Discussion:
[SCRUMDEVELOPMENT] Bug Triages
hennadii.omelchenko@gmail.com [SCRUMDEVELOPMENT]
2015-02-23 08:19:07 UTC
Permalink
Hello,


Recently I've come across a practice in the organization which has been there from pre-agile times. It is called bug triages. I'm struggling to grasp the role of this one in Scrum as I've never met in trusted sources any reference to 'bug triages'.
is this practice to be considered a part of product backlog refinement? won't bug triages meeting create another standing session in addition to 'official' ceremonies?
how does it fit in Scrum (by spirit and letter), are there any alternatives to 'bug triage' meeting? I would greatly appreciate all the opinions and thoughts!
steve@ootac.com [SCRUMDEVELOPMENT]
2015-02-23 16:51:01 UTC
Permalink
Hi


What is the purpose of the 'bug triages'?


My understanding of 'triage' is to place a list of 'things' into some sort of order to be dealt with.


"Bug triages" would imply placing all the bugs into some order to be fixed.


Are your teams doing testing just before the end of the Sprint? It would appear so if they come up with a list of bugs.


IMHO, they should be constantly testing (TDD is a very good process) and fixing as bugs are found; there would never be a 'bug list' to triage.


Am I missing something here?
steveropa@gmail.com [SCRUMDEVELOPMENT]
2015-02-23 17:01:05 UTC
Permalink
Personally, I would see “bug triage” as a natural part of backlog grooming. Not necessarily having a separate meeting for either, but an organic activity of determining what next to work on. I don’t see any value in calling it out as a separate activity. If we have that many bugs, as Steve suggests, maybe its a good idea to look at your development/testing practices. Putting time in getting those where you want them to be would probably be more valuable and more rewarding than setting aside times for even more meetings.


Steve R.






Sent from Surface





From: ***@yahoogroups.com
Sent: ‎Monday‎, ‎February‎ ‎23‎, ‎2015 ‎9‎:‎51‎ ‎AM
To: ***@yahoogroups.com









Hi




What is the purpose of the 'bug triages'?




My understanding of 'triage' is to place a list of 'things' into some sort of order to be dealt with.




"Bug triages" would imply placing all the bugs into some order to be fixed.




Are your teams doing testing just before the end of the Sprint? It would appear so if they come up with a list of bugs.




IMHO, they should be constantly testing (TDD is a very good process) and fixing as bugs are found; there would never be a 'bug list' to triage.




Am I missing something here?
Laurent Bossavit mylists@bossavit.fr [SCRUMDEVELOPMENT]
2015-02-23 20:33:45 UTC
Permalink
• are there any alternatives to 'bug triage' meeting?
My long-time favorite: write fewer bugs.


Cheers,
Laurent



------------------------------------

------------------------------------

To Post a message, send it to: ***@eGroups.com
To Unsubscribe, send a blank message to: SCRUMDEVELOPMENT-***@eGroups.com
------------------------------------

Yahoo Groups Links

<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/SCRUMDEVELOPMENT/

<*> Your email settings:
Individual Email | Traditional

<*> To change settings online go to:
http://groups.yahoo.com/group/SCRUMDEVELOPMENT/join
(Yahoo! ID required)

<*> To change settings via email:
SCRUMDEVELOPMENT-***@yahoogroups.com
SCRUMDEVELOPMENT-***@yahoogroups.com

<*> To unsubscribe from this group, send an email to:
SCRUMDEVELOPMENT-***@yahoogroups.com

<*> Your use of Yahoo Groups is subject to:
https://info.yahoo.com/legal/us/yahoo/utos/terms/
Cass Dalton cassdalton73@gmail.com [SCRUMDEVELOPMENT]
2015-02-23 20:43:17 UTC
Permalink
I've never heard the term "bug triage" before, but this is my perspective
on bugs.


I have advocated for a firm distinction between actual bugs (it crashes
when I hit play) and unimplemented scope (it does nothing when I hit
play). Once you have that, I also advocate that real bugs are always at
the top of the backlog and should usually take precedence above any other
stories even within the sprint. This is the Lean idea of stopping the
assembly line when someone finds a defect. Sure it will slow progress in
the short term, but it gets people in the "no escaped defects" mindset.


The bug/scope distinction is important because many issues that a tester or
manager might consider bugs are actually scope that has intentionally not
been implemented yet. Everyone has a different opinion of what is
"critical", but the PO's is the only one that counts during development.
(Obviously it's the user's/customer's for the end product, but the PO needs
the ability choose scope incrementally)


-Cass
Post by Laurent Bossavit ***@bossavit.fr [SCRUMDEVELOPMENT]
• are there any alternatives to 'bug triage' meeting?
My long-time favorite: write fewer bugs.
Cheers,
Laurent
------------------------------------
------------------------------------
------------------------------------
Yahoo Groups Links
steve@ootac.com [SCRUMDEVELOPMENT]
2015-02-23 21:37:11 UTC
Permalink
The Definition of Done comes to mind; do you use it? All stories are worked on until they are Done which usually means there are no bugs.


If you use the DoD, surely there cannot be a list of bugs to triage.


Un-Done stories go back on the PB for consideration for the next Sprint; what could be simpler and more effective?
Cass Dalton cassdalton73@gmail.com [SCRUMDEVELOPMENT]
2015-02-23 21:46:47 UTC
Permalink
As much as I believe that a good Definition of Done is an absolute
necessity for success, it feels optimistic to believe that simply stating a
DoD and enforcing it will make bugs disappear. No matter how much
diligence you give to testing, something will always slip through. The DoD
helps remove ambiguity and set expectations, but it doesn't eliminate human
error, and some systems are complex enough that the edge case that produced
the bug either wasn't thought of (human error) or didn't occur often enough
in some stochastic aspect of the system (RF input for DSP, load testing for
web systems, etc).


So I assert that the team needs to have a plan for when a bug is discovered
in stories that were previously considered "Done". The belief that "Done"
means guaranteed bug free is an illusion. And when a bug does show up, the
team will find themselves unprepared for the situation and put themselves
in "triage" mode because the didn't have a plan or expectations for how to
proceed.


-Cass
Post by ***@ootac.com [SCRUMDEVELOPMENT]
The Definition of Done comes to mind; do you use it? All stories are
worked on until they are Done which usually means there are no bugs.
If you use the DoD, surely there cannot be a list of bugs to triage.
Un-Done stories go back on the PB for consideration for the next Sprint;
what could be simpler and more effective?
steve@ootac.com [SCRUMDEVELOPMENT]
2015-02-23 22:23:50 UTC
Permalink
Ah - so you are supposing that these bugs are discovered in stories that are Done but are found after the Sprint has been completed; am I correct?


Perhaps the original poster would care to enlighten us.


I would contend that such bugs become new stories and are ordered along with all other PBI. That's the way I handle them and it seems to work very well.


Of course, 'showstoppers' need to be fixed ASAP; the team have to decide how to handle them in the context of the Sprint
Wouter Lagerweij wouter@lagerweij.com [SCRUMDEVELOPMENT]
2015-02-23 23:03:11 UTC
Permalink
Hi,

If at all possible, I try and steer towards a stop-the-line approach. If
you're in a bad situation already, this can be almost impossible to
achieve. But I'd still recommend trying to get there, by some focused
effort (separate team, dedicated 'bandwidth') on reducing the number of
defects.

