jamais sans mes scrumbuts
DESCRIPTION
slides supporting discussion & workshop for 2013 ScrumDay session @ IBM Paris Good or Bad ScrumButs? What can we learn?TRANSCRIPT
Skype://[email protected] Twitter: @ckeromenwww.ckti.com
Christophe Keromenhttp://fr.slideshare.net/ckti/
Jamais sans mes ScrumButs
1
LE FSUG remercie ses sponsors pour leur soutien
2
Présenta1on de la séance
Défini1on et exemples de ScrumButs
recueil d'autres exemples (6 mn)
deux tas (1 mn)
1) ScrumBut temporaire
2) ScrumButs assumés
La voix de son maître : la posi1on de Ken Schwaber
LeGer-‐Vo1ng (2 mn)
Scrum (S)
pas Scrum (!)
Objec1f Scrum ?
Mais Inspect & Adapt, agilité
Quel nom donner ?
ScrumBut = Changer Scrum pour ne pas changer
Devia1on is bad devia1on
Changer le contexte
Comment les ScrumButs transitoires ont été dépassés ? (10 mn)
De bons ScrumButs?
Pragma1sme
Paradoxe
Les bons et les méchants?
Comment les ScrumButs assumés ont apporté de l'améliora1on ? (10 mn)
Comment discriminer ?
Scrum ou pas Scrum (J Sutherland)
Risque -‐ Caractérisa1on (agileforest)
Diagramme d'effets (Laurent Bossavit)
Étude de cas
Ensemble
Discussions
Pour aller plus loin
Scrum Checklist
ScrumBut Test
3
ScrumButs• http://www.scrum.org/scrumbut
• (ScrumBut)(Reason)(Workaround)
4
scrumbut [skruhmbut] noun.
1. A person engaged in only par1al Agile project management or development methodologies
2. One who engages in either semi-‐agile or quasi-‐waterfall development methodologies.
3. One who adopts only SOME tenants of the SCRUM methodology.
4. In general, one who uses the word "but" when answering the ques1on "Do you do SCRUM?"
5
TimeBox : 3+3 mn
Recueil de ScrumButs sur feuilles A4
6
TimeBox : 1 mn
Deux collections :
ScrumButs transitoires
ScrumButs assumés
7
Nous u1lisons Scrum, mais
nous avons besoin d’une grosse équipe pour avoir toutes les compétences indispensables au projet.
Ainsi nos équipes sont + nombreuses que les 9 personnes recommandées.
ScrumBut
8
We use Scrum, but
having a Daily Scrum every day is too much overead,
so we only have one per week.
ScrumBut
9
We use Scrum, but
Retrospec5ves are a waste of 1me,
so we don't do them.
ScrumBut
10
We use Scrum, but
we can't build a piece of func1onality in a month,
so our Sprints are 6 weeks long.
ScrumBut
11
We use Scrum, but
we have a list of bugs to fix,
we do two normal sprints and one bugfix sprint.
ScrumBut
12
Nous u1lisons Scrum, mais
des changements d'exigences doivent être prises en compte fréquemment,
aussi le périmètre du Sprint change en cours de Sprint.
ScrumBut
13
We do Scrum, but
we do not have anyone available to serve as a product owner
so we communicate by email, no ordered backlog
ScrumBut
14
We do Scrum, but
we cannot split our stories
so we cannot finish them in one sprint
ScrumBut
15
We do Scrum, but
hard precise deadlines are defined by management with hardened scope
so we just code
16
We do Scrum, but
we need to include/provide items from/to another team
so we just change the sprint dura5on from sprint to sprint
we do all our planning up front
ScrumBut
17
We do Scrum, but
we need extra-‐material to demo
so we demo items on our local environment and you cannot play with it
ScrumBut
18
We do Scrum, but
we are always saying the same things in our Retrospec5ves
so we don't do them anymore
ScrumBut
19
ScrumButs can come from many sources
• The business doesn’t want to be involved
• Everyone wants their features first and can’t agree on a priority
• Teams don’t know how to self-‐organize
• People aren’t available to work on teams full-‐1me
• Timeboxes arent adhered to
• Teams don’t see a need for a daily Scrum
• Teams can’t get a piece of func1onality done in one Sprint
• Teams don’t have the skills to “do” something
hGp://fr.slideshare.net/tangram77/scrum-‐buts-‐but-‐scrum
20
ScrumButs can come from many sources
• Teams can’t fit tes1ng into the same Sprint as development
• The Scrum Master tells the team what to do and how to do it
• Other managers can’t stay out of a Sprint
• Important things come up that require interrup1ng the Sprint
• The Sprints can’t start un1l all of the other groups do their up-‐ front work
• Other groups are building hardware or using waterfallhGp://fr.slideshare.net/tangram77/scrum-‐buts-‐but-‐scrum
21
“Each component within the framework serves a specific
purpose and is essen1al to Scrum’s success and usage.” —
Scrum Guide
"Les ScrumButs constituent les raisons qui empêchent de tirer parti de Scrum pour résoudre les problèmes et réaliser les bénéfices."
Ken Schwaber - Scrum.org
22
“Scrum is like chess.
You either play it as its rule
s state, or you don’t.”
K. Schwaber
http://agiletrail.com/2013/03/27/blasphemy-they-call-it-scrum/
You can’t call it Scrum if you’re not doing it by the book.
23
Scrum’s roles, artifacts, events, and rules are immutableand although implementing only parts of Scrum is
possible, the result is not Scrum. Scrum exists only in its entirety and functions well as a
container for other techniques, methodologies, andpractices.
24
• Do we really need a Scrum Master?
• Is it OK to have a part-‐1me Product Owner?
• Can we split people across Teams?
• Can we work on several projects at once?
• Can we have itera1ons for doing design and analysis?
Are We Doing Scrum If...
hGp://agileatlas.org/commentaries/ar1cle/are-‐we-‐doing-‐scrum-‐if25
TimeBox : 2 mn
"Letter-Voting" sur chaque ScrumBut
Scrum (S)
pas Scrum (!)
26
Objectif Scrum ?BUT Inspect & Adapt?
27
Retrospec1ve (2001: Norm Kerth)
Scrum evolution
hGp://www.mountaingoatsovware.com/blog/gasping-‐about-‐the-‐product-‐backlog
28
• The priori5sed product backlog was replaced by an ordered product backlog
–hGp://www.scrumalliance.org/ar1cles/367-‐its-‐ordered-‐-‐not-‐priori1zed
• The term “commitment” regarding the work selected for a sprint was ditched in favour of the term “forecast“.
–hGp://www.scrum.org/About/All-‐Ar1cles/ar1cleType/Ar1cleView/ar1cleId/95/Commitment-‐vs-‐Forecast-‐A-‐subtle-‐but-‐important-‐change-‐to-‐Scrum
Scrum deviations (2011)
29
Alors ?
Dévoiement de l'agilité ?
Adaptation agile ?mais
30
Names and what they refer to are important for
communication.
Si chacun utilise un autre nom pour sa déclinaison de Scrum,
comment va-t-on communiquer ?
mais
31
ScrumBut= changer Scrum pour ne pas changer ?
32
Mike Cohn
"To me a ScrumBut is a devia1on from the minimal core set of Scrum prac1ces that is obviously a bad devia1on"
33
ScrumBut => Scrum mal appliqué => Scrum ne marche pas => L'agilité c'est nul
scrum-fall
it’s waterfall with a few Scrum practices woven in.
34
Ces questions révèlent des problèmes dans votre
organisation
35
Rupture ou petits pas ?
http://www.agilex.fr/2013/01/demarrer-avec-scrum-ou-kanban/36
Le contexte• Vous ne pouvez pas changer le contexte avant de rechercher une améliora1on agile ?
–commencez là où vous en êtes
–ne pas se sa1sfaire du statu-‐quo–remontez les problèmes, rendez visible
• transparence–inspectez & adaptez
37
TimeBox : 10 mn
Comment les ScrumButs transitoires ont-ils été
dépassés ?
38
De bons ScrumButs?
39
Jurgen Appelo
IMHO it is the perfect term for overzealous SM's who focus on process instead of people and agile values.
40
Jurgen Appelo
• The real op1mal method always depends on the team's structure and environment.
• In short: the op1mal method will always be an adapta1on...
• And that's what retrospec5ves are for!
41
Dennis Stevens
The goal is for the team to be able to iden1fy the capabili1es and outcomes that are most important to improving the teams delivery
42
Alan Atlas
my philosophy about scrum: do it well, give it a good try, but don’t kill yourself and if it’s not perfect, it’s s1ll good.
hGp://agileatlas.org/commentaries/ar1cle/what-‐scrum-‐means-‐to-‐me
43
Pragmatisme
44
Unless you are lucky enough to work for a company where top management supports Scrum, middle management also supports Scrum, and the whole company or IT department has been completely reorganized for Scrum, your project or company
environment may not be as perfect as Scrum prescribes.Scrum In Action, chap. 12
http://www.amazon.com/Scrum-Action-Andrew-Pham/dp/143545913X
45
Church volunteers had their daily stand up only once every Friday when they came to church because they
did not come to the church every day.Arline C. Sutherland "Scrum in the Church"
http://docs.google.com/viewer?url=http://jeffsutherland.com/scrum/SutherandScruminChurchAgile2009.pdf
46
http://commons.wikimedia.org/wiki/File:110228-F-3188G-115-C-130H3.jpg
http://commons.wikimedia.org/wiki/File:ShuHaRi.png
“Individuals and interactions over processes and tools”http://en.wikipedia.org/wiki/Cargo_cult
47
ButScrum• But Scrum says we es1mate
–I have personally been around agile teams that dont es1mate all of their items in their backlog. Why? Mul1tudes of reasons.
• A couple of examples: –Teams break down their work far enough (day range) where the variance in the es1mates would be minuscule.
–A team works really well together and can accurately predict how much they can get done, without worrying about points
hGp://fr.slideshare.net/tangram77/scrum-‐buts-‐but-‐scrum
48
Les bons et les méchants ?
49
Positive ScrumButsThey make a team's performance better.
AdaptationExperimentation
Innovation
50
(Positive). Good adaptation of Scrum that helps get the team moving forward despite some constraints they
have to deal with in the real world.
51
An example for a good deviation is the replacement of the Daily Scrum with a standup every 1.5 hours, because all of the intentions behind the original element are still (and even better!) met with the use of the new element.
http://agiletrail.com/2013/03/27/blasphemy-they-call-it-scrum/52
Negative ScrumButsThey make a team's performance worse.
Bad excusesWrong application
53
The deviation is bad, when one or more of the intentions behind the original element are not met anymore.
For example, replacing a single person PO by a PO board is a bad deviation, because one of the intentions behind the original element (fast decision making) are not met anymore.
http://agiletrail.com/2013/03/27/blasphemy-they-call-it-scrum/54
TimeBox : 10 mn
Comment les ScrumButs assumés ont apporté de
l'amélioration ?
55
Comment discriminer ?
56
a “C” if it is caused by Scrum (in other words, without Scrum it would not be
happening), or an “E” if it is exposed by Scrum
(in other words, it would be happening with or without Scrum, but Scrum makes it known to the Team),
or a “U” if it’s unrelated to Scrum (like the weather).
http://jeffsutherland.com/ScrumPapers.pdf
57
• Pourquoi cette règle Scrum ?
• Ken Schwaber expose un principe
• Quels sont les risques à ne pas la suivre ?
• Ken Schwaber utilise le mot Conséquence
Scrum is making problems visible and requires you to
change your behavior and way of thinking
Ken Schwaber http://www.youtube.com/watch?v=tgBkvS-q_fA&lr=1
http://www.youtube.com/watch?v=-MZdKaqY6wU&lr=1
58
59
60
what (approach)
why (purpose/intent)
who (is involved and to what extent, RACI is a good model for this)
when (how often, how does the time impact other elements)
how (does it feel when it is working right versus working poorly)
http://agileforest.com/2013/03/29/scrum-intentions-scrumbut-scrumand-and-shu-ha-ri/
http://racichart.org/
61
http://commons.wikimedia.org/wiki/File:RACI_Matrix.png
The RACI Matrix shows responsibilities for a task or set of deliverables (i.e. who is Responsible for doing the work; who will beAccounted for it; who will be Consulted for it; and who will be Informed
once it has been completed).
62
Laurent Bossavit:
each agile prac1ce is an interven1on, discovered through real-‐world experience of veteran developers, which repairs a frequently encountered, dysfunc1onal system structure in sovware development.
Building your own agile process
hGp://www.scandevconf.se/db/BYOAP.pdf
63
Diagrammes d'effets
hGp://www.scandevconf.se/db/BYOAP.pdf64
Résolution de problèmes• 5 why, Ishikawa (Fishbone)•A3 thinking
–hGp://www.crisp.se/gra1s-‐material-‐och-‐guider/a3-‐template
•…
hGp://agilitrix.com/2010/06/go-‐faster-‐with-‐root-‐cause-‐analysis/65
hGp://scrumorakel.de/blog/index.php?/archives/34-‐Can-‐you-‐change-‐Scrum.html
Do Retrospective!
Whatever they want to change, I never allow them to drop retrospec1ves and analyze with them what happened aver they altered something
66
Escalader
Les ScrumButs sont des signaux d'obstacles organisa1onnels.
Ken Schwaber conseille de créer une équipe dédiée pour les traiter : Enterprise Transi5on Team.
Son objec1f est de supprimer les obstacles remontés par les équipes Scrum.
67
Escalader
Le ScrumBut est-‐il dépendant de l'équipe ?
si oui
l'ajouter au Backlog de l'équipe
sinon
l'ajouter au Backlog de l'Enterprise Transi1on Team
68
Étude de cas
69
• Our Product Owner is an arrogant son of none who won't listen to anyone and who pushes everyone around. Our retrospec1ves just wind up with the PO shou1ng at everyone and forcing the same outcome every 1me, work harder and quit whining.
• It is s1ll Scrum if we exclude the Product Owner from our Sprint Retrospec1ves?
Ron Jeffries: hGp://agileatlas.org/commentaries/ar1cle/is-‐it-‐s1ll-‐scrum-‐if
70
• What is the rule?
• What is the goal of the rule?
• What is the organiza1on issue?
• What is the risk of breaking the rule?
• How can we s1ll target the goal?
71
Yes, the Product Owner needs to be in the Sprint Retrospective or you're not doing Scrum.
72
• What is the rule?
–Sprint Retrospec1ve is for the whole Scrum Team
• What is the goal of the rule?
–best results come from 1ght connec1on between Product Owner and Dev Team, with true collabora1on and coopera1on.
–"Business people and developers must work together daily throughout the project."
–"At regular intervals, the team reflects on how to become more effec?ve, then tunes and adjusts its behavior accordingly."
• What is the organiza1on issue?
–Command-‐and-‐control paGern, no full understanding of Scrum…
• What is the risk of breaking the rule?
–No Working Retrospec?ve
73
How can we still target the goal?
74
• SM teaches the PO
• Get a # PO
• new mee1ng Dev Retrospec1ve without PO
–ScrumBut
• …
75
Discussions…
76
Pour aller plus loin
77
Jeff Sutherland
ScrumBut test
http://www.dennisstevens.com/2009/03/05/were-doing-scrum-but/
http://www.dennisstevens.com/2009/09/01/methods-practices-and-outcomes/
traduction française par Antoine Vernoishttp://antoine.vernois.net/scrumbut/?page=test&lang=fr
78
• No Itera1ons – 0• Itera1ons > 6 weeks – 1• Variable Length < 6 weeks – 2• Fixed itera1on length 6 weeks – 3• Fixed itera1on length 5 weeks – 4• Fixed itera1on length 4 weeks or less – 10
Question 1 –
79
•Limit Work in Process – don’t start anything you can’t finish
80
• No dedicated QA – 0• Unit tested – 1• Feature tested – 5• Feature tested as soon as completed – 7
• Sovware passes acceptance tes1ng – 8• Sovware is deployed – 10
Question 2 – Testing
81
•Tes1ng within the Sprint: Maintain Code Quality – Shorten quality feedback loops to zero to maintain code quality and reduce the waste of rework
82
• No requirements – 0
• Big requirements documents – 1
• Poor user stories – 4• Good requirements – 5
• Good user stories – 7• Just enough, just in 1me specifica1ons – 8
• Good user stories 1ed to specifica1on as needed – 10
Question 3 – Agile
83
•Communicate Business Needs to Development – Ensure developers understand outcomes while minimizing coordina1on and transac1on costs, maintain traceability to business need
84
• No Product Owner – 0• Product Owner who doesn’t understand Scrum – 1
• Product Owner who disrupts team – 2
• Product Owner not involved in team – 2
• Product Owner with clear product backlog es1mated by team before Sprint Planning mee1ng (READY) – 5
• Product Owner with a release roadmap with data based on team velocity – 8
• Product Owner who mo1vates the team – 10
Question 4 – Product Owner
85
•Maintain Product Roadmap – The business communicates a plan of what is being built
86
• No Product Backlog – 0
• Mul1ple Product Backlogs – 1
• Single Product Backlog – 3
• Product Backlog clearly specified and priori1zed by ROI (business value) before Sprint Planning (READY) – 5
• Product Owner has release plan based on Produce Backlog – 7
• Product Owner can measure ROI (business value) based on real revenue, cost per story point, or other metrics – 10
Question 5 – Product Backlog
87
•Priori1ze Investments Based on Return – the business priori1zes work based on a current understanding of business value
88
• Product Backlog not es1mated – 0
• Es1mates not produced by team – 1
• Es1mates not produced by planning poker – 5
• Es1mates produced by planning poker by team – 8
• Es1mate error < 10% – 10
Question 6 – Estimates
89
•Provide Meaningful Effort Es1mates – Understand rate the team can produce work to support planning, investment decisions, and customer commitments
90
• No burndown chart – 0
• Burndown chart no updated by team – 1
• Burndown chart in hours/days not accoun1ng for work in progress (par1al tasks burn down) – 2
• Burndown chart only burns down when task is done (TrackDone paGern) – 4
• Burndown only burns down when story is done – 5
–Add 3 points of team knows velocity
–Add two points if Product Owner release plan based on known velocity
Question 7 – Burndown Chart
91
• Communicate Release Schedule – Be able to predict content and 1ming of future releases. This clarifies correc1ve ac1on, next most valuable investment decisions, marke1ng decisions and customer commitments
92
• Manager or Project Leader disrupts team – 0
• Product Owner disrupts team – 1
• Managers, Project Leaders or Team Leaders telling people what to do – 3
• Have Project Leader and Scrum roles – 5
• No one disrup1ng team, only Scrum roles – 10
Question 8 – Team Disruption
93
•Maintain Produc1ve Work Environment – Management understands and supports the focus on rapid delivery. Trust is established that the team will be able to meet current and future commitments.
