Adeon Writer

Very few resident comments on any Jira were what you would expect to see on a bug tracker.

Things like "Why isn't this fixed yet?" and "The same thing happens to me!" even if well-intended are just noise and pretty much spam.

So I think it wasn't unreasonable to close off open commenting...

But making them unviewable, I can't understand.

Leah Mayo

That is the infantile part. Temper tantrum.

Penny Patton

Altho, to be fair, there's a LOT of bugs for which it is entirely unreasonable that they've not been fixed, or even acknowledged.

LL will never increase their userbase if they just can't be bothered to fix basic features.

__________________
The Digital Pasture - SL from a cow's point of view. When the cow is also a designer.

Adeon Writer

Thing is, bug tracker comments are usually to be read by the software engineers (the people who fix things), not by managers (who tell the engineers what to fix)

So saying "OMG why didn't you fix this yet?!" isn't going to be directed at who you intend.

And if and when a Jira does finally get assigned, they have pages and pages of noise to read on day 1.

Innula Zenovka

But what I don't get is why, since clearly they aren't worried by accusations of being heavy-handed when they moderate the forums, they couldn't have tried rather harder to maintain order in the jira.

__________________

Penny Patton

I'm not disagreeing with that, however wishful thinking doesn't make for happy customers and there has never been an outlet for customers to direct their frustrations at anyone but the engineers.

LL insists on making their engineers their customer service people and apparently no one within LL is telling the engineers what needs to prioritized.

Given SL's unique nature, I can see the wisdom in a public Jira, however it shouldn't be the primary place for the average customer to report issues. There ought to be a bug report process that can be accessed either in-world or from the SL website catering to the average user. All reported issues should get in front of someone with the authority to dictate priorities to the engineers, and that person should not only be viewing SL from the engineer's point of view. The engineers aren't paying LL.

Ashiri

As an aside, the initial response on the LL forum and the later rant look like the product of different people. Second thoughts indeed.
I always found the rants to be extremely annoying; trying to pick through the comments for useful information took patience.
I'm thinking that the reason for the changes will soon be obvious... there have been some interesting suggestions.

KT Kingsley

It always seemed to me that a sure way of getting most non-showstopper JIRAs ignored was to fill the comments with abuse.

Penny Patton

Really? Seems to me there has never been any rhyme or reason to what gets ignored and what doesn't. Showstopper or not.

At any rate, a company which cannot deal with upset customers without getting so butthurt they resort to deliberately ignoring critical issues with their product is a company that will find itself unable to hold on to customers. (How's that new user retention working out? Not so well? Fancy that!)

If people are truly being abusive then you ban them but you still fix the problem.

Nika Talaj

To be honest, I remember thinking that something like this would happen when LL first merged the pJira and their (supposed) actual in-house working Jira database. Exposed LL's engineers to too much ranting.

I'm in a blue-sky mood, so here's a stab at Nika's ideal LL bug reporting method:

1. Customers can report bugs via a simple form. One of the pulldowns on the form is a list of the top n recently reported problems. The filer is given a report number that they can check status on later. There is internal triage of such issues by support people, who hand off a subset of them to be reviewed for bug triage. If a filer checks the status of their issue, they should see the results of support's review, and a bug number, if it was assigned one (or deemed already reported).

2. There is a Jira database, searchable by all unbanned residents. However, the ability to post there is granted by application only. But it should be a pretty liberal application process - for example (just an example; I'm making it up here): you have to be either a TPV developer or support person, or you have to be a scripter. (LL should be able to have a couple of automated jobs that tell them which avatars script).

3. The Jira bug format should always have included a separate section for workarounds. If there was one in LL's Jira, I never saw it used.

4. Under no circumstances should support have ever told users to 'go post on bug <n> on the Jira'. I don't know when this started, but it was just silly to have these bewildered non-tech types show up and say "me too!".

Jira is an engineering tool. But by permitting all and sundry to post to it, in fact ENCOURAGING random users to post, LL created their own mess.

Scripters, mentors/teachers, and estate managers need the ability to see issues, including ones that LL is working on, and we need to see workarounds. Since ALL workarounds that I ever saw were contributed by users, that means that some subset of tech savvy users should have the ability to post Jira comments that are visible to all.

Penny Patton

Jopsy Pendragon

It's a real shame we couldn't flag other's issues and comments as 'helpful' or 'unhelpful'.

Comments tagged as purely 'helpful' (defined as being: on topic, technically relevant, accurate) would be bolded or starred or something to make them stand out.

