Ebbe Holds The Jira Door Ajar

Ok well I’m having to improvise here because Inara Pey has basically hit me with some sort of psychic rays and stolen my planned blog post! Inara’s post of Ebbe: the promise of better communications and a more open JIRA is basically the post I had planned, obviously far more polished and eloquent in Inara’s format, but that’s the post I had planned, including the Innula Zenovka citation too, I will still use that.

So I’ll just concentrate on the Jira aspect here. In a forum response to Second Life builder Pamela Galli regarding the nerfing of the Jira, Linden Lab CEO Ebbe Altberg responds with:

Funny, both engineering and product heads here also didn’t like that jira was closed and want to open it up again. Proposal for how is in the works! I hope we can figure out how to do that in a way that works/scales soon. 

This isn’t funny really, it’s quite alarming. However let’s jump into the history machine, which shouldn’t be confused with the mystery machine, we’re not taking Fred or Daphne with us on this journey and there will be no doggy snacks. I am so easily sidetracked, where was I? Oh yes, back in time to examine when Jira’s go bad.

First of all let’s look back at what the Jira was and in many ways still is. The Jira was the place to report bugs and issues. People could also suggest new features there. The problem was that the Jira had a voting system and wasn’t exclusively used for bug reports and new feature suggestions. People would report their pet policy peeve on the Jira and gather a lot of votes. Now although people thought votes were important, the reality was that votes were not important in action being taken, although issues with high votes would obviously attract the attention of Linden Lab.

Then there were the Jira wars where people with admin rights would close a Jira and someone else would re-open it, a lot of commentary would be about opening and closing the Jira, people trading insults, in short there was a lot of off topic commentary. On top of that the Jira was, and remains, a clunky platform for people to use in general, although the more you use it, the easier it becomes to get your head around.

Despite all these issues, the Jira was a bloody useful resource. I used it as an official knowledge base when I ran into technical issues because it was far more informative on technical issues than the knowledge base is. I would also try and search for my issue before reporting it. The Jira worked well in this capacity. However some Lindens appeared to be tired of the off topic commentary, flame wars and lack of efficiency of the Jira. Now here’s where it gets interesting and a tad worrying, in a blog post in February 2011 entitled Improving our Lines of Communication with the Community Linden Lab announced that voting would be removed from The Jira :

Submit Bug Reports in JIRA: For those that aren’t familiar with JIRA, it is our public bug tracking system and it’s the best place to go to let us know about bugs or suggest new features that are proposed in User Group meetings or outside of those sessions.

Our development teams actively review the bugs that you submit in JIRA and do one of several things: place the item onto a development team backlog to address, ask the submitter to contact Support or provide more information if the developer cannot reproduce the bug, or close it and explain why we aren’t going to tackle this particular bug or idea.

It’s also important to note that we are going to remove the “voting” feature in JIRA in one month. Today, we do not use voting to triage or to make product decisions and the last thing that we want to do is set false expectations. So, when you are interested in what action we will take on a particular JIRA, use the JIRA “Watch” feature so that you will be immediately updated in email when there are new comments on that particular JIRA issue. We will continue to use the number of Watchers as an indication of the level of interest.

This was foolhardy for two reasons, the first that people started to watch instead of vote, under the false belief that the number of watchers would mean action would be taken. The second reason it was foolhardy was because it turned out that it wasn’t possible to remove voting from the Jira. The reason I said earlier that this is troubling is that Ebbe is looking at improving Linden Lab’s lines of communication, stop me if you’ve heard this one before.

Anyway plan 1 to nerf the Jira wasn’t successful, so Linden Lab turned to plan 2. In September 2012 Linden Lab put the boot into the Jira in a blog post entitled JIRA Update: Changes to The Bug Reporting Process Linden Lab announced:

Following is a summary of the JIRA changes:

  • All bugs should now be filed in the new BUG project, using the more streamlined submission form.
  • Second Life users will only see their own reported issues. When a Bug reaches the “Been Triaged” status, they will no longer be able to add comments to their issue.
  • Once a Bug reaches the “Accepted” or “Closed” status, it will not be updated. You can watch the Release Notes to see when and if a fix has been released for your issue.
  • Existing JIRAs will remain publicly visible. We will continue to review and work through these.

