still chaotic-good Profile picture
Jun 16, 2021 26 tweets 8 min read Read on X
@Rubberduck203 @agileschools @JonAlb @DPradko @paul_boos @ClayDowling @Attila226 I'll write a long thread w my knowledge of the background and the history, will I hope clarify a lot ... hoping it will reduce the general hostility around the topic - I expect you will see some surprises in there :)

can't now, but later
@Rubberduck203 @agileschools @JonAlb @DPradko @paul_boos @ClayDowling @Attila226 so here goes (taking a minute to warm up his fingers...) (and take off all the respondent names...
Possibly the most important thing to think about when looking at Scrum, XP and "Agile" now is that, as they started, they were very anti-management.

so when you say, Scrum/XP/Agile helps "management" oppress the workers, recognize that is how bad management coopted the ideas
side note 1/ i'm writing this stream of consciousness, so no back-editing (

side note 2/ I'm noted for saying,

"Your ideas will either be misquoted and misused, or ignored. And you don't get to choose."

Power seeks tools to retain power.

That agile got misquoted and misused only means it was successful, otherwise you wouldn't hear of it.

Ken Schwaber wasn't shy that his goal was to get rid of project managers (Jeff S. I don't know about). He wanted to throw them out the window, as a profession, basically.

So Scrum was not invented to help management or PMs.

XP same way. the original XP replaced the PM with a junior-level "Tracker" person, who went around every day and asked ppl how far done they were, that being the perception of what PMs did with their time.

Scrummasters were given zero power and zero votes on all issues.

The job of a SM was to run around the org getting things signed, keeping people away from the team, whatever needed so the ppl could "get work done".

Hopefully that establishes that any interpretations about Scrum helping mgmt press down are pointed in the wrong direction

Scrum struck a magnificent bargain in hostile territory:

Mgmt got 12 times/yr (1ce /mo) to change direction in any way they wanted;

Team got 1 mo of total quiet time w no interruptions or changes of direction to do heavy thinking/working.

No execs ever got a better deal.

so where does "commitment" come in:

Team got to announce what they could and couldn't do in a month, no management interference in their bid.

No dev team ever got a better deal.

only caveat was ...

only caveat was:

If you promise you can deliver this much by end of month, then do that. Honor your commitment.

you F up on your bid, you work overtime. Learn from it for next month. You F up every month, ur a dummy. Learn already.

since mgmnt was not allowed to interfere w the bids, it's only up to the team to decide and declare. If the team consistently overbids, that's on them.

Here's where it starts to get interesting ...

a) teams consistently overbid, even without mgmt pressure. They just do.

Kent Beck and Martin Fowler wrote down (don't know who started it) a brilliant strategy:

Yesterday's Weather - you can only bid what you completely last month/sprint/iteration.

Now you can't overbid repeatedly.

yeah you should do this

side note 3/ as a methodologist, I have a rule that all methodologists overdecorate their methodologies with untested things
me incl

I don't think they tested this commitment thing properly, to know that it works, imho it was a wishful thinking social program.

In XP 2nd ed, 2004, Kent Beck went down the same path, saying estimates should be in hours, not story points.

He explicitly said programmers should learn their profession and take responsibility for their bids.

i.e, grow up

That also didn't work 🤣

but was same thinking

So when was "sprint commitment" taken out?

I started w Scrum in 2008, took a class from Ken himself
He had at that time a ritua: if a sprint was going to be a failure and had to be aborted, everyone should lie down & have a screaming temper-tantrum on the ground.


But I asked him, based on my worries about the negative effects of sprint "failure", whether it was actually a good idea.

His answer shocked me:

"What would it mean to the team was told their work was a failure? How depressing would that be?"

so he knew, or sensed

I never taught sprint commitment, I'll tell you why in a mo...

It was a few revisions after that that "sprint commitment" vanished from the guide and was replaced by ever-softer tones of "do what you do, and learn during the reflection."

So if anyone brings it up, tell them they are 10 years out of date and need to catch up.

now for why I never taught it...

I do project interviews. Some teams tell me they like to overbid the sprint, some that they like to underbid.

note that they *like* it, not that management pressured them. Please.

The ones who like overbidding said "We never want to have to ask, "What should we do next?""

the ones who like underbidding say, "We don't like to fail. If we get done, we can always pull whatever's next."

To me, this is where the action is.

mmm, cuz i can't back edit, I'll put those 2 sentences together:

The ones who like underbidding say, "We don't like to fail. If we get done, we can always pull whatever's next."

The ones who like overbidding say "We never want to have to ask, "What should we do next?""
They don't care about "failure"

who prefers which on your team?

Notice all of this is all team-internal.

Good mgmt tries to engage the team in this dialog, get them to take ownership, get them to learn, participate.