And issue contributors/commentors could be marked with the stigma of how many helpful/unhelpful contributions/comments they've added to the jira.

Sure it could be gamed, but people caught gaming the system, or repeatedly flagging 'unhelpful' content as 'helpful' (or vice versa) could easily be flagged by LL as a jira abusers... and their helpful/unhelpful tags completely excluded from issues/comment/users displayed totals.

Basically, it would be an incentive to use the jira the way it's intended.

Jopsy Pendragon

On a separate note... is anyone planning on importing their hot JIRA issues into the new bug reporting system? I have a strong feeling that the jira is now officially 'last priority', and still I have quite a few nuisances I'd like resolved.

KT Kingsley

It's just a vague feeling I had. I guess it's difficult to disentangle cause and effect, too.

Cerise

They didn't get there in one step with the forums.

They shut down huge parts they didn't like and said to go play elsewhere, initially offering to maintain a list of places like here. For a time that list was maintained and some forum bans were handed out to keep it running the way they wanted, then after about a year LL got bored. The unmaintained list vanished, moderation gradually evaporated as LL lost interest, and even the software only got cursory filter type patches because no one wanted to update it for real.

When a colo reshuffle finally forced then to do something about the vB system, they slapped together an awkward implementation on Jive's system. That varied from month to month from gung ho moderation, to apparently none at all, to what appeared to be some useless automated system. There could have been humans behind the "automated" period, but it was sufficiently indistinguishable from a coarse filter for that detail not to matter.

After an absence of months and months, they tried to restore all the old forum and blog content that people relied on as much as those JIRA entries that are now disappearing. Much of this content was rendered useless because all the attachments were lost (beyond the ones they lost earlier due to neglect) and some formatting mistranslated; residents were left to reverse engineer the hashing scheme in order to recover what information was left. Google eventually took up the slack from the absent index, and some kind anonymous Linden eventually implemented a clone of the resident tools to retrieve stuff by thread, but the damaged content remains broken to thius day. Old blog entries are still difficult to find, the only saving grace there is that the original Wordpress site was never wiped out.

It is only with the next Lithium system that they tried to get serious about moderation, and of course there was yet another round of permanently damaged content when the Jive content was migrated there.

You weren't expecting lessons to be learned from all that, were you? @_@

Evola

In this week's Deploy thread:

I don't know if that's always included in the post or if it's new, I don't read the release notes that often. I've also been seeing more people in the beta group inworld, while perving profiles lately. That might be the option to take?

Jack Abraham

It's always been there, Evola. What's new is Oskar taking a hard line with people who disrupt the technical discussion. If only Oskar ran JIRA...

Adeon Writer

The beta sandboxes are starting to get mainstream SL notice, so there's people joining the SL group just to get the sandboxes, for better or worse.

Void

Not I, I've long since abandoned the idea that it's my responsibility to suffer double the effort to repair nonsensical damage LL does on a whim

__________________
- These eyes can do more than see

Darien Caldwell

It was my understanding LL was still keeping the older JIRA issues and will still be resolving them. The new policy is just going foward. There shouldn't be any need to 're-import' issuse. They are already there and will still (not?) be worked on.

from: JIRA Update: Changes to The Bug Reporting Process - Second Life

__________________
[H]arsh Styles

Jopsy Pendragon

I laugh because, frankly, I've had a few trivially easy fix issues in the jira for ages, acknowledged and ignored. I doubt they'll ever get addressed.

I'm not going to let them forget that I still want them to be fixed though.

Void

ya mean like precision casting a float vs a rotation or vector to a string... they're not only different between MONO and LSL, but even within each VM... but that was blown off as "we don't feel a need to modify MONO's built in conversion to match LSO"... which I would've been fine with... except it's not even self consistent. Seriously, dumbest bug ever since it's obviously using different methods in different places.

Jopsy Pendragon

No even easier than that: Put the damned cursor/focus into the Inventory Search box when someone hits Ctrl-I.

Right now keyboard focus goes somewhere else, not really clear where, but it certainly isn't local chat or navigation. For a while focus was on the inventory tree and you could use the arrow keys to browse your folders.. but it seems to have moved elsewhere since.

*DRIVES ME FECKING NUTS*

Cerise

Yeah. I don't know who was thinking what, or if any thinking happened at all, but you have to press tab three times to get the cursor to the right place. That works for inventory and search. It became too much of a hassle too long ago to wrestle with JIRA (not the software, but LL) to bother filing a report.

Qie Niangao

Except to comment. But that can be manual, and just reference the old jira issue.