94
• Tasks assigned to individual during Sprint Planning – 0
• Team members do not have any overlap in their area of exper1se – 0
• No emergent leadership – one or more team members designated as a direc1ve authority – 1
• Team does not have the necessary competency – 2
• Team commits collec1vely to Sprint goal and backlog – 7
• Team members collec1vely fight impediments during the sprint – 9
• Team is in hyperproduc1ve state – 10
Question 9 – Team
95
• Develop an Empowered Team – The team feels empowered to make decisions about how to move forward. Management has provided sufficient guidance and direc1on that they trust the team will make opera1onal and tac1cal decisions aligned with the best interest of the business.
96
• If you score less than a 6 on average, you probably aren’t doing Scrum.
• if you aren’t doing Scrum, don’t call it Scrum
97
Scrum Check-ListCheck-‐list Scrum non-‐officielle par Henrik Kniberg : faites-‐vous vraiment du Scrum ?
hGp://www.crisp.se/gra1s-‐material-‐och-‐guider/scrum-‐checklist
• hGp://www.fabrice-‐aime�.fr/dotclear/index.php?q=mindmap
• hGp://thierrycros.net/public/docs/Evalua1on_Scrum_-‐_v1.ods
98
GASPsGenerally Accepted Scrum Prac5ces
• User Stories: hGp://agileatlas.org/gasps/ar1cle/user-‐stories
• Release Sprint: hGp://agileatlas.org/gasps/ar1cle/release-‐sprint
• Starup Sprint: hGp://agileatlas.org/gasps/ar1cle/startup-‐sprint
hGp://agileatlas.org/gasps
99
Scrum Alliance: Agile Atlas
• Ron Jeffries: hGp://agileatlas.org/commentaries/ar1cle/is-‐it-‐s1ll-‐scrum-‐if
100
• hGp://www.noop.nl/2009/09/scrumbuts-‐are-‐the-‐best-‐part-‐of-‐scrum.html
• hGp://www.dennisstevens.com/2009/03/05/were-‐doing-‐scrum-‐but/ (Jeff Sutherland ScrumBut Test)
• hGp://jeffsutherland.com/SutherlandScruminChurchAgile2009.pdf
• hGp://prezi.com/4jap4v487ykg/scrum-‐but/
• hGp://agileforest.com/2013/03/29/scrum-‐inten1ons-‐scrumbut-‐scrumand-‐and-‐shu-‐ha-‐ri/
• hGp://agiletrail.com/2013/03/27/blasphemy-‐they-‐call-‐it-‐scrum/
• Laurent Bossavit : Building your own agile process hGp://www.scandevconf.se/db/BYOAP.pdf
• hGp://fr.slideshare.net/tangram77/scrum-‐buts-‐but-‐scrum
• hGp://pierrefauvel.wordpress.com/2013/03/04/583/
Ressources
101
hGp://crea1vecommons.org/licenses/by-‐sa/3.0/
This presenta1on was inspired by the works of many people, and I cannot possibly list them all. Though I did my very best to aGribute all authors of texts and images, and to recognize any copyrights, if you think that anything in this presenta1on should be changed, added or removed, please contact me at [email protected].
102