
I'm not sure those two mean real downside protection - but your point is correct - the VCs are betting on the very small chance that Slack is actually adopted by a huge number of businesses. Right now I'd guess it has a fairly narrow adoption among leading edge companies and SF/Valley natives. I'm sure there are exceptions to this - or maybe they truly have moved beyond that core audience and that's what is driving the valuation.
But that's what VCs do - make a large number of bets that will fail and one that blows it out the park.
Their valuation is a reflection of the growth in Slack's business - meaning they have strong enough growth that they can command that kind of valuation on a huge raise. It is also like a reflection on the limited number of companies that have that kind of growth.
But these kind of valuations also create a high-wire act for the companies in my experience. Growth must be maintained at all costs to justify the valuation. They spend like they're going out of style because of this.
From an information management perspective. Fewer clients (that Slack supports) and oauth2 integration (identity mgmt/password policies). Fewer opportunities to shoot yourself in the foot managing it, IMO.
Chrome lets you create apps that open in separate windows.[0] Then it's precisely as bad as any other IM app.
> you need to keep an eye on that tab in case something comes up
If your team members don't @ you when something might be relevant to you, sure. However, if they don't, you can just check every hour or so and get back to them. That's the beauty of Slack - it sits somewhere between IM and email, not everything has to be instant (because everything's saved and you can read it at your pace) but it has the ability to be used as such when enough people are around.
Eleven Giants, the team working on repairing Tiny Speck's previous game, uses Slack like this. Nobody's ever all going to be on at the same time, but everyone can join discussions.
> the 'notifications' don't work all the time (Archlinux + Firefox)
Never had this issue - I'd blame you using a "bleeding edge" distro.
I'm sure Firefox can do the same thing (allow you to create a shortcut to open a Firefox window in chromeless mode to a website), but I'm not sure how - hence only mentioning Chrome.
I'm with you. Maybe I'm too old to 'get it'. The progression of chat tools I've used for work collaboration: AIM -> Jabber -> HipChat -> Slack.
HipChat & Slack have some convenient features like emailing you if someone @'s you while you are away. Otherwise, I've used them all exactly he same way. I really don't get the hype.
I find the design and polish of Slack to be significantly superior to that of HipChat. Feature-wise though, I'm with you and couldn't pinpoint the differences.
HipChat is significantly cheaper and purports to offer video chat (didn't work for us the one time we tried it).
I too generally find Slack to be a more appealing user experience, though along with others I find the advantages over, say, Skype plus email to be fairly nugatory.
Because I work on a bunch of different teams for different organisations, I currently have open Slack, HipChat, Skype, and Apple's Messages, and I have to use Google Hangouts (which periodically completely kills Chrome for me) for a daily standup. I sort of miss those happy days when you could just use Adium for everything.
I still use Adium for most of my chat needs. Aim, Facebook, Gchat, IRC, Slack, etc. I'm agnostic to what service everyone else is using as long as I can login from Adium :)
The thing that I find really pleasant about Slack is that it syncs state between multiple clients really well. If I get a direct message and read it on my phone, my desktop client doesn't show it as unread. It doesn't send push notifications to my phone when I've got the desktop client open. It seems to strike a good balance between getting my attention for stuff that matters and letting me ignore stuff that doesn't.
Other group chats I've used don't do this as well. Skype is particularly awful in this regard.
For the most part, this is true, but they don't have it working quite right on Android. Opening the app from the launcher doesn't clear active notifications as it would if you launch the app by clicking on the notification. I've also gotten into a broken state where it tried to open one #channel in the wrong organization and just gave up until I quit the app and tried again.
Hopefully with this new money they can invest a bit more in their Android development efforts.
That's one of the main shortfalls I find in IRC as well. If it weren't for that (and basically requiring a bouncer/remote tmux session), I'd probably stick to IRC for quite a bit longer.
My experience is quite the opposite. I frequently get push notifications to my phone 15 minutes after reading and replying to PMs on my OS X native client.
I had the same general reaction when we tested it at our office, but I'm open to the idea that perhaps I'm missing the point.
In the press release, Butterfield says: "As the leader of a brand new product category, we have a huge advantage right now."
Does anyone know what product category he's referring to?
Enterprise collaboration seems like a quite old category to me, but maybe this is where I'm missing the point. Would be interested in hearing where I'm wrong.