Now the real problem was the part about users only seeing their own reported issues. This seriously nerfed the Jira as a knowledge base style tool. Ok it wasn’t intended to be one on its own, but it was one. This also meant people struggled to share their findings. I reported an issue regarding disappearing Dwarfins. Judy Chestnut of the Dwarfins team had spoken to me about this issue, but she couldn’t add any comments to my Jira, she had to create her own and neither of us could view the others report.

This meant extra work for residents, extra work for Lindens and many frustrations. As Inara reported in her post, Innula Zenovka highlighted the problems of this on the forum:

I understand, though I don’t agree with, the reasons the jira was closed. However, it really is a major inconvenience when confronted with unexpected behaviour from scripts or the viewer, not to be able look it up in the jira to see if it’s a known issue or not, and what work-rounds might be available.

A while back, a friend contacted me to ask why a teleporter he’d made was working in some sims and not others. We spend about half an hour trying to make it work, and decided it must be sort of bug. I spent another hour or so playing with it in various sandboxes, and then writing up a jira about it, explaining what the problem was and which server channels were affected, only to discover it was a known issue, was to be fixed in the next week’s roll-outs and that I’d just wasted a couple of hours because I couldn’t look it up for myself (which, in the past, would have been the first thing I’d have tried).

Scripters in particular, but all serious content creators, I think, used to use the jira as an important reference tool when investigating problems, and I’m very unhappy we no longer have it.

It is frustrating for someone to spend so much time investigating an issue to discover it’s already a known issue. In many cases this will lead to a person not bothering to report an issue next time as they won’t feel it’s worth the time. Ebbe did acknowledge that he understood the issue, which is encouraging. Ebbe also has a background of working for Yahoo! and Microsoft, two companies whose users can be grumpy, engage in flame wars and create off topic debate on their user issue platforms etc.

Yahoo have Yahoo User Voice, Microsoft have Technet and Microsoft Answers. Both companies face the wrath of their users at times. Second Life is far from unique in that aspect and generally, although tempers can be raised, Second Life users are generally well behaved compared to other platforms.

Yahoo! have annoyed users with changes to mail, groups, Flickr and their homepage. They don’t please everyone, they never will. I found the changes to Flickr particularly annoying and cumbersome but it should be noted some people like the changes.

Microsoft have annoyed users with Windows 8 and the lack of a start button, Office 2013 and the lack of colour themes, that really is a shocker by the way, we refused to roll Office 2013 out at work and went to 2010 instead when we upgraded from Office 2007  because we knew users would complain in droves about the way Office 2013 looks, despite the fact that it has some very nifty features. I have no idea what Microsoft are thinking there at all. However some people like the minimalist style.

However the point is, the Jira will never be perfect, users will be grumpy, they will complain, they will argue, but alongside that there will be a hell of a lot of decent discussion, solutions and information and that is worth the pain points to preserve it as a useful shared resource. Minimising the pain points is of course an admirable aim, but they will never all be removed.

The Jira was better in its previous form, there has to be a better way forward than where it stands now. Good grief this turned into a long post!

4 Replies to “Ebbe Holds The Jira Door Ajar”

  1. You drilled down the way I was going to initially on the JIRA matter (then I decided to cover the comments and feedback on communications as well). Perhaps we should consider doing a combined blog after recent events! 🙂 .

  2. So, Ebbe is reviewing some decisions taken under Rod’s era – decisions on LL’s products, on JIRA… I wonder if the reviewing process will reach last names too. It would be interesting to investigate that, wouldn’t it?

Leave a Reply