Bad management uses it as a tool to beat people.
But that they'll do no matter what tools you give them.

Remember, Scrum was invented to function in hostile environments, it's a really interesting contract between hard-pushing execs and devs needing time to think.

hmmm, let's see what I left out....

lol who says you can't back-edit in Twitter hahahaha


• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with still chaotic-good

still chaotic-good Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @TotherAlistair

Sep 7, 2023
I've been watching this type-checked / dynamic language battle for <ow>years\h\h\hdecades</ow> here are my useless thoughts: I can't make head or tail of it.

I'm a non-type-checked guy from the get go. Early C with Lint, worked great for me. Never liked when C got into the
type checking business. Then REXX, Prolog, Smalltalk, Ruby.

Did learn Pascal. I refused to touch C++ from the very beginning and learned why over time.

since forever, I have been hearing that type-checking will "do something good" for me.

that it will reduces errors, do <stuff no doubt great stuff>.

At the same time, I watch people delivering fabulous apps in LISP, Prolog, Smalltalk, Ruby, Python, REXX, you name it dynamic languages.

so evidently there's some mismatch between the promised
Read 26 tweets
Jan 4, 2023
because you peeps are both fans, sharp readers and fair critics, I'll post here the pages I have so far for comment
This is 1st draft, so It shows what's in my head.

Comments on tech content or intention, etc welcome.

Probably can leave typos alone, so much will change.

tia ImageImageImageImage
pp 11-18 ImageImageImageImage
pp 19-26 ImageImageImageImage
Read 4 tweets
Apr 3, 2022
if you want some hard reading:
"Elements to a Theory of Software Development" (2016)…
as an experiment, I'll put the key sentences from there in a thread, slowly over days, so you only have to read one sentence a day, most days :) heehee.

Starts w a mouthful tho:

(what is a 'theory'?)

We would like a theory of software development
• (T1) explain why successful projects succeed
and failing projects fail,
Read 18 tweets
Aug 15, 2020
This is the sort of historical revisionism and weak reporting that seriously pisses me off. Sutherland did not invent Scrum in the 80s, Scrum is not the founder of agile, there were 17 people involved, etc etc. That HBR is passing this along is seriously bad history/reporting
ps wrote this in the article comments: This is the sort of historical revisionism and weak research that seriously pisses me off, me being one of the authors of the agile manifesto. Sutherland did not invent or use Scrum in the 80s, it was in the 90s, Scrum is not the founder...
of agile, there were 17 people involved, with diverse histories and origin stories. That HBR is passing this false history is seriously wrong. I have to write to correct you. This article is online, so Yes you can amend it. Do some research please. Alistair Cockburn, Co-author..
Read 4 tweets
Apr 26, 2020
@tpierrain @JuanMGarridoPaz @DDDesignFR @duchessfr @DorraBartaguiz @Audrey_Neveu @estelandry @JeromeAvoustin @lucas_courot @JulienTopcu @Lilobase @BodySplash @pauline_io @jgrodziski @brunoboucard I had a LONG conversation w Thomas about this
Two things showed up.
1) he is writing a data-driven meta-model application. The inputs are data defining the data that will be received. So he needs a module that catches model inconsistencies... he calls this an adapter (i don't)
@tpierrain @JuanMGarridoPaz @DDDesignFR @duchessfr @DorraBartaguiz @Audrey_Neveu @estelandry @JeromeAvoustin @lucas_courot @JulienTopcu @Lilobase @BodySplash @pauline_io @jgrodziski @brunoboucard I consider it part of the domain (maybe), for sure not part of "the technology of the world that will be swapped out".

however, we are not done with the discussion, nor do I have a modified, better arch to show him .. :))

2) more significantly...
@tpierrain @JuanMGarridoPaz @DDDesignFR @duchessfr @DorraBartaguiz @Audrey_Neveu @estelandry @JeromeAvoustin @lucas_courot @JulienTopcu @Lilobase @BodySplash @pauline_io @jgrodziski @brunoboucard 2) the economics and pragmatics of writing the same tests at two very close interface points: just inside the adapter, and just outside, are unfavorable. Maybe 1 person can do that, but a whole team, unlikely ...

... understanding the pragmatics doesn't change the pattern,

Read 6 tweets
Sep 3, 2019
Happy to tell the story, @Heart_of_Agile supports Mónica Delsouc reaching the summit of Elbrus, highest mountain in Europe:
(thread w pics)
Monica (from Mendoza, think Malbecs you love) had already climbed Aconcagua, the highest peak in North and S America. I decided to support her, she, already a supporter of the #HeartOfAgile, offered to carry the Heart of Agile banner to the top.
This pic St Pete Beach, FL
Here in Mendoza airport, with family and friends, getting ready to go.
Read 8 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!


0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy


3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!