This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2020-06-25
Channels
- # ask-the-speaker-track-1 (280)
- # ask-the-speaker-track-2 (292)
- # ask-the-speaker-track-3 (326)
- # ask-the-speaker-track-4 (212)
- # bof-arch-engineering-ops (11)
- # bof-covid-19-lessons (18)
- # bof-cust-biz-tech-divide (3)
- # bof-leadership-culture-learning (3)
- # bof-next-gen-ops (1)
- # bof-overcoming-old-wow (2)
- # bof-project-to-product (7)
- # bof-sec-audit-compliance-grc (4)
- # bof-transformation-journeys (2)
- # bof-working-with-data (3)
- # discussion-connect-february (1477)
- # games (131)
- # games-self-tracker (9)
- # happy-hour (38)
- # help (85)
- # hiring (11)
- # networking (24)
- # snack-club (22)
- # sponsors (41)
- # summit-info (263)
- # summit-stories (17)
- # xpo-datadog (1)
- # xpo-digitalai-accelerates-software-delivery (9)
- # xpo-github-for-enterprises (9)
- # xpo-gitlab-the-one-devops-platform (14)
- # xpo-itrevolution (1)
- # xpo-launchdarkly (5)
- # xpo-planview-tasktop (1)
- # xpo-slack-does-devops (5)
- # z-do-not-post-here-old-ask-the-speaker (2)
@mik and @genek is it possible that because of the pace of technology we are in both the end of the transition of the information age and in the installation phase of the digital age?
@mik and @genek is it possible that because of the pace of technology we are in both the end of the transition of the information age and in the installation phase of the digital age?
FYI, most of the talk is in #ask-the-speaker-keynote
The Perez book makes it clear that the ICT Age (which @mik wonderfully renamed Age of Software and Data) began in 1970s. So itβs all really one age. Her assertion is that weβre in the middle of the Turning Point. Prediction is that 20-50 years of prosperity associated with the Deployment Period is next. Iβd check out the Jerry Neumann blog post and podcast.
Engage with @mik and @genek101 in #ask-the-speaker-keynote!
Looking forward to this @natasha.anderson
Make All the Things Continuous! Yes!
@natasha.anderson how did you differentiate between the the practice view with that of the customer/product teams?
@natasha.anderson how did you differentiate between the the practice view with that of the customer/product teams?
Hi Matt, over simplified, the practice is responsible for defining what good is, creating platforms for the product teams to share and work with each other. The product teams, constructed of feature and system teams, are responsible for delivery.
We also have the responsibility of defining the Engineering Standards for the Group
Thank you would you also sometimes help in the building of those capabilities? Also where applicable helping to enable the outcomes through organization conversations
Yes we do, my team are technical coaches, we either engage adhoc on short term quick consultations, or longer 10week deep dives
Thatβs pretty much we have in Nationwide; WoW playing the role of practice and then we got Squads responsible for their products. We deliver the CI/CD Pathfinder MVP which runs upto10 to 12 weeks, depending on VSM and outcome.
gosh yes! The amount of βare we done yet with all this uncomfortable βPeople Practiceβ obsession of yours now then?!β We get every week:joy:
@8mckeem8 Team leaders that get in our Dashboard then glare βl tried but it wasnβt like an Apple product, looked like l had to do work on that data. Then some work again afterβ
@8mckeem8 Team leaders that get in our Dashboard then glare βl tried but it wasnβt like an Apple product, looked like l had to do work on that data. Then some work again afterβ
No one but the team sees it - Bubble level only in out case
@natasha.anderson letβs do a show me your call - canβt see if you measure Psychological Safety but dashboard and measurements comparing alone should be fun
is about creating communities and connections. I also spend alot of time establishing working groups
*show me yours-lβll show you mine that is (mustnβt multitask:see_no_evil:)
How did you manage to get the buy in from product teams (which I assume had embedded engineers from your practice)
@natasha.anderson so refreshing to hear on your transformation. Banks especially in Africa are considered dinosaurs when it comes to technology
@natasha.anderson so refreshing to hear on your transformation. Banks especially in Africa are considered dinosaurs when it comes to technology
I am from Cellulant and work with banks across the continent on payment aggregation. I cannot tell you how good it feels to hear about your journey
its very high levl here, if you would like to see a bit more let me know, we are only to happy to share
Yeah I would. Especially where you started coz you must have a large Dev team and lots of apps
Its defined as part of the organisational architecture and we are part of the Group's CTO
@natasha.anderson... it makes me proud to see you representing the Bank so positively! I have seen first-hand how the Bank has led the DevOps movement in South Africa. Even more so, you had the courage to move to a growth mindset and be led by metrics and culture. Your leadership team has been exemplary.
