i love it when i find the integration process of DA with FLEX finds errors (yes, something wrong) in FLEX and not just improvements to it. Finding what's wrong with your approach is something to look at.
I tend to think of DA FLEX like I would a service. When I tell a service provider something's wrong I expect them to listen and say "oh, we'll fix that. Thanks for your input." 2/
I don't expect them to tell me "ah, you're just bashing." Or, "well we could do that, but then it wouldn't be our service."
The first means that they're not listening, the 2nd means i should look for another provider.
When you hear people tout the simplicity of an approach, it is often an indication that they've lost sight of the goal - the effectiveness of the approach.
how something is defined and how something is used may be quite different 1/2
Things often start simple (1 gear car), turn more complicated (multiple gears), then become simple to use (automatic transmission).
Use a DA consultant to help you figure out what's simple for you. Then use that. A little foresight goes a long way. 2/3
Maybe the understanding in "simple to understand, difficult to master" is the wrong understanding. We need to understand what works for us. Then it's not so difficult. Choice is power. Choice is fit for purpose.
when your problem is simple, basing a solution on specific practices may work. But when the problem is complex, more of a thinking and education approach is required.
our approaches should be about teaching us how to improve our results. Not presume following them provides us good results.
in '00 a framework within which to figure out what to do made sense because we didn't understand the theories under software development. Now we do. Using those theories directly is more effective. You can start with a set of often useful practices if you want to. then adjust
timeboxing/iterations are practices. not inherent to Agile. When you require using them you liimit your choices. What you want is: quick feedback, managed WIP, visibility, quality workflow, cadence. work on those directly - lean/flow/ToC provide guidance.
nothing wrong w/ timeboxing. However, you shouldn't drive from it. Drive from flow. Timeboxing's a practice that provides discipilne &structure that may be needed. When build Scrum on flow you realize you want short cycle times for stories. Can eliminate timeboxing when useful.
most companies I talk to have more problem on missing, unclear, changing requirements than with their teams. Focusing on the teams and not portfolio/product management is like trying to put a gasoline fire out from a leaking pump without stopping the pump from leaking.
an example of why we need more than empiricism. Imagine the theory underneath this about sub-atomic particles that you cannot observe. This theory is useful, but not observable - only (in)validatable with experiments.
Consider how this odd behavior can now be understood, how its implications (death by radiation poisoning) can be understood and dealt with (lead as a protective shield).
Now, consider a Scrum team process. If you get wierd behavior what do you do? Without theory you just have to stick to Scrum practices or insert other practices. But, since Scrum creates a psychological barrier to stay within its sopce, you may never find a way out. 3/
Trump's covid plan is apparently to let people get sick, and use his miracle cure. Of course, part of his plan is to eliminate insurance for people with pre-existing conditions. And he ignores the 10s of thousands of dollars in healthcare people are gong to have to pay.
this is a big issue. He's hoping all this fuss will have people ignore he owes almost half a billion dollars to people (likely the russians). And that he's trying to take away women's rights. Not to mention he's creating a white terrorist group to do his biding.
his gaslighting on ballots (no swing state sends ballots without request) takes our eyes of the GOP voter suppression. The environment, bounties, disrespect for our vets and many other things are being ignored due to cognitive overload.
Lean theory as to why Mob Programming works. Waste is directly proportional to the number of handoffs, handbacks and delays in workflow. Mob-programming eliminates all of these. The waste from these can represent 70-90% of the work being done.
@JoshBurdick72@PavelASamsonov which is part of why visibility is so important. and maybe, dare I say it, it might be useful to have a manager at a daily standup hearing "I'm blocked because I don't get enough time with a PO" Oh, but we can't do that - it'd be horrible.
@JoshBurdick72@PavelASamsonov what would have more impact to a manager?
1-hearing 3rd hand-via SM-that a team's having troubles or
2-seeing 1st hand a team is having troubles
this whole keep the manager out of the daily standup is an acccommodation to the fact that Scrum takes a negative attitude to managers
@JoshBurdick72@PavelASamsonov i'm not saying managers should be there either. I'm saying you have to do what works. and you can't prescribe that.
if you're smarter than someone who has had decades more training and experience than you then what are you not smartest in? Challenging experts is always good if you're trying to learn something. But recognize who is probably right while you're learning.
This goes for everywhere
the exception to this is when there's a mindshift involved. So when 5% disagree w/95%, one should look to see if the 5% are coming from a different place. If so, worth examing that diff. For example, Agile started out with a minority, but was based on different undrestanding
however, when two disparate groups are interpreting the same data with the same science and one side has 95% agreement, they are virtually always correct. When the 5% has been correct they've been able to make more accurate predictions as to what's going to happen.
I am really amazed how so many people just want to be right about masks. It's not a big deal to wear them. I don't care if it's right or not. It seems a more rational decision when risks are concerned. Not to mention those who've studied these diseases think it's best. ...
if you think you're smarter than someone who's studied something for decades then what are you not the smartest person about? This is a statement that you know more than anyone else about everything. It's incredibly arrogant and like a bull in a china shop. ...
there's no rationality here. It means you think what you think is the truth. Oh yeah, you can find evidence that meets your beliefs, but cognitive bias has it meet that. That's still you saying you can interpret data better than an expert. ...
