Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Slack Platform Launch

slackhq.com

Post reply on HN
Print view View original

61–70 of 226 posts

Re: Slack Platform Launch

#61

Post by mnutt »

nkozyra wrote: post #43
studentrob wrote: post #35

Earlier quoted context omitted.

Less than 30 minutes for your tech team. 1) Install Sandstorm (curl https://install.sandstorm.io | bash) 2) Install Rocket Chat (one click) 3) Set up a domain. Done Regarding maintenance: I set it up 2 months ago and it's still running smoothly.

This is how long it takes to spin something like this up, but you're forgetting maintenance, domain knowledge, internal documentation, etc. People underestimate the costs of "rolling your own" or even going FOSS. It's absolutely never "set it and forget it." Most of these services that charge nominal fees ($5 - $100/mo) are interesting because you don't have to control your own support procedure and ostensibly they'r…

Sandstorm's updates are closer to Chrome's than regular open source software. Stick to the 'release' channel can you can be sure that you're always up to date. And Sandstorm locks its apps down so tightly that they often don't suffer from the same security vulnerabilities the same apps do outside Sandstorm.

That being said, to roll your own you'll need to run a box somewhere, select secure defaults, and keep on top of patching the box itself. It's not worth it unless you're really willing to spend the time to do it correctly.

Re: Slack Platform Launch

#62

Post by pallian »

chuhnk wrote: post #2

I feel like the world has lost it's mind slightly. When a 2 year old startup launches an $80m fund. I mean I understand they want to be a platform and can see the strength in funding projects that will empower a platform model but still. This is the point at which I think we're really in a boom heading in that bubble direction, and I was never one to fall in to the trap of calling it a bubble before. Love slack, love…

At 2M DAU, there's a lot of upside in building a bot for that audience. We just launched a simple expense management slack bot last week and in the past few days, our engagement has gone up... will post some numbers shortly. https://medium.com/trippeo-blog/introducing-expense-manageme...

Re: Slack Platform Launch

#63

Post by lotyrin »

jzelinskie wrote: post #52
chlestakoff wrote: post #44

Earlier quoted context omitted.

Will somebody please explain the use cases for slack bots to me? If they are basically glorified command line scripts, the I get it. But are there genuine and important use cases that require/can benefit from a richer dialogue?

There's a pretty good talk on "chatops" by the guys at GitHub[0]. Basically, it provides a dialog around a "shared command-line" where developers from learn from each other by simply watching chat. I think it's more beneficial to distributed companies. However, my question for those using chatops is basically how do you keep the repository for the bot "tame"? It needs access to _everything_ if you want it to be usefu…

Give the bot runtime (not in the repository, environment variable or something) credentials for a secret storage tool of some kind, where it can then fetch credentials for whatever it's allowed to do.

Re: Slack Platform Launch

#64

Post by debaserab2 »

Fishkins wrote: post #58
michael_fine wrote: post #21

Earlier quoted context omitted.

Genuine question -- how how is it been a revelation? It seems to be very similar to HipChat, which was just an iteration on previous chat systems.

For me, by far the biggest difference is the notifications. In HipChat, notifications are both buggy and almost completely unconfigurable[0]. If I sent a message to a colleague while they were offline, they would never get notified of it[1]. Slack notifications always work how I expect and are configurable to the degree I desire. Slack also has a nicer UI and better integrations for us, and generally just always work…

We've been on hipchat for about two years. Never had a problem with offline notifications. I get both an e-mail instantly and an IM when I log back onto the chat client. I don't recall this ever being a problem or ever not being there.

I dunno what notification settings hipchat is missing, but I'm not sure I need whatever it is missing. If I get tagged in a message, I get notified. Not sure that I want more than that. We have our deploys and a few other things hooked up to hipchat - at first when we started using it we went crazy with the third party integrations, than realized we didn't actually need all the noise the integration notifications gave us and turned a lot of it off.

We don't even pay for it and we get all of this.

Re: Slack Platform Launch

#65

Post by btown »

thatindiandude wrote: post #57

I don't think it's reasonable to say we're in a bubble because of a $80M fund around developing for a new platform. Few companies truly become platforms, and it's a misnomer to call it a bubble if this only happens to one company. It takes a lot of money or momentum to develop one. Facebook had by far the most compelling one in the last ten years, and the obvious incentive there was it's >300M users at the time. The…

A good comparison that hasn't been mentioned much in this thread is Salesforce's Salesforce1 Fund, which launched with $100M in 2014 to encourage people to build around the B2B app-creation platform it had launched a year earlier [0]. As the parent post mentions, it's important to make the case "compelling" for building best-in-class integrations with a relatively-recently-launched platform (whether or not that platform is sponsored by a historically huge corporation or not).