@natasha.anderson... it makes me proud to see you representing the Bank so positively! I have seen first-hand how the Bank has led the DevOps movement in South Africa. Even more so, you had the courage to move to a growth mindset and be led by metrics and culture. Your leadership team has been exemplary.
see also "leadership is about how you make people feel" from Jon Smart's talk
Hi, i will have to get some history on you for that. it was part of the journey before i got there. We do have a mixed bag of tools so right now while we have Bamboo, we also have Azure DevOps
@natasha.anderson by the way, I really love your choice of a Black mural artist at work for your background image on Leadership
@natasha.anderson by the way, I really love your choice of a Black mural artist at work for your background image on Leadership
John Adams is phenomenal! Have one of his in my study π Let me share
Please do! May I also recommend Sonia Jones? She's a former colleague and has done a couple of murals in DC recently. https://www.instagram.com/soniajonestheartist/
At my company we've had a 40 hour mandated minimum target for career learning and development called 'Think 40' and we reward learners for exceeding that target. These last three days here have been some of the most important and influential I've ever learned from in my multi-decade career so far, so thanks to my boss and colleagues for enabling me to attend. Many thanks @natasha.anderson for sharing a truly inspirational presentation π
Thanks @natasha.anderson - that was very inspiring and relatable. Thanks for sharing π
@natasha.anderson Great talk and valuable lessons for leaders and all DevOps enthusiasts
Thank you @natasha.anderson! Welcome @suzette.johnson5 and @robinyeman!
Thank you @natasha.anderson! Welcome @suzette.johnson5 and @robinyeman!
Pleasure, thank you all! i'll be off but feel free to contact me on LinkedIn https://www.linkedin.com/in/n-natasha-anderson-a4a0222b/
Ruth is amazing.
So many gems in https://ruthmalan.com/Journal/2019/20190629SlideDocTechnicalLeadershipDecisions.pdf
(CC @genek101 - Ruth Malan would be an incredible speaker for DOES some time)
Culture change can be slow. Where did you start when changing the culture and why?
erm... I just taught from SAFe slides that look spookily similar 4 weeks ago π
@robin.yeman, @suzette.johnson5 "The context of the system is amplified in when you consider culture" - v useful!
WRT culture I use a couple change models as a foundation as I put together the cultural roadmap; such as ADKAR and John Kotter @linyd
Strangler pattern πΊ
Yes we have also been working with SAFe on their upcoming Agile for hardware class
Yes. understanding the value stream is important in understanding flow and where or how can remove dependencies
I really like the explicit approach to addressing Culture using the tools like roadmaps that we already know
@suzette.johnson5 is the standard Value Stream Identification toolkit of safe enough or needs more tools to address the cultural aspect?
We start with the VS identification and to your point we then elaborate and dig deeper into VSM
Thank you @robin.yeman and @suzette.johnson5 - it was very good to hear about how devops approaches can be used in systems that span hardware and software, and which have safety critical elements - v useful! ππ
Such a great quality video @ track1. Good job Kira Barclay and John Deere!
High quality talk done faster than expected. The John Deere story checks out π
@andrew.salt - yes - will export the Mural board and pop it in this channel
2This bit of the Unicorn Project is my fave too: elevate the practice of improving daily work over daily work itself.
This bit of the Unicorn Project is my fave too: elevate the practice of improving daily work over daily work itself.
I believe this but yet find the ability to keep pushing back against the flood of external asks and management priorities.
Improvement (βdebtβ) budget formally agreed at a high level and tracked as a balanced KPI to senior management, I think is the only way to manage the incentives. Itβs the only one Iβve found.
Easier said than done :-/
@dominica In Mural you have the Outline function. That helps to jump around in the Mural board by this you do not have to fight with the scroll and zoom.
@dominica In Mural you have the Outline function. That helps to jump around in the Mural board by this you do not have to fight with the scroll and zoom.
Thanks for the tip @rene.lippert π
15Anytime before your reservation at the Restaurant at the End of the Universe is later.
@dominica Do you suggest to have limits around groups like Debts, Defects? So if there are more than x amount of Depts you are not allowed to work on Features!
The improving daily work and the reference to Risto Siilasmaa is interesting. Before being the chairman at Nokia he founded F-Secure. One of the original values of F-Secure that was never written in the corridor walls was something along the lines of "Instead of constantly asking for the time, build a clock so that anyone can see the time." The anti-malware industry has been heavy on automating for a long time. I find that one of the values in the DNA of F-Secure when I used to work there that drove it into a DevOps way of working as an early adopter... well, that found its way back home through the pen of @genek
@dominica do you have any data around the relationship between team size and unknown dependencies? Is it linear or exponential?
