Hacker News new | past | comments | ask | show | jobs | submit login

Every single one of my clients uses Jira, and nobody I've ever met had a problem with it. It works, it's flexible, it's pretty easy to use. It's the industry standard, and for good reason, as far as I can tell.



Meet me.

I'm not sure what you're talking about. Confluence is OKish. But Jira is absurdly bad. I'm talking about the UI for creating Issues and Epics etc. The front-end devs that wrote it simply didn't have the ability. The parser for entering markup such as preformatted code blocks just doesn't work a lot of the time. The newer "Visual mode" just doesn't work a lot of the time. It simply needs to switch over to markdown and use a 3rd-party parser and renderer. The Visual Mode preview doesn't render using a fixed-width font. have you ever clicked on the little "Link" symbol in the top right of the text entry box? Obviously that should copy the current URL to the system clipboard. But they didn't know how to look that up on StackOverflow and instead made it a normal link to the current page (so you reload the page accidentally), with the link title saying

> title="Right click and copy link for a permanent link to this comment."

! You enter `bq.` to quote a line of text. This is all just some crap that someone with no design sense or standards came up with after 10 seconds thought.

I'm talking specifically about the quality of the UI. It is far, far, below the quality of UIs put out by respected modern products.


Not that it really matters much, but a lot of the design issues you're highlighting are due to the age of the software and Atlassian's seeming commitment to not breaking backward compatibility. In particular, Jira predates Markdown, so the software adopted the text formatter of the day, which was textile[0]. This is where the `bq.` syntax comes from. Jira didn't invent it from whole cloth -- it was adopted because that was the standard of the day. Likewise, it predates StackOverflow by a good 7 years. Some of the JavaScript used is old enough to be a college freshman.

As and end user, you may not (and probably should not) care about the historical context of its design decisions. But it's hardly the case that they hired a bunch of inept engineers. They've simply placed a large premium on backward compatibility and are still around today in large part because of that. Having said that, they really should find a way to support both Textile and Markdown if for no other reason than Bitbucket uses Markdown and it's confusing as hell having to switch between the two syntaxes if your company uses both products.

[0] -- https://textile-lang.com/


That's a helpful comment. Nevertheless

bq. commitment to not breaking backward compatibility

Backward compatibility with what? People's brains? We're talking about markup language and rendering right, which is not an API consumed by machines.

bq. it's hardly the case that they hired a bunch of inept engineers.

So why is the new "Visual Mode" WYSIWYG text entry mode so terrible? And why that absurd "link" icon in the top right of the text entry widget?


Backwards-compatible with people's brains is one aspect of it, sure. No one likes a constantly evolving UI that shuffles things around. But, also backwards-compatible with already entered issues. I wholeheartedly agree they should support Markdown, but I don't think they can just dump Textile in the process either, since it'd affect a load of already entered issues.

As for the link issue, I'm not entirely sure what you're referring to. I have an icon that looks like the Android "share" icon and that drops down a dialog with a link to the current page and a target user field. The link icon in the text entry field just adds a textile formatted link. I'm probably just overlooking something, but I'm not seeing what you described. And I never use the visual editor, so I can't speak to its quality.

I should note that I don't work for Atlassian and never have, so I don't have a horse in this race. But I have been using Jira since maybe 2004 due to its early adoption by the Apache Software Foundation. Jira is hardly perfect, but it's the least bad issue tracker I've used. At some level, I'm sure it's just a matter of preference. E.g., I know plenty of people that laud the GitHub issue tracker and I don't get it. It works well enough for small projects, but is too limiting for any project of non-trivial size, IMHO. I also find more than 2 or 3 labels in the issue list to just be a distracting sea of colors.

I hope you're able to find something that works well for you. I'll add that if you're using an on-premise version of Jira in your company, there's a high likelihood that you're running a dated release. I've found that some of the more aggravating issues people run into have actually been fixed, but not deployed in their environment. If you can find access to a running instance of the latest version, you might find it to be a more less frustrating experience.


I've never heard anyone call Jira the industry standard. There's way too much fragmentation in that market for anyone to be able to make that claim.

Jira can work well or it can work very poorly. It really depends on what you're trying to do with it and what resources you're willing to pour into it. That's why some people love it while others hate it.


It looks like the industry standard from where I'm sitting. Of all the companies I've worked for in the past 15 years, both as employee and as freelancer, I think only 2 didn't use Jira. 3 if I count my private projects (I used Pivotal).

In all that time, the only thing I've really heard people complain about, was when it was slow or down.


I'll see your anecdotal experience and raise you mine: in 9 years I've only worked with one company that did use Jira.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: