The PO's job is to make educated guesses, not to create a definitive product definition. No matter how much work you do, you don't know that it's the "right thing" until you build it, deploy it to your customers, & get some feedback. Up until that point, it's just guesswork.— Allen Holub (@allenholub) September 17, 2020
Visualization isn’t the goal.— Cliff Hazell (@cliffhazell) September 17, 2020
It’s the conversations you have creating and using them, that are the most valuable
This morning, was working w/ a team collaboratively capturing user stories, and it occured to me that the techniques of #MobProgramming applied in spades in this context. #Mobbing is a general-purpose collaboration technique with applications way more diverse than programming. 1/— Allen Holub (@allenholub) September 16, 2020
Code is inventory— Steve Smith (@SteveSmith_Tech) September 16, 2020
Inventory is a liability, not an asset https://t.co/ozyLBQoeOb
Yes! Stories are descriptions of what your customers DO (in the domain) to solve their problems.— Allen Holub (@allenholub)September 18, 2020
They do not describe a computer program. https://t.co/qWaFmNNDZV
"A ll defects have a common source....— Tim "Agile Otter" Ottinger (@tottinge) September 18, 2020
... we write them."@mcarlson_sb
The agi le lesson is not "don't design." Rather, it's "design enough." When you start, design enough to get started. No more. As you work and learn, incrementally improve the design. A simple design is easier to refactor.— Al len Holub (@allenholub) September 25, 2020
I get s o tired of this BS revisionist history pushed by the snake-oil salesmen who sell Scrum. The origianl HBR article that introduced the term Scrum described a completely different way of work (that was better), and was an obscure article nobody knew about. 1/ https://t.co/GAQ6xGFIbz— Allen Holub (@allenholub) September 27, 2020
"Well, you can add it to the backlog, but that does not mean it will ever be prioritized."— Charles Lambdin (@CGLambdin) September 25, 2020
If your POs do not have the authority to say this to higher-ups, then they are not POs.
Putting people in a group doesn’t make them a team.— Adam Grant (@AdamMGrant) September 27, 2020
A team is a collection of people with a shared identity who collaborate to achieve a common mission. Each member makes a unique contribution.
Turning a group into a team starts with clarifying core values, goals, and roles.
Big stories hide lots of low-value junk. One of the best reasons to split/narrow stories into much smaller stories is that the splitting process uncovers the junk, which you can thow out. The most efficient way to speed development is to not develop low-value stuff.— Allen Holub (@allenholub) September 27, 2020
Metrics are NOT Pokemon.— Cliff Hazell (@cliffhazell) September 22, 2020
You don't need to "catch em all".
The goal is to Learn.
Not merely to have them.
Catching them, is expensive.
It's likely you need more learning, not more metrics.
The most important skill a programmer can have is conversation. We need to talk to each other, to our customers, to the business people who surround us in order to be even minimally effective. The tech stuff is important, but secondary to communication.— Allen Holub (@allenholub) September 28, 2020
You have 1M lines of code that nobody fully understands written in a language nobody uses any more leveraging libraries nobody knows and no unit tests. Do you actually think that it's not cost effective to replace it?— Allen Holub (@allenholub) September 29, 2020
The term Minimum Viable Product (MVP) has a lot of cr***p surrounding it. The phrase has been twisted around to the point where people think it means exactly the opposite of what it really is. It is NOT version 1. 1/2— Allen Holub (@allenholub) September 30, 2020
Do yourself a favor and spend 3 minutes with this video: https://t.co/efgJfEqIzz . If you have more than 3 minutes, read The Lean Startup. 2/2— Allen Holub (@allenholub) September 30, 2020
Math is inclusive. We prove things right. Science is exclusive. We prove things wrong.— Uncle Bob Martin (@unclebobmartin) September 30, 2020
Software is a science, our tests only prove our programs wrong.
We trust science because our test coverage is very high.
We can’t trust software with low test coverage.
“I want you to work together, on these Individual goals”— Cliff Hazell (@cliffhazell) September 30, 2020
Fundamental failings of modern management practice.
If we don’t know which bits are working, simply doing more is a waste.— Cliff Hazell (@cliffhazell) September 30, 2020
Doing less, unsurprisingly, allows time to reflect, and learn.
It’s usually only a few things that have the (largest) impact.
Take time to focus on them.
How to LEARN faster with the Feynman technique:— Richard Feynman (@ProfFeynman) September 30, 2020
> Pick a topic you wanna understand and start studying it
> Pretend to teach the topic to a classroom
> Go back to the books when you get stuck
> Simplify and use analogies!
Teaching a powerful tool for learning. 🧠