@dominica on charting flow distribution, example over 30days, you said if they have that data in their toolsets they copy and paste it in... what if they don't? Do they typically just estimate the aggregate %s, or go and try to build it bottom up from each work item?
@dominica on charting flow distribution, example over 30days, you said if they have that data in their toolsets they copy and paste it in... what if they don't? Do they typically just estimate the aggregate %s, or go and try to build it bottom up from each work item?
If they don't have this data in their toolset (and most don't capture debt work), then I ask them to take their best guess.
Thanks. Mik Kersten say he advocates just counting work items and "by the law of large numbers" that's an adequate proxy for effort, "it just works"... do you agree?
Every dependency doubles the probability of staying late or finishing late.
@dominica What is the best way to visualise the work of an expert who works across multiple product teams?Β Currently this person has their own Kanban board where their work is visualised.Β This seems wrong to me as I think it encourages siloed thinking and inhibits collaboration. Long term, I want this expert to shadow others to help teams develop the capabilities themselves.Β But until then, whatβs the best way to visualise this persons work?Β Is it by including it on the product teams Kanban Boards?Β Or is there a better way? π
@dominica What is the best way to visualise the work of an expert who works across multiple product teams?Β Currently this person has their own Kanban board where their work is visualised.Β This seems wrong to me as I think it encourages siloed thinking and inhibits collaboration. Long term, I want this expert to shadow others to help teams develop the capabilities themselves.Β But until then, whatβs the best way to visualise this persons work?Β Is it by including it on the product teams Kanban Boards?Β Or is there a better way? π
What makes this across multiple teams mode of working more desirable than the context switching caused?
What do you gain through the visualisation?
Is the work similar enough with the product teams' work to follow the same WIP limits?
@ferrix The across multiple teams mode is certainly not something we see as desireable. The problem is as this person is the only one who can perform a particular task, he is very much in demand by multiple product teams.
@ferrix We want to visualise his work because as with all the work we visualise, we want to see where it gets stuck, where there might be dependencies, conflicting priorities etc.
Does the person know without visualisation or with their own Kanban where they get stuck?
@dominica Do you recomend to differentiate in the dependency excersise between technical/software dependencies and team/personal dependencies?
@dominica Do you recomend to differentiate in the dependency excersise between technical/software dependencies and team/personal dependencies?
The dependency matrix exercise can be done with roles, skillsets, or architecture code components.
I like to do roles or skillsets when teams have long wait times between different roles/skillsets.
@dominica Thank you so much for the PDF - Iβm going to give this a try!
Thanks for all your great questions - I will be doing a Q&A session over zoom in 30 minutes: https://virtualdevopsenterprisesumm.sched.com/event/cq99?iframe=no
<!here> Join @dominica for Q&A during networking time! https://sched.co/cq99
β’ We have mentors doing onboarding for new employees. this way onboarding process can be fixed by someone who knows how and who to ask for help. β’ We provide one week of training about company and products. β’ We got architecture sessions provided by the team that the engineer is being onboarded to. What are some of other ways you guys improve onboarding?
Specially when you have 20 tools in a pipeline, and you need access for all of them, one-by-one. Good to have a centralized service for that
Specially when you have 20 tools in a pipeline, and you need access for all of them, one-by-one. Good to have a centralized service for that
And then somebody thought SSO is too expensive π
Oh, and add a few Bots to the picture, who I use to automate processes. They also need access rights π
nothing more motivating then working with a bot to try and get them to understand my request
Real life problem: "accessing my password manager from the data center when the network is down is not possible"
Not AI bot, just a bot to do automatic steps, like creating a Jira ticket automatically, or send a Notification to Slack. Or a Test User. All need their own account...
Try joining a US Federal project where it can take new developers as much as five months to start developing because they're waiting for clearances, badging and issuance of government-furnished laptops
Try joining a US Federal project where it can take new developers as much as five months to start developing because they're waiting for clearances, badging and issuance of government-furnished laptops
Happy devs deliver more secure, higher quality solutions!
Happy devs deliver more secure, higher quality solutions!
Joking aside we all need to eat, folks will put up with sub par experience because they need to pay the bills
Joking aside we all need to eat, folks will put up with sub par experience because they need to pay the bills
@chris.leeworthy Agree. But what happens is that we all become actively disengaged. Thats worse than if we just left.
