Jerome Rasky

Software Engineer

On Slack's Bait and Switch

Posted at — Mar 13, 2018

I’ve been reading for a couple of days about the news that Slack is dropping support for IRC in their chat clients. Lobsters is definitely a biased place, but the sentiment there in the upvoted comments I think represents a non-trivial portion of at least the Lobsters community. Comments below come from the Lobsters thread on the news.

In particular, people are upset because they had a certain workflow with their own IRC or other client, and their workflow is now going to change.

This isn’t a new lesson with mission-critical, proprietary software. Anyone relying on profit-hungry, 3rd parties is just asking for it.

The issue I take with this kind of reasoning is that free software isn’t any different. Does nobody remember the whole Gnome 3 debacle, or the systemd fight, or pulseaudio, or any other of the myriad of tools that are controversial? Even just a cursory look at the landscape reveals that people get upset about their workflow changing no matter if a “profit-hungry 3rd party” is behind it.

In general, there are two ways to respond when the world changes. Either you adapt the world, or you adapt your behavior. Some people, in particular those complaining about Slack dropping support for IRC, are at least on this topic firmly in the “adapt the world” camp. Sometimes that’s reasonable, and I’ve certainly changed tooling or customized the tools that I have to achieve that. I use dark userstyles on websites, for instance, since I enjoy that look more.

I’m not a fan of Slack, particularly their godawful clients, but I think this article falls into the classic “It is what I want, therefore it is what everyone wants” fallacy. As my boss at Apple once told me, “we’d go broke if we made products for you.”

This comment gets at the idea of adapting your behavior. It turns out that people behave differently, and if you’re trying to accomodate a large number of people, you have to sacrifice certain niches. Furthermore, almost every time I’ve run into people using Slack it’s been through their client, and not through IRC. It takes time and resources to federate; at minimum, you have to thoroughly test your platform against whatever other implementations you’re federating against. Would those that want IRC support be willing to pay for its development? I highly, highly doubt it.

I’m not sure about the “godawful” clients argument. I know people don’t like electron apps, and that native apps a lot less resource intensive. But this begs the question, what sitation are you in that it matters so much? Is your workstation so resource constrained you can’t run an electron app? Why not change your workstation, in that case? Electron apps might be less performant, but how much time do you really spend chatting with people? Certainly, if others find Slack more usable, is it not worthwhile to adapt your workflow as a power user to suit them?

Finally, the people working for companies like Slack are like everybody else. Certainly they aren’t monsters, and I don’t doubt there is a contingent at Slack that is upset at the decision. Lots of choices that companies make can seem very backwards from the outside but be good, and conversely choices that seem outwardly good can be inwardly bad. I don’t think I know enough to really pass judgement on Slack’s choices, but it’s likely wise to think twice before imputing bad motives to them.

comments powered by Disqus