It's so easy to allow defects to live. So tempting to prioritise new
features over fixing issues. It's very hard to keep high quality when you
allow defects to accumulate.

Don't triage, prioritise. And prioritise defects high.If you don't, they
will eat your progress, take your speed, and you'll have to start all
over...

Wouter
Post by ***@ootac.com [SCRUMDEVELOPMENT]
Ah - so you are supposing that these bugs are discovered in stories that
are Done but are found after the Sprint has been completed; am I correct?
Perhaps the original poster would care to enlighten us.
I would contend that such bugs become new stories and are ordered along
with all other PBI. That's the way I handle them and it seems to work very
well.
Of course, 'showstoppers' need to be fixed ASAP; the team have to decide
how to handle them in the context of the Sprint
--
Wouter Lagerweij | ***@lagerweij.com
http://www.lagerweij.com | @wouterla <http://twitter.com/#!/wouterla>
linkedin.com/in/lagerweij | +31(0)6 28553506
Cass Dalton cassdalton73@gmail.com [SCRUMDEVELOPMENT]
2015-02-23 22:51:50 UTC
Permalink
Maybe I misunderstood and they are saving the day as a bug analysis /fixing

session. I agree that this is bad/wrong/unagile. That would make each
sprint just a mini-waterfall complete with a final integration phase. We
have had teams do that as well as we grew into agile. If you are doing
that, stop. Just stop. The Scrum guide doesnt mention this practice
because it has no place in agile.
Post by ***@ootac.com [SCRUMDEVELOPMENT]
Ah - so you are supposing that these bugs are discovered in stories that
are Done but are found after the Sprint has been completed; am I correct?
Perhaps the original poster would care to enlighten us.
I would contend that such bugs become new stories and are ordered along
with all other PBI. That's the way I handle them and it seems to work very
well.
Of course, 'showstoppers' need to be fixed ASAP; the team have to decide
how to handle them in the context of the Sprint
firepoet78@yahoo.com [SCRUMDEVELOPMENT]
2015-02-25 13:10:46 UTC
Permalink
While I can wholeheartedly agree that bug triage is certainly a smell that there is something blocking the organization from becoming aware of and correcting defects sooner, there is a dogmatism to this response that is striking. The linkage between behaviors in the Scrum Guide and the overall notion of "Agile" is harmful to the overall community. Pair Programming is also not in the Scrum Guide -- does that mean it has "no place in agile?"

Others have their own beliefs about how to embody the Manifesto -- judging their activities against the Scrum Guide is just plain shortsighted and reinforces the mistaken notion that the two are one and the same. They aren't. And the sooner we can embrace that and embody the distinction with our clients, the better off they will be.

I would also be extremely careful about calling people "unagile" -- this practice may be the next step for them in creating an authentic, human-oriented process that accepts and responds to feedback as quickly as it possibly can. While we should be holding everyone to a higher standard, it is extremely unlikely they will come along if we bully them by telling them they are wrong all the time.

My challenge for everyone who coaches Scrum and other kinds of agile teams: what is the good in what they are doing? How can you take that strength and call it out in them? What is the best way to motivate people to move even further beyond their current predicament?

The questions I would have for this particular thread: What problem are they trying to solve with bug triage? How might they get to the root cause of the problem? What would allow them to use this solution to its best effect without becoming addicted to it so they can later wean themselves off of it?

Wielding the hammer of "that's unagile," in my view, is not likely to get them there.
Cass Dalton cassdalton73@gmail.com [SCRUMDEVELOPMENT]
2015-02-25 19:04:12 UTC
Permalink
Obviously, the Agile community is passionate about their beliefs and it's
hard for any of us to lay out our opinion without sounding dogmatic. But I
don't know that people are saying that if it's not in the scrum guide then
it's not agile. I think what people mean when they say "un-agile" is that
the topic somehow goes against the grain of the philosophies behind the
agile movement. Very often, traditional organizations address problems
with traditional thinking even after having "transitioned" to agile.
Usually when people here say that something is "un-agile", they are saying
that the way an issue is being addressed is rooted in the old, traditional
mindset, not using "agile" thinking. How you handle a situation can lean
more traditional or more agile. For example, leaving bug fixing to the end
of the sprint leans heavily in the traditional direction. Underlying that
practice is a lack of focus on a real definition of done (the code is
"done" but it's not integrated, so it's not really done). I would call
that practice "un-agile" not because it's not in the scrum guide, but
because it is ignoring some very important ideology that is at the heart of
the Agile movement.


No one wants to be in a situation where bug triage is an option, but the
reality is that sometimes bugs show up, and sometimes they show up in
overwhelming numbers. The way you deal with the situation can lean more
traditional or more agile.


If I were coaching a team that was starting work in a legacy code base, I
would suggest the following categories of "bugs", each one with a different
plan for addressing them:
1) Legacy bugs - Those that existed before the start of the effort
2) New bugs - Those that were introduced by the new work
3) Unimplemented scope - Deficiencies in the product, usually with the new
work, that are really use cases or features that have not yet been
implemented.


3 is easy. This is really just a PBI on the backlog to be prioritized.
2 is also easy. The team should take a stop-the-line approach with these.
These are escaped defects. Teams should ultimately strive for a zero
escaped defects policy (i.e. do no new harm)
1 is the contentious one, but I draw from something that Ron said in a
previous thread. The thread was about writing "code improvement" stories,
but I think it is pertinent here:
[source:
https://groups.yahoo.com/neo/groups/scrumdevelopment/conversations/topics/55626
]


"Then what do we do about this bad code? I'm glad you asked.


If our Product Owner's stories do not cause us to make changes to the bad
code, we leave it alone. It is not hurting anyone, and it will not rot. It
will just sit there. Cleaning it up will take time away from stories, and
it will pay off, if it ever does, an unknown amount at an unknown time in
the future. Bad investment.


If stories cause us to work in the bad code, the bad code will slow us
down. It will make us throw up, causing extra breaks in the rest room. We
will be sad. Therefore, follow the "Boy Scout Rule", "Always leave the
campground cleaner than you found it". Clean up the code in a reasonable
fashion. Do not take ten days to rewrite it. We don't know how and anyway
it's inefficient. Just leave it better than it was. Over time, all the code
that is troubling us will get better. "


In summary, if you have bug in the legacy code base you are forced to work
in, you fix the bugs that need to be fixed for the story you're working
on. For example, if you are adding a codec to a media player application
that crashes whenever the user changes the volume, the first few stories
may be about playing a file but the DoD does not include changing the
volume. As soon as the PO's DoD requires that the user be able to change
the volume, the crash must be fixed as part of the story. The analysis of
prioritizing bugs is happening here just like in an independent "bug
triage" activity, but you're linking the analysis to providing prioritized
value. THAT is an agile solution to the problem. I would argue that any
activity that decouples work from prioritized value is "un-agile" and one
should be aware of that "un-agile" smell to see if there is an alternate
way of addressing the situation that is more in line with the core agile
values.


-Cass
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell that
there is something blocking the organization from becoming aware of and
correcting defects sooner, there is a dogmatism to this response that is
striking. The linkage between behaviors in the Scrum Guide and the overall
notion of "Agile" is harmful to the overall community. Pair Programming is
also not in the Scrum Guide -- does that mean it has "no place in agile?"
Others have their own beliefs about how to embody the Manifesto -- judging
their activities against the Scrum Guide is just plain shortsighted and
reinforces the mistaken notion that the two are one and the same. They
aren't. And the sooner we can embrace that and embody the distinction with
our clients, the better off they will be.
I would also be extremely careful about calling people "unagile" -- this
practice may be the next step for them in creating an authentic,
human-oriented process that accepts and responds to feedback as quickly as
it possibly can. While we should be holding everyone to a higher standard,
it is extremely unlikely they will come along if we bully them by telling
them they are wrong all the time.
My challenge for everyone who coaches Scrum and other kinds of agile
teams: what is the good in what they are doing? How can you take that
strength and call it out in them? What is the best way to motivate people
to move even further beyond their current predicament?
The questions I would have for this particular thread: What problem are
they trying to solve with bug triage? How might they get to the root cause
of the problem? What would allow them to use this solution to its best
effect without becoming addicted to it so they can later wean themselves
off of it?
Wielding the hammer of "that's unagile," in my view, is not likely to get them there.
Ron Jeffries ronjeffries@acm.org [SCRUMDEVELOPMENT]
2015-02-25 19:10:22 UTC
Permalink
FP,
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell that there is something blocking the organization from becoming aware of and correcting defects sooner, there is a dogmatism to this response that is striking.
It is important to distinguish between dogmatism and an experience-based firmly held opinion.

Ron Jeffries
ronjeffries.com <http://ronjeffries.com/>
I'm really pissed off by what people are passing off as "agile" these days.
You may have a red car, but that does not make it a Ferrari.
-- Steve Hayes
Adam Sroka adam.sroka@gmail.com [SCRUMDEVELOPMENT]
2015-02-25 21:24:44 UTC
Permalink
If you are doing bug triage you have too many bugs. That is why the
comparison to mass casualties is apt. If you have too many bugs there is a
flaw in your team's communication that is allowing potentially costly
mistakes to get too far down the line, possibly (probably) into the hands
of the customer. Whether you are a developer, a customer, or a business
owner, you don't even have to know what a manifesto is to see the problem
there.

If you are not doing bug triage that does not necessarily mean that you
don't still have too many bugs. You need to work on your team's
communication process to eliminate the source of the bugs (By
"communication process" I mean to include planning, testing, and handling
feedback from customers.) You may also want to pay particular attention to
bugs that are costing you money without derailing the rest of your process.
Triage may be an effective stop gap if used appropriately.

All of this is somewhat orthogonal to Agile. I'm actually using more of the
Lean part of my brain above, but I don't think there is any conflict
between the two on this (nor most) issue(s.)

The main symptom is you have too many defects. The solution is two-pronged:
figure out why you are creating them and address the root cause (hint: it's
a communication problem,) and build a safety net to catch them earlier
(TDD, pair programming, continuous integration, etc.)
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell that
there is something blocking the organization from becoming aware of and
correcting defects sooner, there is a dogmatism to this response that is
striking. The linkage between behaviors in the Scrum Guide and the overall
notion of "Agile" is harmful to the overall community. Pair Programming is
also not in the Scrum Guide -- does that mean it has "no place in agile?"
Others have their own beliefs about how to embody the Manifesto -- judging
their activities against the Scrum Guide is just plain shortsighted and
reinforces the mistaken notion that the two are one and the same. They
aren't. And the sooner we can embrace that and embody the distinction with
our clients, the better off they will be.
I would also be extremely careful about calling people "unagile" -- this
practice may be the next step for them in creating an authentic,
human-oriented process that accepts and responds to feedback as quickly as
it possibly can. While we should be holding everyone to a higher standard,
it is extremely unlikely they will come along if we bully them by telling
them they are wrong all the time.
My challenge for everyone who coaches Scrum and other kinds of agile
teams: what is the good in what they are doing? How can you take that
strength and call it out in them? What is the best way to motivate people
to move even further beyond their current predicament?
The questions I would have for this particular thread: What problem are
they trying to solve with bug triage? How might they get to the root cause
of the problem? What would allow them to use this solution to its best
effect without becoming addicted to it so they can later wean themselves
off of it?
Wielding the hammer of "that's unagile," in my view, is not likely to get them there.
Tim Wright tim@tfwright.co.nz [SCRUMDEVELOPMENT]
2015-02-26 17:00:01 UTC
Permalink
Hi,


There's another reason to do bug triage that hasn't come up yet. It's when
you have too many bug reports - but not many actual bugs.


This can happen when your project is working in a context like this:


* The large client has very aggressive deadlines
* Several business units inside the client will benefit if the project fails
* Your business has signed SLAs on defect fixing (in integration test
environments)
* The client raises every defect as "severity one" giving you 2 days to fix
each defect.
* 95% of the defects are in the order of "working as designed",
"duplicate", "this was your requirement", or "function not implemented in
this drop as agreed".
* The client refuses to listen to anyone except for a member of the
development team. Also, few people in the organisation have sufficient
knowledge to classify the defects properly anyway.


The only way we could handle that particular environment was to setup daily
defect triage meetings with the client while we increased knowledge in our
org and tried to manage the client better.4


Tim
Post by Adam Sroka ***@gmail.com [SCRUMDEVELOPMENT]
If you are doing bug triage you have too many bugs. That is why the
comparison to mass casualties is apt. If you have too many bugs there is a
flaw in your team's communication that is allowing potentially costly
mistakes to get too far down the line, possibly (probably) into the hands
of the customer. Whether you are a developer, a customer, or a business
owner, you don't even have to know what a manifesto is to see the problem
there.
If you are not doing bug triage that does not necessarily mean that you
don't still have too many bugs. You need to work on your team's
communication process to eliminate the source of the bugs (By
"communication process" I mean to include planning, testing, and handling
feedback from customers.) You may also want to pay particular attention to
bugs that are costing you money without derailing the rest of your process.
Triage may be an effective stop gap if used appropriately.
All of this is somewhat orthogonal to Agile. I'm actually using more of
the Lean part of my brain above, but I don't think there is any conflict
between the two on this (nor most) issue(s.)
The main symptom is you have too many defects. The solution is
two-pronged: figure out why you are creating them and address the root
cause (hint: it's a communication problem,) and build a safety net to catch
them earlier (TDD, pair programming, continuous integration, etc.)
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell
that there is something blocking the organization from becoming aware of
and correcting defects sooner, there is a dogmatism to this response that
is striking. The linkage between behaviors in the Scrum Guide and the
overall notion of "Agile" is harmful to the overall community. Pair
Programming is also not in the Scrum Guide -- does that mean it has "no
place in agile?"
Others have their own beliefs about how to embody the Manifesto --
judging their activities against the Scrum Guide is just plain shortsighted
and reinforces the mistaken notion that the two are one and the same. They
aren't. And the sooner we can embrace that and embody the distinction with
our clients, the better off they will be.
I would also be extremely careful about calling people "unagile" -- this
practice may be the next step for them in creating an authentic,
human-oriented process that accepts and responds to feedback as quickly as
it possibly can. While we should be holding everyone to a higher standard,
it is extremely unlikely they will come along if we bully them by telling
them they are wrong all the time.
My challenge for everyone who coaches Scrum and other kinds of agile
teams: what is the good in what they are doing? How can you take that
strength and call it out in them? What is the best way to motivate people
to move even further beyond their current predicament?
The questions I would have for this particular thread: What problem are
they trying to solve with bug triage? How might they get to the root cause
of the problem? What would allow them to use this solution to its best
effect without becoming addicted to it so they can later wean themselves
off of it?
Wielding the hammer of "that's unagile," in my view, is not likely to get them there.
--
Tim
021 251 5593
http://www.linkedin.com/in/drtimwright
Adam Sroka adam.sroka@gmail.com [SCRUMDEVELOPMENT]
2015-02-26 20:43:03 UTC
Permalink
Yeah, having too many bugs that aren't bugs is essentially the same problem
I was describing with a similar solution.

