Why Socio-Technical Practice Is So Important For Engineers | Jessica Kerr On Valued Capabilities

Jessica Kerr, Engineering Manager at Honeycomb, talks to Dave about her fascination with software development as a socio-technical exercise.

--------------------------------------------------------------------------------------

⭐ PATREON:

Join the Continuous Delivery community at access extra perks & content, join in our CD Discord discussions and support the CD YouTube channel.

--------------------------------------------------------------------------------------

🔗 LINKS

-------------------------------------------------------------------------------------

👕 T-SHIRTS:

A fan of the T-shirts I wear in my videos? Grab your own, at reduced prices EXCLUSIVE TO CONTINUOUS DELIVERY FOLLOWERS! Get money off the already reasonably priced t-shirts!

🚨 DON'T FORGET TO USE THIS DISCOUNT CODE: ContinuousDelivery

-------------------------------------------------------------------------------------

🙏The Engineering Room series is SPONSORED BY EQUAL EXPERTS

[KairosDCC]

Spent a decade building technical solutions and systems in education before going into other verticals. You couldn't be more right. Great software development requires a purposeful learning system with evolutionary and intentional revaluation of presumptions in order to succeed at scale over time.

[stephendgreen]

I like the thoughts here, as I also like the Doug Engelbart ideas of producing tools which people can use to manage their information complexity problems, and tools to help those producing the tools which allow collaboration to find new ways forward and boost collective IQ (bootstrapping as he called it in those early days). So yes, we as software developers don’t solve problems directly as our initial approach but rather we can produce tools which cause synergies which can ultimately help solving of problems, such as information management, bringing minds together to that end.

[marshalsea]

"No Software Without People" is the key statement; the moment you go into a team you are part of a human system.
Many, if not most, of the Stakeholders in your system will be untechnical and therefore there is a communication barrier that needs to be bridged.
This goes in both directions > the resistance of "Non-Techies" to be dismissive of the skills are techies is a major issue > cue psychology on typical reaction when you feel dumb.

The average experience of reading a Stakeholder's requirements is analogous to Bernard Black reading his tax form (youtube Bernard Black: Taxes).

Therefore, the way to bridge this is to introduce a common language: standards and common pseudo-language such as Stories and GWT... but how often are these done to a good standard?

[Nohbdy_Ahtall]

Whoah, good clips indeed. Convincing people - how, how. Without biz or corp moneying involved, just tools for grassroots and decentralized & mutual inter-dependent coordination. First barrier is no one sees the value in trying, let alone imagining it.

[pavallok]

Hey Dave, amazing channel, can't stop watching your videos, will buy your book eventually...
I know is a bit off your domain but can you talk more about the internal structure of a team please.
E.g. I'm a sw eng whom don't believe in manual testing, neither complex automated testing suites, so I don't see much point in having "testers". I met amazing quality engineers that were crossing over the product owner and ux realms though. I'd be curious to know if anybody has analysed these structures perhaps at the level that the dora team has done, and what the outcome of that study was.

[aaaowski]

>our job is to provide capabilities
yeah, only if you enjoy being a cog and/or are devoid of leadership skills.
probably the reason you hold agile in such high esteem too, as it is a stand in for organisational know-how.
id say disappointing piece, but i knew what to expect from the title.

as for twitter...
cmon. theyre a bunch of half-baked webmasters. theyre not creating software. they do arts and crafts with existing solutions. did.
given that twitter is now being merely maintained (barring elon musks' new "features"), there is literally no reason to be 10k+ workers.
edit: i see no reason to be 10k even when creating the site.
even if everything down to the os the servers run on was to be written from scratch you still dont need 10k people.
im pretty sure 500 people would do. tops. and even that sounds vastly exagerated.

>how twitter hasn't come apart at the seams yet with such layoffs?
yeah, programming isnt shoveling coal, lady. and even then, most of twitter lay-offs were political hires. nothing of value was lost.