Okay, I’m a bit taken by surprise on this one, but let’s quickly take this from the top. Firstly, Linden Lab has added a simplified reporting form for Second Life JIRA issues. That part is pretty okay (and some might say that it is many years overdue) being that it is the sort of bug-reporting form that you might see for many major pieces of software, but that increases the triage-level workload on reports significantly. Now, that triage process has so far been split between the Second Life JIRA users and Linden Lab.
There’s a couple problems with bug triaging though. The first is that the Second Life JIRA users do it badly. The second is that Linden Lab do it badly. Okay, so issue triaging becomes more complicated now. So, ouch, right there.
Badly? How?
Well, the triage process has traditionally closed service-critical issues in error, erroneously marked issues as duplicates when they weren’t, and so on, while duplicate issues have – in turn – burgeoned, because the actual quality of reporting of issues is also … well, “a little bit shit” (in the ancient language of my people).
Now, the second part of this is that submitters no longer have access to reading or updating Second Life JIRA entries – that means that the community can no longer provide any input or assistance in the triage process. That’s where things start to get really awkward. That exponentially increases the cost and burden of triage and pushes that all onto Linden Lab, while simultaneously magnifying the impact of any triaging errors that the Lab’s triage staff makes.
That seems <sarcasm>awesome</sarcasm> right there.
So, the quick summary: Triaging gets more error-prone, difficult and time-consuming for Linden Lab – and Second Life users get to deal with any of the consequences of mistakes or delays.
Is that ideal? Heck no.
[via Second Life community site – thanks to Inara Pey]
This sounds like the type of decision PR or lawyers or some other clueless decision maker up top forces down on the company without even listening to the opinions of the company’s own experts on the product/service they provide…
Either that or they are intentionally selfdestructing.
“… or they are intentionally self destructing.”
I gotta add my vote to this conclusion Tigro. I can’t for a second believe anyone actually involved in the process would allow this change in basic operation to reach daylight. No matter how I try to conjure up a rational reason, I just cannot find one that has reaffirming evidence sufficient to let me pick it as the most probable. After eliminating all the patently improbable and obviously unbelievable ideas, the only one that keeps its place firmly atop the “Hill of Rightness” is the one that involves a sharp blade to LL’s throat .. by their own hand.
So much value gets lost with this move: referring people to known issues, often finding workarounds in Jira comments, collaborating on finding solutions…
Let me point one an example, which I was following recently:
https://jira.secondlife.com/browse/VWR-28843
So people experienced a crash when uploading stuff to SL (images, anims, etc.)
Someone figures out that the cause is installation of Microsoft Skydrive
Jira title gets changed
I play with registry and figure out the workaround which worked for many people judging from the comments
Whirly Fizzle points out the root cause. tcmalloc messing up windows function call tables
Linden Lab updates tcmalloc fixing this and several related issues (also linked on top).
All this now gone…
Whomever came up with this “brilliant” new process needs to find new employment ASAP. I really hope we’ll see some sign of adult supervision at Linden Lab soon.
Excellent example of what LL just lost, Latif.
Two thumbs up. Indeed, LL needs a few adults (real adults) to come in and make the important decisions.
Nice to see you posting again!
Sorry about the long gap. Medical appointments, project work, hectic schedule, and all of that.
It’s good to see you.
I really wish there was better news to post on SL, but knowing you’re still there helps a little.
I’m glad you’re well!
Glad to see that, no matter how many times they poke, prod, photograph and irradiate you, you’re still kicking bootie and taking names. *smiles*
As awful as the public JIRA was (awful in the sense that it was difficult to use and challenging to follow), this is no improvement. It will make it even more difficult for users of the system to understand why things stop working or find solutions. It will also isolate users and maybe that’s the intention. The JIRA was really the only semi-public space left where the Lab’s users could interact directly with the Lab with any hope of reply. Its sad, to see how the Lab is circling the wagons, further cutting themselves off from their customers.
Such a pompous decision.
From corporate point of view is a sure way to make annoying non issues such as bring back second names finally go away for good. The arrogance of the residents who simply won’t accept that surnames were eliminated for the residents own good and to make them better consumers.
The manager who took this decision is so detached from the community model of SL that the two sides can’t even communicate meaningfully with each other.
Bah humbug!
What left me stunned, is LL’s apparent lack of clarity as to how this will work.
At a User Group meeting, Alexa Linden (who appears responsible for the new system) was asked how LL will identify multiple duplicate reports on a the same issue if(/when) something serious enough goes wrong for people to file multiple reports.
The reply was, “We don’t know.”
That isn’t the most reassuring answer in the world.
The next worry is how long will JIRA items remain visible. Over the last few weeks there has been an odd drift in JIRA issues suddenly moving to a “restricted” access. MAINT-628 (snapshot tiling issue) is a case in point. This was a public JIRA Runitai announced a fix which was going to QA … now the JIRA has gone. No way to access it to see if (perchance) there has been any update from LL on the status of the fix.