BTW, I don't really recommend having bug triage "meetings". The way I would
prefer is to look at the bugs right when they arrive, stop the line if they
are critical, put them on a card and give it to the PO if they need to be
fixed but aren't critical, push the delete button if they are duplicates or
you won't fix them. That's my rather extreme :-) version of triage.

Also, bug tracking tools are awful. They seem like they're a way to enable
communication, but they're actually a way to delay it. Put important stuff
on the wall where people can see it. Let the customer talk to you in a chat
window, on the phone (or video conference), or preferably in person. Just
say no to bug tracking tools.
Post by Tim Wright ***@tfwright.co.nz [SCRUMDEVELOPMENT]
Hi,
There's another reason to do bug triage that hasn't come up yet. It's when
you have too many bug reports - but not many actual bugs.
* The large client has very aggressive deadlines
* Several business units inside the client will benefit if the project fails
* Your business has signed SLAs on defect fixing (in integration test
environments)
* The client raises every defect as "severity one" giving you 2 days to
fix each defect.
* 95% of the defects are in the order of "working as designed",
"duplicate", "this was your requirement", or "function not implemented in
this drop as agreed".
* The client refuses to listen to anyone except for a member of the
development team. Also, few people in the organisation have sufficient
knowledge to classify the defects properly anyway.
The only way we could handle that particular environment was to setup
daily defect triage meetings with the client while we increased knowledge
in our org and tried to manage the client better.4
Tim
Post by Adam Sroka ***@gmail.com [SCRUMDEVELOPMENT]
If you are doing bug triage you have too many bugs. That is why the
comparison to mass casualties is apt. If you have too many bugs there is a
flaw in your team's communication that is allowing potentially costly
mistakes to get too far down the line, possibly (probably) into the hands
of the customer. Whether you are a developer, a customer, or a business
owner, you don't even have to know what a manifesto is to see the problem
there.
If you are not doing bug triage that does not necessarily mean that you
don't still have too many bugs. You need to work on your team's
communication process to eliminate the source of the bugs (By
"communication process" I mean to include planning, testing, and handling
feedback from customers.) You may also want to pay particular attention to
bugs that are costing you money without derailing the rest of your process.
Triage may be an effective stop gap if used appropriately.
All of this is somewhat orthogonal to Agile. I'm actually using more of
the Lean part of my brain above, but I don't think there is any conflict
between the two on this (nor most) issue(s.)
The main symptom is you have too many defects. The solution is
two-pronged: figure out why you are creating them and address the root
cause (hint: it's a communication problem,) and build a safety net to catch
them earlier (TDD, pair programming, continuous integration, etc.)
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell
that there is something blocking the organization from becoming aware of
and correcting defects sooner, there is a dogmatism to this response that
is striking. The linkage between behaviors in the Scrum Guide and the
overall notion of "Agile" is harmful to the overall community. Pair
Programming is also not in the Scrum Guide -- does that mean it has "no
place in agile?"
Others have their own beliefs about how to embody the Manifesto --
judging their activities against the Scrum Guide is just plain shortsighted
and reinforces the mistaken notion that the two are one and the same. They
aren't. And the sooner we can embrace that and embody the distinction with
our clients, the better off they will be.
I would also be extremely careful about calling people "unagile" -- this
practice may be the next step for them in creating an authentic,
human-oriented process that accepts and responds to feedback as quickly as
it possibly can. While we should be holding everyone to a higher standard,
it is extremely unlikely they will come along if we bully them by telling
them they are wrong all the time.
My challenge for everyone who coaches Scrum and other kinds of agile
teams: what is the good in what they are doing? How can you take that
strength and call it out in them? What is the best way to motivate people
to move even further beyond their current predicament?
The questions I would have for this particular thread: What problem are
they trying to solve with bug triage? How might they get to the root cause
of the problem? What would allow them to use this solution to its best
effect without becoming addicted to it so they can later wean themselves
off of it?
Wielding the hammer of "that's unagile," in my view, is not likely to get them there.
--
Tim
021 251 5593
http://www.linkedin.com/in/drtimwright
Cass Dalton cassdalton73@gmail.com [SCRUMDEVELOPMENT]
2015-02-26 20:48:41 UTC
Permalink
RE: tracking tools:
We have to use a CMMI compliant issue tracker to track stories, bugs, etc
to meet contractual requirements. But my teams always also use stickies on
a whiteboard and the SM keeps the two in sync. The issue tracker (Jira
Agile) is tied into the code review and CM tool (Stash), so the issue
tracker is part of the workflow, but I like to keep something that gives an
"always on" and visceral representation of the work.
Post by Adam Sroka ***@gmail.com [SCRUMDEVELOPMENT]
Yeah, having too many bugs that aren't bugs is essentially the same
problem I was describing with a similar solution.
BTW, I don't really recommend having bug triage "meetings". The way I
would prefer is to look at the bugs right when they arrive, stop the line
if they are critical, put them on a card and give it to the PO if they need
to be fixed but aren't critical, push the delete button if they are
duplicates or you won't fix them. That's my rather extreme :-) version of
triage.
Also, bug tracking tools are awful. They seem like they're a way to enable
communication, but they're actually a way to delay it. Put important stuff
on the wall where people can see it. Let the customer talk to you in a chat
window, on the phone (or video conference), or preferably in person. Just
say no to bug tracking tools.
Post by Tim Wright ***@tfwright.co.nz [SCRUMDEVELOPMENT]
Hi,
There's another reason to do bug triage that hasn't come up yet. It's
when you have too many bug reports - but not many actual bugs.
* The large client has very aggressive deadlines
* Several business units inside the client will benefit if the project fails
* Your business has signed SLAs on defect fixing (in integration test
environments)
* The client raises every defect as "severity one" giving you 2 days to
fix each defect.
* 95% of the defects are in the order of "working as designed",
"duplicate", "this was your requirement", or "function not implemented in
this drop as agreed".
* The client refuses to listen to anyone except for a member of the
development team. Also, few people in the organisation have sufficient
knowledge to classify the defects properly anyway.
The only way we could handle that particular environment was to setup
daily defect triage meetings with the client while we increased knowledge
in our org and tried to manage the client better.4
Tim
Post by Adam Sroka ***@gmail.com [SCRUMDEVELOPMENT]
If you are doing bug triage you have too many bugs. That is why the
comparison to mass casualties is apt. If you have too many bugs there is a
flaw in your team's communication that is allowing potentially costly
mistakes to get too far down the line, possibly (probably) into the hands
of the customer. Whether you are a developer, a customer, or a business
owner, you don't even have to know what a manifesto is to see the problem
there.
If you are not doing bug triage that does not necessarily mean that you
don't still have too many bugs. You need to work on your team's
communication process to eliminate the source of the bugs (By
"communication process" I mean to include planning, testing, and handling
feedback from customers.) You may also want to pay particular attention to
bugs that are costing you money without derailing the rest of your process.
Triage may be an effective stop gap if used appropriately.
All of this is somewhat orthogonal to Agile. I'm actually using more of
the Lean part of my brain above, but I don't think there is any conflict
between the two on this (nor most) issue(s.)
The main symptom is you have too many defects. The solution is
two-pronged: figure out why you are creating them and address the root
cause (hint: it's a communication problem,) and build a safety net to catch
them earlier (TDD, pair programming, continuous integration, etc.)
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell
that there is something blocking the organization from becoming aware of
and correcting defects sooner, there is a dogmatism to this response that
is striking. The linkage between behaviors in the Scrum Guide and the
overall notion of "Agile" is harmful to the overall community. Pair
Programming is also not in the Scrum Guide -- does that mean it has "no
place in agile?"
Others have their own beliefs about how to embody the Manifesto --
judging their activities against the Scrum Guide is just plain shortsighted
and reinforces the mistaken notion that the two are one and the same. They
aren't. And the sooner we can embrace that and embody the distinction with
our clients, the better off they will be.
I would also be extremely careful about calling people "unagile" --
this practice may be the next step for them in creating an authentic,
human-oriented process that accepts and responds to feedback as quickly as
it possibly can. While we should be holding everyone to a higher standard,
it is extremely unlikely they will come along if we bully them by telling
them they are wrong all the time.
My challenge for everyone who coaches Scrum and other kinds of agile
teams: what is the good in what they are doing? How can you take that
strength and call it out in them? What is the best way to motivate people
to move even further beyond their current predicament?
The questions I would have for this particular thread: What problem are
they trying to solve with bug triage? How might they get to the root cause
of the problem? What would allow them to use this solution to its best
effect without becoming addicted to it so they can later wean themselves
off of it?
Wielding the hammer of "that's unagile," in my view, is not likely to get them there.
--
Tim
021 251 5593
http://www.linkedin.com/in/drtimwright
George Dinwiddie lists@idiacomputing.com [SCRUMDEVELOPMENT]
2015-02-26 21:02:36 UTC
Permalink
Cass,

Heh! Your contract and your SCAMPI appraiser may demand an issue
tracker, but the phrase "CMMI compliant issue tracker" is an oxymoron.

- George
Post by Cass Dalton ***@gmail.com [SCRUMDEVELOPMENT]
We have to use a CMMI compliant issue tracker to track stories, bugs,
etc to meet contractual requirements. But my teams always also use
stickies on a whiteboard and the SM keeps the two in sync. The issue
tracker (Jira Agile) is tied into the code review and CM tool (Stash),
so the issue tracker is part of the workflow, but I like to keep
something that gives an "always on" and visceral representation of the work.
__
Yeah, having too many bugs that aren't bugs is essentially the same
problem I was describing with a similar solution.
BTW, I don't really recommend having bug triage "meetings". The way
I would prefer is to look at the bugs right when they arrive, stop
the line if they are critical, put them on a card and give it to the
PO if they need to be fixed but aren't critical, push the delete
button if they are duplicates or you won't fix them. That's my
rather extreme :-) version of triage.
Also, bug tracking tools are awful. They seem like they're a way to
enable communication, but they're actually a way to delay it. Put
important stuff on the wall where people can see it. Let the
customer talk to you in a chat window, on the phone (or video
conference), or preferably in person. Just say no to bug tracking
tools.
__
Hi,
There's another reason to do bug triage that hasn't come up yet.
It's when you have too many bug reports - but not many actual bugs.
* The large client has very aggressive deadlines
* Several business units inside the client will benefit if the
project fails
* Your business has signed SLAs on defect fixing (in integration
test environments)
* The client raises every defect as "severity one" giving you 2
days to fix each defect.
* 95% of the defects are in the order of "working as designed",
"duplicate", "this was your requirement", or "function not
implemented in this drop as agreed".
* The client refuses to listen to anyone except for a member of
the development team. Also, few people in the organisation have
sufficient knowledge to classify the defects properly anyway.
The only way we could handle that particular environment was to
setup daily defect triage meetings with the client while we
increased knowledge in our org and tried to manage the client
better.4
Tim
__
If you are doing bug triage you have too many bugs. That is
why the comparison to mass casualties is apt. If you have
too many bugs there is a flaw in your team's communication
that is allowing potentially costly mistakes to get too far
down the line, possibly (probably) into the hands of the
customer. Whether you are a developer, a customer, or a
business owner, you don't even have to know what a manifesto
is to see the problem there.
If you are not doing bug triage that does not necessarily
mean that you don't still have too many bugs. You need to
work on your team's communication process to eliminate the
source of the bugs (By "communication process" I mean to
include planning, testing, and handling feedback from
customers.) You may also want to pay particular attention to
bugs that are costing you money without derailing the rest
of your process. Triage may be an effective stop gap if used
appropriately.
All of this is somewhat orthogonal to Agile. I'm actually
using more of the Lean part of my brain above, but I don't
think there is any conflict between the two on this (nor
most) issue(s.)
The main symptom is you have too many defects. The solution
is two-pronged: figure out why you are creating them and
address the root cause (hint: it's a communication problem,)
and build a safety net to catch them earlier (TDD, pair
programming, continuous integration, etc.)
__
While I can wholeheartedly agree that bug triage is
certainly a smell that there is something blocking the
organization from becoming aware of and correcting
defects sooner, there is a dogmatism to this response
that is striking. The linkage between behaviors in the
Scrum Guide and the overall notion of "Agile" is harmful
to the overall community. Pair Programming is also not
in the Scrum Guide -- does that mean it has "no place in
agile?"
Others have their own beliefs about how to embody the
Manifesto -- judging their activities against the Scrum
Guide is just plain shortsighted and reinforces the
mistaken notion that the two are one and the same. They
aren't. And the sooner we can embrace that and embody
the distinction with our clients, the better off they
will be.
I would also be extremely careful about calling people
"unagile" -- this practice may be the next step for them
in creating an authentic, human-oriented process that
accepts and responds to feedback as quickly as it
possibly can. While we should be holding everyone to a
higher standard, it is extremely unlikely they will come
along if we bully them by telling them they are wrong
all the time.
My challenge for everyone who coaches Scrum and other
kinds of agile teams: what is the good in what they are
doing? How can you take that strength and call it out in
them? What is the best way to motivate people to move
even further beyond their current predicament?
What problem are they trying to solve with bug triage?
How might they get to the root cause of the problem?
What would allow them to use this solution to its best
effect without becoming addicted to it so they can later
wean themselves off of it?
Wielding the hammer of "that's unagile," in my view, is
not likely to get them there.
--
Tim
021 251 5593
http://www.linkedin.com/in/drtimwright
--
----------------------------------------------------------------------
* George Dinwiddie * http://blog.gdinwiddie.com
Software Development http://www.idiacomputing.com
Consultant and Coach http://www.agilemaryland.org
----------------------------------------------------------------------
Adam Sroka adam.sroka@gmail.com [SCRUMDEVELOPMENT]
2015-02-26 21:10:59 UTC
Permalink
We use Jira where I work too. I hope one day our process will improve
enough that we can throw it away. My point is that throwing it away is a
noble goal. The best way that I've seen that done is to never start using
it in the first place. There are other ways, though...
Post by Cass Dalton ***@gmail.com [SCRUMDEVELOPMENT]
We have to use a CMMI compliant issue tracker to track stories, bugs, etc
to meet contractual requirements. But my teams always also use stickies on
a whiteboard and the SM keeps the two in sync. The issue tracker (Jira
Agile) is tied into the code review and CM tool (Stash), so the issue
tracker is part of the workflow, but I like to keep something that gives an
"always on" and visceral representation of the work.
Post by Adam Sroka ***@gmail.com [SCRUMDEVELOPMENT]
Yeah, having too many bugs that aren't bugs is essentially the same
problem I was describing with a similar solution.
BTW, I don't really recommend having bug triage "meetings". The way I
would prefer is to look at the bugs right when they arrive, stop the line
if they are critical, put them on a card and give it to the PO if they need
to be fixed but aren't critical, push the delete button if they are
duplicates or you won't fix them. That's my rather extreme :-) version of
triage.
Also, bug tracking tools are awful. They seem like they're a way to
enable communication, but they're actually a way to delay it. Put important
stuff on the wall where people can see it. Let the customer talk to you in
a chat window, on the phone (or video conference), or preferably in person.
Just say no to bug tracking tools.
Post by Tim Wright ***@tfwright.co.nz [SCRUMDEVELOPMENT]
Hi,
There's another reason to do bug triage that hasn't come up yet. It's
when you have too many bug reports - but not many actual bugs.
* The large client has very aggressive deadlines
* Several business units inside the client will benefit if the project fails
* Your business has signed SLAs on defect fixing (in integration test
environments)
* The client raises every defect as "severity one" giving you 2 days to
fix each defect.
* 95% of the defects are in the order of "working as designed",
"duplicate", "this was your requirement", or "function not implemented in
this drop as agreed".
* The client refuses to listen to anyone except for a member of the
development team. Also, few people in the organisation have sufficient
knowledge to classify the defects properly anyway.
The only way we could handle that particular environment was to setup
daily defect triage meetings with the client while we increased knowledge
in our org and tried to manage the client better.4
Tim
Post by Adam Sroka ***@gmail.com [SCRUMDEVELOPMENT]
If you are doing bug triage you have too many bugs. That is why the
comparison to mass casualties is apt. If you have too many bugs there is a
flaw in your team's communication that is allowing potentially costly
mistakes to get too far down the line, possibly (probably) into the hands
of the customer. Whether you are a developer, a customer, or a business
owner, you don't even have to know what a manifesto is to see the problem
there.
If you are not doing bug triage that does not necessarily mean that you
don't still have too many bugs. You need to work on your team's
communication process to eliminate the source of the bugs (By
"communication process" I mean to include planning, testing, and handling
feedback from customers.) You may also want to pay particular attention to
bugs that are costing you money without derailing the rest of your process.
Triage may be an effective stop gap if used appropriately.
All of this is somewhat orthogonal to Agile. I'm actually using more of
the Lean part of my brain above, but I don't think there is any conflict
between the two on this (nor most) issue(s.)
The main symptom is you have too many defects. The solution is
two-pronged: figure out why you are creating them and address the root
cause (hint: it's a communication problem,) and build a safety net to catch
them earlier (TDD, pair programming, continuous integration, etc.)
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell
that there is something blocking the organization from becoming aware of
and correcting defects sooner, there is a dogmatism to this response that
is striking. The linkage between behaviors in the Scrum Guide and the
overall notion of "Agile" is harmful to the overall community. Pair
Programming is also not in the Scrum Guide -- does that mean it has "no
place in agile?"
Others have their own beliefs about how to embody the Manifesto --
judging their activities against the Scrum Guide is just plain shortsighted
and reinforces the mistaken notion that the two are one and the same. They
aren't. And the sooner we can embrace that and embody the distinction with
our clients, the better off they will be.
I would also be extremely careful about calling people "unagile" --
this practice may be the next step for them in creating an authentic,
human-oriented process that accepts and responds to feedback as quickly as
it possibly can. While we should be holding everyone to a higher standard,
it is extremely unlikely they will come along if we bully them by telling
them they are wrong all the time.
My challenge for everyone who coaches Scrum and other kinds of agile
teams: what is the good in what they are doing? How can you take that
strength and call it out in them? What is the best way to motivate people
to move even further beyond their current predicament?
The questions I would have for this particular thread: What problem
are they trying to solve with bug triage? How might they get to the root
cause of the problem? What would allow them to use this solution to its
best effect without becoming addicted to it so they can later wean
themselves off of it?
Wielding the hammer of "that's unagile," in my view, is not likely to
get them there.
--
Tim
021 251 5593
http://www.linkedin.com/in/drtimwright
'Tim@tfwright.co.nz' tim@tfwright.co.nz [SCRUMDEVELOPMENT]
2015-02-26 21:01:52 UTC
Permalink
I do agree. Made harder to use cards when client, development, and product ownership are in different countries. The one thing you don't want is 4 different bug tracking tools...