there is more cause &effect in complex systems than there is in simple or complicated systems. It's just that the complex events that occur attract our attention. writing a blog now. this is very important insight. I've been saying this for 2 decades by not this clearly before
there are 4 types of cause and effects: 1) simple (obvious) 2) complicated (straightforward, but requires some thinking) 3) complex (can't been seen, made worse by coupling) 4) chaotic
The recourse for types 3 & 4 are:
* feedback (to dampen affect)
* small steps
I'm not as interested in what type of system we have (all development is complex) as much as I am as to what type of cause and effect is present and how to compensate for bad reactions
i have always disliked agile's attitude that you flip the iron triangle. Writing something on that. Consider the iron triangle with the word "risk" in the middle. You affect time using flow. You address scope via MBIs and you address cost with Lean. 1/2
you drive from value by focusing on what you can deliver quickly. Doing this by limiting time is not the best way. Focusing on most important value to deliver first is. 2/2
Agile with iterations requires you to flip the iron triangle. Flow lean and toc can dissolve it while lowering risk
you don't get to simple by taking things away. There is no guarantee that a simple solution exists as a subset of a complicated one. You get to simple by first understanding your problem and seeing what is required and not going beyond that.
In fact I inclined to think of someone has made an overly complicated solution there is no reason to believe there is a simple solution embedded in it. Why would you expect this?
This also explains why taking a subset of a complicated solution is probably still complicated ...
forget the quality of the men involved, look at the issues: 1) COVID. Trump actively trying to suppress info, discouragin use of masks, not making good tests available. putting faith in vaccines that the scientists who are creating them don't have. Biden-pro masks, pro science 1/
jobs: Trump lost 4.7 million jobs. Even before covid Biden had gained more jobs in last 3 years than trump in his first 3.
future jobs: Trump looking at past technologies - will result in job losses. Biden looking ahead. Also acknowledges a transition is needed. 2/
environment - Trump hurting water supply and air. Not even looking for easy gains that would improve economy. Biden - wants to help environment and invest to use green tech to create jobs.
Social injustice - Trump doesn't think there is any. BIden wants to improve things 3/
Unbelievable statement by Trump. Blaming someone who isn't even in office for not doing what is his job.
Remember his conversation with Newsom where he demonstrated his opinion that he doesn't think people should wear masks.
also, i suspect i eat out a lot more than the prez. The watiers are great. When was the last time he went to a public restaurant? He has service in the white house, he owns his country clubs and can tell waiters what to do. 1/2
I am certain he just fabricated up this whole story. Would love him to name a restaurant he's been to.
Remember his claim that you needed a driver's license to buy grocerices? 2/2
@afhill@christineluc@cydharrell thx for asking. Let's go through the roles: 1) PO-team should know more about what's being built, but often doesn't. 2) UX-devs are notorious for building hard to use sw. Not saying a UX person shouldn't be available, but the degree of the need for one illustrates the lack of 1/
@afhill@christineluc@cydharrell dev empathy for the user. If a person acted like software does they'd be punched out or fired in 15 minutes - "oh, you mean 9A doesn't come before 10?" 3) SM-they facilitate process and ceremony but can't/don't provide coaching for how to get the job done. Devs aren't taking 2/
@afhill@christineluc@cydharrell responsibility for how they work because 1) they aren't trained in what good workflow would be 2) they are pressured by management to keep their heads down and work.k
I'm not saying we don't need comparable roles to PO,UX,SM,dev but much of what we have is because 3/
run into almost always the same problems. Much of my analysis was about the reliance on empiricism without any understanding of Leam (this was before Don R's brilliant book on Flow). Since then I've learned that empiricism combined with a framework that has 10/
immutable roles, events, artifacts and rules is a dangerous combination. But until later that I learned the inherent challenges of frameworks. OK, now on to SAFe. 11/
In '11 I met with Dean to see about us joining forces on whether there was an advantage in us joining forces. At the time, he and Net Objectives had done more successful large scale adoptions than anyone else. I have since found that one of his BMC, wasn't his, but was 12/
@KavehKalantar@snowded@scottwambler@Mark_Lines First of all, I’m not going to try to convince anyone out there who is skeptical about the PMI. I’ll just ask you to be open that the new PMI is different in several ways. First of, @SunilPrashara, our new CEO, is a visionary. I love working for him and he has as big a vision 2/
@KavehKalantar@snowded@scottwambler@Mark_Lines@SunilPrashara as I do, maybe bigger. I haven’t been able to say that about many people. Second, PMI’s recent acquisitions, Disciplined Agile and Net Objectives, are the only two popular approaches that are not based on frameworks and both are method agnostic. This is not a coincidence. 3/
one reason bad ScrumBut happens (e.g., we're not finishing our stories at the end of a sprint so let's go to Kanban) is that we're focused on a practice - completing a story - instead of flow itself (short cycle time, proper WIP). If we realized the right issues we'd do better 1/
doing better might be don't allow devs to start stories until testers have completed theirs, or go to 1 week sprints. Or, it may be to use kanban if that gets focus on completion. our focus should be on improving our methods, not improving how we follow an approach 2/
imagine you're doing this - attending to Flow, WIP, ... (a good thing) and someone injects into the conversation - "But then wed not be doing Scrum" would that be a good thing or not?
One reason boundaries around an approach can be detrimental 3/3
Been working on Inherent Simplicity and how to work with complex systems. Got question why my comment that one of biggest inventory/queues/delays causes waste is doing work in iterations. So that's one topic. But related is how frameworks based on practices create constraints /1
to use both from a "what's possible" point of view (especially when we're told not doing those practices means you're not doing the framework anymore) and what we attend to. Constraints in design are good actually, as long as they are the right constraints. Using constraints /2
based on practices (which are not generally applicable) means the constraints limit where your approach can work. But having constraints based on how the world works would be useful. So the thought is, let's do design thinking on our process. Our "client" is our way of working/3