The Problem with Slack at Scale
liorn.substack.com
The Problem with Slack at Scale
1–10 of 107 posts
Re: The Problem with Slack at Scale
#2I mean, good luck…
Getting into enterprise IM in 2022 is complete idiocy, but… Good luck.
We already have the solution, it’s zulip, yet no one uses it or cares.
> Slack is based on IRC
Not true any more, maybe in ancient history.
> It was built for synchronous communication.
Also wrong.
Re: The Problem with Slack at Scale
#3I'd add security to the list of things to address.
Aggregating all of the institutional knowledge of the company to one perpetually searchable global workspace hosted in a US company's shared cloud vulnerable to being downloaded and abused by a global breakin or any compromised employee account is bad.
Re: The Problem with Slack at Scale
#4Quill tried to solve this, but they eventually got acquired https://web.archive.org/web/20220122150126/https://quill.cha...
Re: The Problem with Slack at Scale
#5I mean, good luck… Getting into enterprise IM in 2022 is complete idiocy, but… Good luck. We already have the solution, it’s zulip, yet no one uses it or cares. > Slack is based on IRC Not true any more, maybe in ancient history. > It was built for synchronous communication. Also wrong.
Totally agree with this, not to mention that chat apps are some of the most boring, uninteresting, and braindead "problems" to solve and they have been basically completely "solved" over and over again since bulletin boards.
Re: The Problem with Slack at Scale
#6One thing I think Teams does well in this space, and where I feel Slack falls short, is the organisation of channels.
With Teams you can put people in... well teams, and each team has channels and other data associated with it. This hierarchy makes it easy for people to keep a multitude of channels and data organised without actually having to do it themselves.
Another thing I have to (begrudgingly) concede to Teams is that it's essentially "free" and from the perspective of an exec it does the same thing as Slack + has better video conferencing. Even simple things like meeting with 3rd parties is very easy to do on Teams - just send an invite. Whether or not the A/V actually works during the call or peoples PC's don't run out of memory mid-meeting is another matter, but on paper it looks good.
Re: The Problem with Slack at Scale
#7>Slack is based on IRC and is a messaging client. It was built for synchronous communication and not for collaboration. As a result, the way to scale Slack for a large company is to create more and more channels. This increases the amount of information generated in a company. Whilst this can be a good thing, Slack users don’t have the ability to filter high-quality information from low-quality. [emphasis mine]
This is the crux of the issue for me personally. The way Slack handles notifications and badges creates basically only three levels of importance: read this immediately, read this relatively quickly, or never read this. That works for small teams in which the gap between the second and third option is rather small. However as companies grow they generate more data and a huge gulf starts to form between the second and third option. There is now a wealth of information that necessitates a "you might want to read this eventually" category. Yet Slack provides no native way to handle this and the end result is that either you subject yourself to frequent interruptions for messages that aren't relevant to you or you ignore messages that are.
Re: The Problem with Slack at Scale
#8I'd add security to the list of things to address. Aggregating all of the institutional knowledge of the company to one perpetually searchable global workspace hosted in a US company's shared cloud vulnerable to being downloaded and abused by a global breakin or any compromised employee account is bad.
I thought much the same. Controlling the custody and flow of information is impossible with tens of thousands of users and pretty much no policies or significant monitoring in place.
Re: The Problem with Slack at Scale
#9I mean, good luck… Getting into enterprise IM in 2022 is complete idiocy, but… Good luck. We already have the solution, it’s zulip, yet no one uses it or cares. > Slack is based on IRC Not true any more, maybe in ancient history. > It was built for synchronous communication. Also wrong.
I remember at a previous company there were discussions on what to go with (HipChat/Stride had just been killed by Atlassian) and Zulip seemed to be the best free solution. But ultimately we went with Slack.
Re: The Problem with Slack at Scale
#10I mean, good luck… Getting into enterprise IM in 2022 is complete idiocy, but… Good luck. We already have the solution, it’s zulip, yet no one uses it or cares. > Slack is based on IRC Not true any more, maybe in ancient history. > It was built for synchronous communication. Also wrong.
Totally agree with this, not to mention that chat apps are some of the most boring, uninteresting, and braindead "problems" to solve and they have been basically completely "solved" over and over again since bulletin boards.
And yet Atlassian managed to not solve it with HipChat and not solve it again with Stride.