Hacker News Reader

Comment by albertgoeswoof | Hacker News Reader

Slack pretty much killed hipchat and drove directly into enterprises - which should have been atlassian sales. Once you have your tentacles in enterprise (and slack are almost there), it’s very easy to sell huge contracts of pretty average software. JIRA is great, but it’s not that complex, and it would not take much to start chipping it away

View on HN

JIRA is quite complex, but it suffers from a Swiss army knife complex, and at some point you end up "programming in JIRA" instead of using a much simpler purpose-built tool, or even building one.

i.e. the Inner Platform Effect (https://en.m.wikipedia.org/wiki/Inner-platform_effect), and all the worse for the programming being done without the benefit of programmers, in this case.

If your processes are complex, you're likely need flexible tool which allows arbitrary workflows. Of course you might argue that complex processes should bend to match tools, but not everyone who pays for software will agree, so Jira definitely has its market as a flexible tool even if it means that there should be dedicated Jira developers.

I heard from a friend at a fortune 100 company you would all know, that they tried piloting hipchat internally that hipchat simply did not scale well. Their servers choked under the load so frequently they got out of the contract.

> JIRA is great, but it’s not that complex

You and I must use a very different JIRA.

Jira is complex but no one needs that complexity. Jira is built to sell to executives, not for anyone to actually use. Executives need to know that they'll be tracking their employees time down to the second with Jira.

But then they're never going to look at the UI for Jira either because they're too busy for that.

I'm not gonna defend Jira. I don't like it. But this is the same thing people like to say about Office. It turns out that the 80/20 rule (80% of users only use 20% of features) never guarantees that even 50% use the _same_ 20%. You need to implement a lot more of Jira's complexity than you might think to get a significant part of Jira's market share.

(Slack is another good example. Slack is "just" an IRC client with better emoji/GIF support. But I'll be damned if Microsoft Teams doesn't _really suck_, even with the benefit of knowing how Slack does everything. Good software is hard and takes work.)

> I'm not gonna defend Jira. I don't like it. But this is the same thing people like to say about Office. It turns out that the 80/20 rule (80% of users only use 20% of features) never guarantees that even 50% use the _same_ 20%. You need to implement a lot more of Jira's complexity than you might think to get a significant part of Jira's market share.

Trello was the counterexample: it implemented less than 20% of Jira's features and was all the more useful for it. (thus Atlassian bought it and have started ruining it with bloat).

Not complex? We have to hire entire teams of people just to handle that confangled mess of options and wierd defaults and strange behavior to not slow down the developers and everyone else.

If Slack had an on-prem product, HipChat would be toast.

Hipchat is already toast, it was being replaced with Stride, and Stride got the axe. Atlassian has now partnered with Slack for chat.

And now they don't have an on-prem solution. Very upsetting.

Reply on news.ycombinator.com