It's an interesting point that bringing multiple sources of external capital to bear on an ecosystem via a multi-sponsor fund, rather than into the business itself via an IPO, allows the business to grow at a natural rate, and avoids Slack contributing to a "bubble" in which companies race to go public to launch their own platforms. I think we've learned from the last time around that the public markets aren't the best place to manage expectations in technology.

As for why the sponsors made this investment, there's a very good case to be made that (a) efficient free-form communication within agile teams at small and large companies is a pattern as crucial to developing modern businesses as a CRM, and (b) Slack is perfectly poised to capture and hold this new market in the same way that Salesforce was. Certainly worth a $13mm commitment that's diversified amongst companies whose teams could still pivot should Slack not work out.

[0] https://www.salesforce.com/company/news-press/press-releases...

Re: Slack Platform Launch

#66

Post by gkop »

adamseabrook wrote: post #34

No need to build your entire startup on Slack but a solid integration can drive serious initial revenue for your startup especially if you do not get a lot of direct traffic. "One click install" is what helped Woocommerce and others get massive growth on the back of Wordpress through top exposure in the plugin directory. Slack is still very much at the bottom of the growth curve. I have seen electrical contractors wh…

I too am bullish on Slack and am with you right until "wiped out email." I've heard this said about Slack before and don't get it. Asana is a real email killer in that it's both realtime (instant) and structured (i.e. it wholly replaces email threads). On a meta level, why is it that it's so important for Slack's marketing that they "replace email"? Why isn't it enough for them to "go beyond email" (which is totally realistic and honest and still a potentially huge business)?

Re: Slack Platform Launch

#68

Post by mowgles »

chlestakoff wrote: post #44
fideloper wrote: post #26

2 points: 1. Remember when Dropbox was dumb, because rsync? (Bunch of naysayers here citing fee alternatives). From what I'm seeing, bots and integrations are great and here to stay. Businesses will gladly pay money in exchange for time and complexity not spent rolling your own. 2. This seems like a boon for us happy slack users!

Will somebody please explain the use cases for slack bots to me? If they are basically glorified command line scripts, the I get it. But are there genuine and important use cases that require/can benefit from a richer dialogue?

The New York Times has a queryable Slack bot that can tell their editors what topics/stories are likely to be popular. The backend is a machine learning algorithm.

This is essentially a command line script, as you say. But having that functionality and being able to discuss the results amongst your fellow editors is very powerful.

http://www.niemanlab.org/2015/08/the-new-york-times-built-a-...

Re: Slack Platform Launch

#69

Post by kentonv »

mnutt wrote: post #61
nkozyra wrote: post #43

Earlier quoted context omitted.

This is how long it takes to spin something like this up, but you're forgetting maintenance, domain knowledge, internal documentation, etc. People underestimate the costs of "rolling your own" or even going FOSS. It's absolutely never "set it and forget it." Most of these services that charge nominal fees ($5 - $100/mo) are interesting because you don't have to control your own support procedure and ostensibly they'r…

Sandstorm's updates are closer to Chrome's than regular open source software. Stick to the 'release' channel can you can be sure that you're always up to date. And Sandstorm locks its apps down so tightly that they often don't suffer from the same security vulnerabilities the same apps do outside Sandstorm. That being said, to roll your own you'll need to run a box somewhere, select secure defaults, and keep on top o…

Eh, honestly, if you're not running anything else on the box, you probably don't have to be overly proactive about maintaining it. Sandstorm handles updating its own dependencies and mitigating most kernel bugs. Maybe if you're leaving sshd open to the internet that's something to worry about, although if you read HN you'll surely know when that needs patching. :)

That said, we plan to build a "Sandstorm OS" that auto-updates everything including the kernel, and doesn't even have an SSH interface (wouldn't be useful anyway).

Re: Slack Platform Launch

#70

Post by btown »

rfrank wrote: post #55
radlad wrote: post #22

Earlier quoted context omitted.

The difference is people like using it. UX counts for a lot. The massive amount of integrations for it helps a lot too. Their goal, from what I've read, is to be a giant log for everything in your company, not just a chatroom service.

"Their goal, from what I've read, is to be a giant log for everything in your company..." Why would you ever want to entrust a log of everything in your company to a third party service? Honestly curious.

Because the expected loss from a malicious actor with access to those logs cloning your product (without any access to the actual people who thought up your product!) is often more than balanced by the expected benefit from getting everyone on a knowledge base with a brilliant UX.

Sure, you wouldn't have wanted to host the Apollo Program's lab notes on Soviet servers, but your cat selfie startup isn't quite as high-stakes as that.

Unless your cats are taking selfies on Mars. In which case, yeah, maybe developing an in-house chat service is worth the investment.

Post reply on HN