Tim
Sent from phone.
Yeah, having too many bugs that aren't bugs is essentially the same problem I was describing with a similar solution.
BTW, I don't really recommend having bug triage "meetings". The way I would prefer is to look at the bugs right when they arrive, stop the line if they are critical, put them on a card and give it to the PO if they need to be fixed but aren't critical, push the delete button if they are duplicates or you won't fix them. That's my rather extreme :-) version of triage.
Also, bug tracking tools are awful. They seem like they're a way to enable communication, but they're actually a way to delay it. Put important stuff on the wall where people can see it. Let the customer talk to you in a chat window, on the phone (or video conference), or preferably in person. Just say no to bug tracking tools.
Hi,
There's another reason to do bug triage that hasn't come up yet. It's when you have too many bug reports - but not many actual bugs.
* The large client has very aggressive deadlines
* Several business units inside the client will benefit if the project fails
* Your business has signed SLAs on defect fixing (in integration test environments)
* The client raises every defect as "severity one" giving you 2 days to fix each defect.
* 95% of the defects are in the order of "working as designed", "duplicate", "this was your requirement", or "function not implemented in this drop as agreed".
* The client refuses to listen to anyone except for a member of the development team. Also, few people in the organisation have sufficient knowledge to classify the defects properly anyway.
The only way we could handle that particular environment was to setup daily defect triage meetings with the client while we increased knowledge in our org and tried to manage the client better.4
Tim
If you are doing bug triage you have too many bugs. That is why the comparison to mass casualties is apt. If you have too many bugs there is a flaw in your team's communication that is allowing potentially costly mistakes to get too far down the line, possibly (probably) into the hands of the customer. Whether you are a developer, a customer, or a business owner, you don't even have to know what a manifesto is to see the problem there.
If you are not doing bug triage that does not necessarily mean that you don't still have too many bugs. You need to work on your team's communication process to eliminate the source of the bugs (By "communication process" I mean to include planning, testing, and handling feedback from customers.) You may also want to pay particular attention to bugs that are costing you money without derailing the rest of your process. Triage may be an effective stop gap if used appropriately.
All of this is somewhat orthogonal to Agile. I'm actually using more of the Lean part of my brain above, but I don't think there is any conflict between the two on this (nor most) issue(s.)
The main symptom is you have too many defects. The solution is two-pronged: figure out why you are creating them and address the root cause (hint: it's a communication problem,) and build a safety net to catch them earlier (TDD, pair programming, continuous integration, etc.)
Post by ***@yahoo.com [SCRUMDEVELOPMENT]
While I can wholeheartedly agree that bug triage is certainly a smell that there is something blocking the organization from becoming aware of and correcting defects sooner, there is a dogmatism to this response that is striking. The linkage between behaviors in the Scrum Guide and the overall notion of "Agile" is harmful to the overall community. Pair Programming is also not in the Scrum Guide -- does that mean it has "no place in agile?"
Others have their own beliefs about how to embody the Manifesto -- judging their activities against the Scrum Guide is just plain shortsighted and reinforces the mistaken notion that the two are one and the same. They aren't. And the sooner we can embrace that and embody the distinction with our clients, the better off they will be.
I would also be extremely careful about calling people "unagile" -- this practice may be the next step for them in creating an authentic, human-oriented process that accepts and responds to feedback as quickly as it possibly can. While we should be holding everyone to a higher standard, it is extremely unlikely they will come along if we bully them by telling them they are wrong all the time.
My challenge for everyone who coaches Scrum and other kinds of agile teams: what is the good in what they are doing? How can you take that strength and call it out in them? What is the best way to motivate people to move even further beyond their current predicament?
The questions I would have for this particular thread: What problem are they trying to solve with bug triage? How might they get to the root cause of the problem? What would allow them to use this solution to its best effect without becoming addicted to it so they can later wean themselves off of it?
Wielding the hammer of "that's unagile," in my view, is not likely to get them there.
--
Tim
021 251 5593
http://www.linkedin.com/in/drtimwright
hennadii.omelchenko@gmail.com [SCRUMDEVELOPMENT]
2015-02-24 06:32:48 UTC
Permalink
Seems like my case falls under the category mentioned before: this is a legacy product with low overall quality and almost without safety net. Bugs pop up here and there even without any changes let alone when refactoring starts.