@chris.leeworthy Agree. But what happens is that we all become actively disengaged. Thats worse than if we just left.
agree, and it's an agreement of tolerance... well you did the onboarding process so you must be patience enough to do the next crappy process. Slippery slope
Absolutely true @tom.sheeran, I spent a lot of time working with folks who just wanted to get by until they retired, I left that job because I didnβt want to become one of those people
Absolutely true @tom.sheeran, I spent a lot of time working with folks who just wanted to get by until they retired, I left that job because I didnβt want to become one of those people
very few, but sometimes you have to get them involved whether you (or they) like it or not!
@chris.leeworthy did the same move 9 month ago. Leaving behind people that were afraid every day of being laid off. And life (well, work) is sooooo much better now π
Itβs the same for me @szilard.szell, the folks at my new place have a passion and enthusiasm that has improved my work life hugely
It was scary after 19 years to leave the big boat, but so much better to work with like-minded people
Sorry for dumb question. What is this Unicorn status of which he speaks? I thought Unicorns were startups reaching $1B valuation... clearly not T Mobile
Sorry for dumb question. What is this Unicorn status of which he speaks? I thought Unicorns were startups reaching $1B valuation... clearly not T Mobile
good point though, a lot think we're already too big to fail from a business standpoint
Fear from failure is so deep in Telco. When you want five nine, and your bonus is based on fulfilling this!
@chill9 do you find it's typically a few passionate people holding up the transformation scaffolding?
@chill9 do you find it's typically a few passionate people holding up the transformation scaffolding?
From a selfish POV, at least we have that going for us. π
Well, more the fact that we are in demand. Helps overcome the fear of being crushed by the Empire. π
However, I want to find ways to highlight the risk of making your billion dollar enterprise's success depend on the passion of so few.
"Why not?" Yes, prove to me that you are special and none of this applies to you.
"Why not?" Yes, prove to me that you are special and none of this applies to you.
btw did you release the TBD checker? we came up with just looking at see how many unique commiters are on the source branch on average
@chill9 do you also have DevOps on Core networks side? With Network function Virtualization the platform is given. How do you see this area?
@chill9 do you also have DevOps on Core networks side? With Network function Virtualization the platform is given. How do you see this area?
it's an area I feel has always been left behind, but there are signals that there is significant improvement in the right direction. Updating network devices 100 times per day is foreign to a lot of firmware developers
How about talking about Policy Based Network automation, to freak out some even more? π
I still believe this is the time to start transformation on the network (and even Radio) side. Are you looking into that, too?
iterating very cautiously and slowly... it's never a network problem until ITS A FREAKIN NETWORK PROBLEM
Still it is a question if you want to deploy Network SW once a year, or maybe every 2 weeks, in collaboration with your multiple telco vendors, to provide fast feedback. Maybe deply to Lab is a good start first
Process Automation, Multi vendor DevOps pipeline, collaborative culture, common goals. Like a few challenges to solve π
AWS fails at bringing something into CORE a lot of times, we typically mostly see the successes IMHO
Building a broad base of allies who help promote the cause is so important. Elevating the success of others.
Building a broad base of allies who help promote the cause is so important. Elevating the success of others.
elevating can equal to negotiation and selling - which not all highly technical people are good at
It's a learned behavior, to be sure. I'm still learning it.
There is much power to be had in giving recognition/kindness/empathy.
Well, I failed a lot until I learned to listen to what people need. I still fail, but not as often.
Still a problem where some of our leaders don't realize the cost of using Saas vs the cost of you maintaining it, or building it yourself.
Still a problem where some of our leaders don't realize the cost of using Saas vs the cost of you maintaining it, or building it yourself.
I brought this up with a previous employer and their response was βWell weβre paying you already, why bother paying someone else to do itβ. :man-facepalming:
We call this the DevOps tax: cost of maintaining the pipeline. Estimation it is 10% of your effort!
@tommy I have this argument frequently with my Auto OEM colleagues. They have these massive monuments to CIO ego called data centers and have very sophisticated analyses that they are better at running them than AWS or Azure or Google
@tommy I have this argument frequently with my Auto OEM colleagues. They have these massive monuments to CIO ego called data centers and have very sophisticated analyses that they are better at running them than AWS or Azure or Google
do the data centers have strong pioneer like names making them sound like they are indestructible
Kelsey Hightower uses a great analogy. It's like you go on a plane, love the speed and convenience in getting you to your destination then decide to bring the 747 home and park it in your driveway.
I have seen same analysis with my previous company π However that was true we were heating a small town from the waste heat of our Test Laboratory
Yes! Cheers for @jessicam @annp @alex and everyone else on staff. Fantastic job
@chill9 Awesome talk. Iβve no doubt youβve won a lot of frustrated developers over as life long followers.
@chill9 Awesome talk. Iβve no doubt youβve won a lot of frustrated developers over as life long followers.
See also #ask-the-speaker-keynote . Most of the comments on the talk are there.