Hacker News Reader

Comment by SheinhardtWigCo | Hacker News Reader

Enterprises don't want federation. They want an easily searchable information repository much more than they want strong E2EE. They also don't see open source as an advantage.

Matrix was focused on things that deter enterprise customers while Slack built a machine for locking them in.

View on HN

> Enterprises don't want federation.

100% of companies require a way of communicating with other companies/people, and federated email and phone calls are extremely common ways (granted that non-federated means are becoming more popular, and I can believe there are a few companies whose only communication with the outside world is via non-federated means).

Companies have lists requirements for their tools and "not federated" isn't one of them, it's just that so far non-federated tools already fullfil the other requirements. Once Element is mature enough (and I'm sorry, but looking at the incoming issues on https://github.com/vector-im/element-web/issues?q=sort%3Aupd..., it doesn't look like it yet), then hopefully more companies will start considering it.

Also, some companies (at least some people in the one I work for, so I'm extrapolating) want to be able to text chat with other companies/people, and they currently do this with Teams (which requires both parties to have a Teams account). Slack allows this too if I understand. Federated chat would also give them this, so it is something they need, even if they don't realise that.

Enterprises may not want federation until they encounter a practical problem that just happens to be solved by federation. E.g. company/team acquisition, need to merge the chat ecosystems; don't want to disrupt the acquired team but you also need to have people in the rest of the company interact with existing channels etc. If both companies uses slack, the integration between such multiple accounts is possible.

Federation is also super useful if you're a multinational, since each geographic location could still operate autonomously in the face of an inability to route to the broader Internet. And with a protocol like Matrix, you're able to recover gracefully when federation is restored.

This is nice to have, but how often does that really happen?

Even once is usually enough, for the larger companies.

Do you think that's what's led to Slack's success though, that enterprises don't want federation?

I don't really think so. If Slack had exactly the same UX as it has but federated, I don't think that would have deterred very many customers. Maybe a few that have come on only after Slack already got so big.

It's not that federation per se is a downside for any particular customer. It's more that federation makes it a lot harder to make changes to your product, and it doesn't usually come with enough business upside to justify that cost.

Agreed 100%.

Federation is actually pretty hard. It's real that federation makes it harder/slower to make changes or add new features -- but on top of that spending time on federation (even on open source projects) necessarily takes time you could be spending on the non-federation UX.

And it's not just enterprise customers for whom in the end federation isn't nearly as prioritized as a good UX in other places.

"Enterprises" aren't a monolithic group.

Reply on news.ycombinator.com