As to your DoD question, the answer is yes, we do use DoD. Nevertheless, sometimes it happens that some people think it is worth to say PBI is done even with bugs found and put those to some sort of 'backlog' (but not product backlog).
steve@ootac.com [SCRUMDEVELOPMENT]
2015-02-24 17:38:20 UTC
Permalink
Hi


Ah - legacy products - what joy (not!)


I recently worked with a team doing product development on a legacy product and transforming to Agile.


One of the first things we did was what might be called a 'bug triage'; there was a huge 'legacy bug list' going back 10 years! The majority were now out of date and many had been 'fixed' with a previous development.


Those left in the list were added to the PB and ordered along with new/revised functionality to be developed; worked very well.


As to bugs in the legacy code that 'pop-up' during a Sprint, well fixing them has to be part of the DoD or they will never get fixed - very much the Scrum philosophy.


I do not quite understand. You say you are using a DoD; does this include something like "All discovered bugs fixed" or "All tests passed". If so, who are these 'people' who say that the PBI is Done when it isn't? The Product Owner is the only person who has authority to make such a decision but it would be extremely rare.


Was the DoD set by the Team and agreed with the PO?


If it was, then anybody except the PO who 'breaks' the DoD for any story is displaying a woeful lack of discipline and needs close mentoring about Agile!!


If it wasn't, then your organisation has arguably a greater problem than the bugs, that of not implementing Agile/Scrum principles and values correctly. If this is not correct, what else is lurking in the shadows.


I would strongly recommend that you get some independent Agile audit.
Cass Dalton cassdalton73@gmail.com [SCRUMDEVELOPMENT]
2015-02-24 18:29:01 UTC
Permalink
I agree that legacy work can be a much different flavor of agile than new
development. I still assert that the PO should look at each legacy bug to
determine if it is really a front and center bug in the required
functionality or if it is a problem in future scope. The converse of that
is that any PO that is dealing with legacy code should take even more care
than usual to focus on thin slivers of functionality/value so that the DoD
for a particular story is narrow enough to allow for deferring of some
legacy "bugs" as future scope. Otherwise, legacy work can be a quagmire
for the team and you will take a morale hit.


-Cass
Post by ***@ootac.com [SCRUMDEVELOPMENT]
Hi
Ah - legacy products - what joy (not!)
I recently worked with a team doing product development on a legacy
product and transforming to Agile.
One of the first things we did was what might be called a 'bug triage';
there was a huge 'legacy bug list' going back 10 years! The majority were
now out of date and many had been 'fixed' with a previous development.
Those left in the list were added to the PB and ordered along with
new/revised functionality to be developed; worked very well.
As to bugs in the legacy code that 'pop-up' during a Sprint, well fixing
them has to be part of the DoD or they will never get fixed - very much the
Scrum philosophy.
I do not quite understand. You say you are using a DoD; does this include
something like "All discovered bugs fixed" or "All tests passed". If so,
who are these 'people' who say that the PBI is Done when it isn't? The
Product Owner is the only person who has authority to make such a decision
but it would be extremely rare.
Was the DoD set by the Team and agreed with the PO?
If it was, then anybody except the PO who 'breaks' the DoD for any story
is displaying a woeful lack of discipline and needs close mentoring about
Agile!!
If it wasn't, then your organisation has arguably a greater problem than
the bugs, that of not implementing Agile/Scrum principles and values
correctly. If this is not correct, what else is lurking in the shadows.
I would strongly recommend that you get some independent Agile audit.
hennadii.omelchenko@gmail.com [SCRUMDEVELOPMENT]
2015-02-24 21:38:33 UTC
Permalink
Steve,

DoD doesn't have a line about ALL bugs, but several to deal with bugs depending on there severity (for example, only bugs of highest severity are to be fixed as a part of work on PBI), so formally it can be said that a PBI is Done according to DoD. So, it is usual thing to move bugs beyond the sprint. As I now figured, this is the very goal of Triage: to deal with severity so that all know what will be moved.


DoD was initially suggested as a template by coach (included 'bug segregation by severity' principle right from the very beginning) and slightly adjusted later by the team.


PO is indeed responsible for making decision for 'moving' bugs beyond the scope of the sprint, and people are bosses, some team members and agile coach who have significant influence on those decisions.


As to a greater problem in orgnization adopting Agile you mentioned, it is widely believed that that organization has been Agile enough just having some people with titles 'Scrum Master' and 'Product Owner', and talking about Agile values and/or principles is excessively idealistic and impractically philosophical. After all, what else does a true Agile organization need but scrum masters and product owners?


I would like to thank all of you who took part in answering my question and expressing your thoughts. I's been really helpful for me. Thank you!
Wouter Lagerweij wouter@lagerweij.com [SCRUMDEVELOPMENT]
2015-02-25 10:00:48 UTC
Permalink
Hi Hennadii,
Post by ***@gmail.com [SCRUMDEVELOPMENT]
DoD doesn't have a line about ALL bugs, but several to deal with bugs
depending on there severity (for example, only bugs of highest severity are
to be fixed as a part of work on PBI), so formally it can be said that a
PBI is Done according to DoD. So, it is usual thing to move bugs beyond the
sprint. As I now figured, this is the very goal of Triage: to deal with
severity so that all know what will be moved.
Many responses here say defect triage can be a useful thing, especially in
a legacy project. And allowing defects to live beyond the sprint they're
found in can be necessary. But allowing defects that you created in the
current sprint to live, and considering a story done while that is the case
is, in my estimation, a very different thing. By doing that you're
systematically allowing the total quality of your system to get worse every
sprint. There's no way I can see that that can end well.
Post by ***@gmail.com [SCRUMDEVELOPMENT]
DoD was initially suggested as a template by coach (included 'bug
segregation by severity' principle right from the very beginning) and
slightly adjusted later by the team.
Are you sure that the intention was to do that for defects found in the
work in progress? Even if it was, it was apparently interpreted differently.