Pathfinding issues have gone the same way. While I can in part understand the reasoning, given the large number of “irrelevant” comments posted on some of the JIRA (vis-a-vis the “FIX IT NOW” comment on my own article), it still gives the impression LL are trying to hide the issues they don’t want people to know about. Given the rising levels of alienation many are feeling towards the company, this isn’t perhaps the wisest impression to give to people.
And totally separate to this: good to see you back here and blogging. You’ve been missed!
[...] Tateru Nino [...]
After listening to thoughts on why this decision may have been made (other than the typical “another Linden Lab brain fart”), I find myself most believing the suggestion of a friend:
“Maybe they don’t want Steam to see how many thousands of unfixed bugs they have on the books.”
Another friend responded:
“As if Steam members aren’t going to recognize a bad software platform when they see it.”
It seems evident that Linden Lab does not like the typical JIRA format– the one that shows major, showstopper-level bugs existent for literally years without being fixed.
My response to this: everyone should pursue the logical consequence of this decision– the predictable one that everyone but Linden Lab seems to be able to foresee… and start JIRA reporting every bug we find so that Linden Lab is flooded with thousands of identical JIRA reports that otherwise would be handled under one report.
Hardly anyone has found the current JIRA system to be beneficial… for the reasons Tateru mentions and more (a major issue has been “Lindens with attitude” treating customers as if we’re their triage slaves rather than customers reporting an evident bug). Linden Lab is not the only company to make bug reporting user-specific or not open to public comment/examination. But they make doing so seem like a direct and intentional affront to their customers. Linden Lab “PR” seems to stand for “public retaliation”.
My take on this: Linden Lab doesn’t like the long list of their JIRA failures, nor a public record of customer reactions to such failures. The current JIRA is basically a historical archive of FAIL, company attitude and customer dissatisfaction– and someone up there realizes it. So in typical LL knee-jerk reaction, they “made the problem go away”… without consideration for the predictable consequences. This is a company that does not look ahead, never has looked ahead, focuses solely on the dollar sign– and as a result repeatedly runs headlong into the brick wall.
In short, this is just another in their long line of “we-say-so”, “Iron Curtain” decisions.
Should we start by reporting the failure of the JIRA search system to report any potential duplicate issues?
Or maybe, if Phoenix and Firestorm can’t see the bugs in the base viewer source, they can’t fix them first and keep leaving the official viewer in the dust.
At least once, the Firestorm team have made the decision to leaving fixing a bunch of bugs to the Labs, rather than have to resolve possibly inconsistent bugfixes in the future.
I think they knew they could fix the bugs, but had no confidence that the Labs would accept their fixes for the codebase. I don’t know if the problem is Oz Linden, or somebody who gives him orders, but Linden Labs has a poor grasp of Open Source concepts.
I think I’ve grown numb to anything Linden Lab does that amounts to it becoming a more insular, faceless, antisocial company. As inefficient as it was for the purpose, the Jira was still one of only a few public squares Lindens engaged us at in a meaningful way. Productively so a lot of the time, and even when Lindens ignored issues we were able to give each other workarounds.
This.
I wonder how hard it would be to setup a resident managed Jira-like site that would get at least almost as much attention from the resis as the official Jira while it was still public…
InWorldz is good……
Where was all this mourning for the JIRA system when people like Prok began spamming it to (literal) death around the clock with insane rants and selfish demands? Personally I think it died back then when it ceased to be a useful feedback tool for developers, and became more of a soapbox for idiots with a few real problems mixed in, deep down in the noise.
Considering the writing has been on the wall for so long, this news should not have come as a surprise to most.
Speaking for myself Angela, it isn’t surprise. It’s a mix of stark outrage at the lack of clear thinking this demonstrates, and a tongue-clucking disbelief at how something so obviously detrimental could even make it past the “drunk on Friday night shooting the breeze about stupid stuff we could do” moment that surely must have been the source of this latest change. If there is any surprise involved, it comes when we actually get to see or hear a Linden speak (for example, Alexa) and they seem to make sense at the moment .. then later turn around and implement changes that simply cannot come from the same state of cognitive awareness.
BTW: IIRC Prok was banned from contributing to the JIRA after her “momentary lack of decorum”. Granted, it wasn’t instantaneous, but it did happen. I believe the delay in taking that action was due to her initially having a real issue that was pertinent to the JIRA process.
WB Tateru.
Well we’ve been here before.
– - New Leadership at LL with a new direction and promises of a new era of cooperation and communication.
– - New technology developed, some desirable, some not, some well done, some not-so-well.
– - Communication channels opened gradually. Relationships with individual Lindens improve, sometimes to the point of real cooperative efforts.
– - It begins to become apparent that the New Leadership, for all it has accomplished, is not really fixing the problems with SL. Concurency and land ownership continues to slip.
– - The Leadership begins to behave erratically, taking extreme, over-the-top measures in the face of problems that require delicate and well thought out adjustment.
– - The conversation begins to turn to the next New Leadership, who they should represent and what they should do.
The “Post Rod Era” is just around the corner.
Of course I could be wrong, the age of steam could usher in thousands of new and permanent users. I just don’t believe it.
The Jira system worked because so many volunteers gave effort. Linking, verifying, combining and contributing. Volunteers gave word-arounds, solutions and suggestions. There is one Linden that does not like public participation and is know to close Jira’s for no reason at all.