I think the category he means is chat-based collaboration with a ton of seamless third-party integrations. Enterprise collaboration is too broad and elides the differences between Slack and, say, email or a wiki.
Of course, you might agree with the person you're responding to and think the differences are being exaggerated. Personally that has not been my experience.
I think he's referring to real-time enterprise collaboration. Things like Slack and Hipchat (and maybe Google Wave, but obviously that wasn't successful).
I am not the fan either, especially that even the "native" version is just a packaged web browser, and some features (like file upload) were broken when I tried. Unfortunately I do think it's a trend. Maybe someone who actually uses it could tell me why and also what makes it a 1B company?
Hm, that's a shame about the notifications. Their Mac desktop app (obviously doesn't apply that much to you) is pretty decent, I just keep it in the same workspace as my email client. I just see it as another way of communicating mostly, nothing necessarily too special. My favourite feature though is their service hooks, we have separate channels for Twitter (posts anytime our Twitter posts or gets @mentioned), Github for any activity on our repos, so I get notifications anytime someone pushes, creates an issue, etc. Obviously GH already sends email but this way all the communication is centralized, plus I hate email notifications.
I can't say I've fallen in love with it, but it's okay. The Mac app provides decent notifications - but I actually don't use it that much, I've installed the Android app on my phone and that works pretty well.
The notifications are clever enough to not ping me again on the phone if the Mac app is already running.
Integration with other things is dead simple - I had Sensu alerts sent to Slack in a matter of minutes.
I don't know. It's instant messaging, it's not like they've reinventing rocket science from scratch, but it's a good implementation of the concept, multiplatform, well integrated.
I also had slack forced on me after moving from IM to partychat, to hipchat and now slack. I didn't think it was all that great at first. However, it's really grown on me. It works really well for teams that include contractors and it's not the same level of distraction as IM.
My experience has been the exact opposite. I was reluctant to try yet another collaboration tool despite my teammates enthusiasm but after giving it a chance for a few days, I found the OSX and iOS apps extremely functional and well designed making Slack a real pleasure to use.
I think this comment will be dragged up in the future every time there's a post about the success of Slack, not unlike that first thread about Dropbox on HN.
Could the fickle finger of fate easily
cross their Buffet moat, i.e., could
this be a short term fad that fades?
If they have something solid, that is,
a much better solution for an important
problem where their solution will have
a significant barrier to entry, maybe
Fred Wilson's "large network of engaged
userrs" who have a significant switching
cost, then okay. Do they?
I think they do. We introduced slack at my employer a few months ago and we now have nearly 400 users on there. Traffic to our core mailing lists has dropped to almost nothing, I've replaced several weekly status meetings with low traffic slack channels and most of my co-workers now consider it indispensable. It's also been crucial in helping us establish a new remote office and some other remote workers.
If it's anywhere bear as transformational for other companies, I can see it continuing to grow for a long time.
I don't know about linux support, but they have native apps. This is much better than having an extra tab open all the time. As for your other concerns, you have listed a bunch of technologies that "already [do] that." Sure, there are plenty of them, but it's nice to have everything together.
Yes, as jscheel points out the native apps are far superior to the in browser option. We also link it into our production services such as pivotal, circleCI, etc to send notifications for build failures of deliverables. It's definitely a service that gets better the more you integrate it with your workflow. One stop shop and ease of use has also been a big selling point for my team
I'm on Linux and I've created a Chrome application shortcut for it (File>Create application shortcuts). At least means I don't accidentally close the tab and can launch it with Kupfer/your launcher of choice.
-you need to keep an eye on that tab in case something comes up,
-the 'notifications' don't work all the time (Archlinux + Firefox)
It's like somebody took all the bad qualities of IRC, and shoehorned it into a web-page and all the horror that brings. The features that I don't understand:
search-able logging of messages. Email and/or Pidgin already does that.
Group messages: Email already does that.
Transferring of files: Email and/or corporate LAN shares already do that.
But it does add the necessity of stopping my workflow every 5-10 minutes so I can check to see if there are any messages that _might_ apply to me.
The quicker it can be killed with fire the happier I will be. Or am I missing the point? curmudgeonly - check
beard - check
Unix admin - check
Perhaps there is no hope for me. Next thing you know people will want to take pictures with their cellphones! =)