Triaging is not a bad practice when needed. And in a legacy system, you can
find defects in existing code that are surfaced by new functionality.
Fixing those kind of issues can be a major effort, so it sometimes makes
sense to postpone that to a later sprint. But defects in new functionality
should not be allowed to escape the sprint. The cost of that is simply too
high in the mid to long term: you'll slow down too much.


Wouter
--
Wouter Lagerweij | ***@lagerweij.com
http://www.lagerweij.com | @wouterla <http://twitter.com/#!/wouterla>
linkedin.com/in/lagerweij | +31(0)6 28553506
Nancy Van Schooenderwoert vanschoo@acm.org [SCRUMDEVELOPMENT]
2015-02-23 23:37:13 UTC
Permalink
Hi -

I agree that bugs need to be prevented, and I also believe it is
possible to keep a codebase far cleaner than many think is possible.

Here is a great blog post by Elizabeth Hendrickson on the topic of bug
triage:

http://testobsessed.com/2012/08/bugs-spread-disease/


Here is where you can download a paper I wrote describing how my team
never had more than 2 open defects all thru a 3-year very challenging
embedded systems software project:

http://www.leanagilepartners.com/publications.html
Download the pdf file "Embedded Agile Project by the Numbers with
Newbies" from that page.


You've asked where bug triage fits in Scrum. If your sprints are
short enough and your code base not too buggy then few bugs will need to
interrupt the current sprint. In that case it's another backlog item to
be prioritized.

In working with legacy codebases, there is an art to balancing the
need to deal with urgent bugs and the need to deliver something
regularly enough to build trust with the team's stakeholders.

When working with teams to turn around a bad situation, I allow room
in each sprint to deal with urgent bugs. If none show up, then we
refactor and/or get some extra stories done.

There is the matter of bugs where you're not sure it's software or
hardware or something else. If an investigation has to be done, that can
be scheduled, and the results of it used to fix or reassign the bug.
With modern Agile dev tools, you can set up your environment so that
investigating the bug can be done with very little work.

Hope this helps!

- njv
Post by ***@gmail.com [SCRUMDEVELOPMENT]
Hello,
Recently I've come across a practice in the organization which has been there from pre-agile times. It is called bug triages. I'm struggling to grasp the role of this one in Scrum as I've never met in trusted sources any reference to 'bug triages'.
is this practice to be considered a part of product backlog refinement? won't bug triages meeting create another standing session in addition to 'official' ceremonies?
how does it fit in Scrum (by spirit and letter), are there any alternatives to 'bug triage' meeting? I would greatly appreciate all the opinions and thoughts!
--
............................................
Agile hardware? Yes! Agile safety-critical Embedded Systems too

Nancy Van Schooenderwoert, Lean-Agile Partners Inc.

US mobile: 781 301 1822 ***@leanagilepartners.com
Twitter: @vanschoo http://www.leanagilepartners.com
............................................
Ron Jeffries ronjeffries@acm.org [SCRUMDEVELOPMENT]
2015-02-24 00:55:12 UTC
Permalink
Right on!
R
Post by Nancy Van Schooenderwoert ***@acm.org [SCRUMDEVELOPMENT]
Here is where you can download a paper I wrote describing how my team
never had more than 2 open defects all thru a 3-year very challenging
Ron Jeffries
ronjeffries.com <http://ronjeffries.com/>
In times of stress, I like to turn to the wisdom of my Portuguese waitress,
who said: "Olá, meu nome é Marisol e eu serei sua garçonete."
Adam Sroka adam.sroka@gmail.com [SCRUMDEVELOPMENT]
2015-02-24 23:32:22 UTC
Permalink
It's not an uncommon practice. It is even a useful practice if you are in a
situation where you have more defects than you can possibly fix in a short
time box. Yes that is a bad situation to be in, but lots of teams are.

The metaphor comes from mass casualty situations where there aren't enough
doctors to treat everyone as they arrive. Certain patients are going to
have their outcomes significantly improve if they are seen quickly (e.g. If
they are bleeding.) Other patients aren't going to get significantly worse
if they have to wait. So, you prioritize the patients who are going to
benefit most from immediate attention.

I've seen defect tracking tools for legacy products that had tens of
thousands of reported defects. That is unmanageable. However, you also
can't just ignore them all and start doing Toyota style Stop the Line. You
need a bridge from one to the other.

One way to think of that bridge is to divide reported defects into
categories. Some defects are going to prevent customers from doing work
and/or seriously injure the business. Other defects represent mistakes in
understanding the stories or the language/tools. We generally want to fix
those, but can think of them like a special kind of PBI that gets planned
and prioritized. Finally, most legacy systems also have defects that likely
won't be fixed or whose fix will be obviated by some other planned change.

These things are not all the same. It's silly to pretend that they are.
It's a much better idea to create policies for each one and then start
analyzing and eliminating root causes until enough are eliminated that the
policy can be simplified.
Post by ***@gmail.com [SCRUMDEVELOPMENT]
Hello,
Recently I've come across a practice in the organization which has been
there from pre-agile times. It is called bug triages. I'm struggling to
grasp the role of this one in Scrum as I've never met in trusted sources
any reference to 'bug triages'.
- is this practice to be considered a part of product backlog
refinement?
- won't bug triages meeting create another standing session in
addition to 'official' ceremonies?
- how does it fit in Scrum (by spirit and letter),
- are there any alternatives to 'bug triage' meeting?
I would greatly appreciate all the opinions and thoughts!
Markus Gaertner mgaertne@gmail.com [SCRUMDEVELOPMENT]
2015-02-26 21:18:21 UTC
Permalink
You may also want to dive into the free available course material from Cem
Kaner's Black-box software testing course. The video material on Lesson 1
in the Bug Advocacy course covers bug triage - from a traditional tester's
point of view, not necessarily from an agile perspective. But this should
enlighten you on why it's used traditionally, and probably in place in your
current workplace.

Here's the whole set of material:
http://www.testingeducation.org/BBST/bugadvocacy/

Best
Markus

Scaling Agile as if you meant it: http://www.scaledprinciples.org
--
Dipl.-Inform. Markus Gaertner
Author of ATDD by Example - A Practical Guide to Acceptance Test-Driven
Development

http://www.shino.de/blog
http://www.mgaertne.de
http://www.it-agile.de
Post by ***@gmail.com [SCRUMDEVELOPMENT]
Hello,
Recently I've come across a practice in the organization which has been
there from pre-agile times. It is called bug triages. I'm struggling to
grasp the role of this one in Scrum as I've never met in trusted sources
any reference to 'bug triages'.
- is this practice to be considered a part of product backlog
refinement?
- won't bug triages meeting create another standing session in
addition to 'official' ceremonies?
- how does it fit in Scrum (by spirit and letter),
- are there any alternatives to 'bug triage' meeting?
I would greatly appreciate all the opinions and thoughts!
Loading...