That one Linden is now defining the new Linden policy (GO AWAY PESKY RESIDENTS) forgetting who really owns Second Life.
Face it, we, the users, own Second Life and we allow Linden Lab to do maintenance. Isn’t it time we send that message to Linden Lab?
Which linden is that?
Yes…but…PATHFINDING!!
I can’t comment on https://jira.secondlife.com/browse/SVC-4968 . Baker Linden needs testers but he can’t receive feedback on Jira. He is requesting feedback with in-world IM’s. Just a display on the improvement. Client participation will grind to a full stop.
Maybe they got embarrassed with being flooded by all the Pathfinding-related bug reports?
…”the community can no longer provide any input or assistance”…
Hasn’t that been company policy for quite some time?
(also, nice to see you back Tateru)
Has this been added to the publication?
http://lindensurvey.force.com/support/JIRASurvey
Rodvik spoke up about this on the SLU forums. He didn’t say anything you wouldn’t expect him to say. He basically flat out lied to everybody and said that this move is supposed to help them fix bugs better. This is the typical LL bullshit that I have come to expect from them. What did surprise me is the number of SLU people who were actually nice to Rodvik. I expected them to come out with torches and pitchforks when I saw Rodvik’s post. I guess a lot of people in SL have a higher tolerance for incompetence than I thought.
Anyway, you can see this trainwreck for yourselves here:
http://www.sluniverse.com/php/vb/general-sl-discussion/76184-jira-shutdown-limit-scripting-smaller.html#post1634050
It’s scary Rod doesn’t use the SL forums.
His statements are correct, it will reduce the workload for LL. The downside is, less bugs will be reported and will remain rogue in the wild and the userbase that maintained the Jira will have left. That’s not a prediction, it’s adding and subtraction, basic math.
It looks as though the Bandwidth Bug has been fix, at last. This was easily causing a 20-fold increase in traffic for Residents who happened to be in the wrong place in a region, and ignored the bandwidth limit set in the viewer.
Linden Labs, despite several anecdotal reports of people being hit by excess usage fees, did nothing to publicise it.
We don’t know what it did to the LL network, and its connections to the wider Internet, but my observation has been that a good many load-associated problems diminished once the fix was finally deployed to the Main Channel. Second Life stopped behaving as though it was over-crowded.
I knew about it because of the JIRA entry. People saw it, and amplified it. I took the trouble to post a warning to the General Discussion forum, referencing the JIRA. It was always possible to avoid the worst of the direct effects.
Since the fix, vehicles have become usable again. Texture loading has been faster. Many features which depend on the communication between computer and Linden Labs myriad servers are working better.
The bug struck at the same time as the completion of the Pathfinding rollout. The indirect effects are likely to have made Pathfinding look bad. I have seen unsupported claims that there was a security fix in that week’s rollout too, which blocked any chance of a roll-back.
On what I know, from what I have seen and the little that Lindens have said, this does not look to have been well-managed. And I am certain that the silence exacerbated the effects. It wouldn’t surprise me if the average network traffic doubled, as if there were twice as many Residents connected.
Would I have known if I hadn’t been able to read the JIRA?
Would I have been believed if I had passed on a warning based on an invisible JIRA?
The first time I saw the surge in traffic, maxing out my internet connection, it looked awfully like a DOS attack.
With or without the full JIRA access of the past, can we trust Linden labs when something goes wrong?
Wolf, the JIRA was a mess anyway and still is. I actually think that it´s a smart move to bury this public bug cemetary completely. And I doubt that more than a superminority of users really contributed anything usable there – however.
The not working functionality of the software does not really need a public rioting and bitching stage like that. I´d prefer something much more diiscrete and more effective, tho I am not sure if the new method will be better than the old one, but we shall see. It depends on how LL will communicate through the now established channels.
My conclusion is that the old JIRA system diid not contribute anything to a positive public image of Second Life and Linden Lab, nor did it really help development. I´´d not classify the shutdown as a sign of incompetence or unwillingness to communicate, but as an attempt to take bug handling to a professional level.
I think you missed the point.
The bandwidth bug had a serious effect on some customers, and the traffic levels seem to have had a noticeable effect on the general behavior of the Grid.
The Lindens chose not to say anything beyond the JIRA and few comments by Oskar Linden on the progress of the bugfix. And maybe something in user-group meetings.
It is as if they don’t care about their customers. Take away the JIRA, and what is left of Linden Lab communications?
If you want to say the JIRA was a lousy system, I can’t disagree. But what we have now looks worse.
Love your posts Tateru but is this it? We’re down to just one post per month from you?
Sorry – things have been very hectic and crazy of late (not the least with what seem like endless rounds of medical appointments). I’m not far, but dealing with overlapping crises has necessarily borrowed a lot of my attention of late.
I’ll dare speak for other fans and followers thirsty for your words and say we’re sorry to hear it’s medical appointments and overlapping crises which keep you occupied in other ways. May things settle and smooth for you soon.
Those words you do manage to pen make my daily visits worthwhile nonetheless and I/we thank you for them.