Every organization gets ~3 "innovation tokens" for non-standard technology choices. Giving example from personal experience Cost of a token: what is it solving? Is its complexity earning its keep? Are we falling for hype? "Doing well at business is what gives you more innov tokens" - @charity.wtf

My hot take here is that going remote has not been, on the whole, a positive change for individual contributors.

Finally moved the homelab to @honeycomb.io and the o11y 2.0 gains for agents hype is real. Game changer using it with an agent. Fixed a weird tracing issue with like 2 prompts with Claude (using the skill/MCP) to review, find and fix.

OH: "I turned off news alerts because alerts are only useful if they're actionable"

@chadfowler.com i am supposed to be writing right now, but i've spent the last 2.5 hours reading the last few months of your writing. i can't believe i am just discovering it now! there is SO much good stuff here! (yes, i've been under a rock since november ✍️📖☠️)

“SaaS is dead. Everyone will just create what they need” I think this underestimates what a PITA it is to create software. Lots more short form video content is created today than ever before. The tools are now ubiquitous. The vast majority still just consume it.

it's almost time for SRECon26, which means it's been less than a year since @ferd.ca and i were delivering the closing keynote at SRECon25. my perspective has changed A LOT since then, and the advice i would give has changed too. i wrote it up here. charitydotwtf.substack.com/p/my-hypothe...

Agentic engineering is driving the cost of code to zero but the cost of good taste and product sense is going to go sky-high.

speaking of asbestos, y'all should read "No More Tears: The Dark Secrets of Johnson & Johnson". www.goodreads.com/en/book/show... I bet you thought you knew about the fucked up American health care system. I bet you thought you knew about regulatory capture. I bet you can still be shocked. I was.

I'm on holiday and catching up on things: a week after @charity.wtf made it public, I'll (re-)inform y'all that I'm amongst the contributors to Observability Engineering, 2nd Edition, along with the likes of @jeremymorrell.dev and @phillipcarter.dev. I will of course be all over User-Focused O11y!

this man is a fucking genius. shovel him some silicon valley dough *stat*

I propose "Human attention is the scarcest resource in your company" as a potential framework for navigating this new world

Outcome Engineering (via @charity.wtf): o16g.com

For those who have to use LinkedIn, here are my survival tips: * Install the "Minimal Theme for LinkedIn" (you have to VPN to a non-US country, then search for it, but it scrubs nearly all ads and noise) * Disable notifications. Like, ALL notifications, except for the few you specifically want

My favorite LinkedIn use case: reading shitposts bsky.app/profile/bram...

This post sold me on @charity.wtf’s new book. If there’s one thing I know after 20 years of working as a director in high tech, it’s that my success depends on how well I can communicate with execs who don’t know my field. That means it’s my job to find common ground. charity.wtf/2026/02/19/f...

i am not a huge fan of the term "technical" as an adjective applied to people... 🤔 i would say, rather, that when you become sufficiently untethered from current workflows, architectures, and pain, your ability to assess risk becomes similarly ungrounded.

Yesterday, I wrote the first half of a two-parter on the second edition of "Observability Engineering". Yes, it's twice as long as the first one (oops) but I think it's a *much* better book. charitydotwtf.substack.com/p/observabil... (by @gmiranda98.bsky.social, @lizthegrey.com, @aparker.io & me)

Least favorite new LinkedIn use case: carpet bombing people with messages asking for comments and likes and RTs to boost attention to newly posted content. Do not want. 🙅 (favorite bsky use case, prolly complaining about LinkedIn)

can you believe it's been _ten years_ since @cyen.bsky.social and i started honeycomb?? jfc. @honeycomb.io was founded to solve the hardest problems in software. in 2016, that was cloud; in 2026, that's AI. the problems have gotten worse; the stakes got higher. www.honeycomb.io/blog/honeyco...

o16g.com is the best thing i've read in a week full of seriously great reads. and now, as of two hours ago...with its own line of merch. do scroll down until you see the "honeycomb edition" shirts.. you will understand my excitement 😍🌈 www.bonfire.com/store/o16g/

Once again noticing that LLMs that write code often get a person’s name and are described as partners, but AI SREs are generally named “AI SRE” and they are there to make sure nobody’s got to stop to do that lowly unproductive work.

People still don't grasp just how much pain and suffering is caused by trying to build software using "observability" tools that are hand-me-downs from ops. The three pillars make total sense if you can't change the code you run. That is the *only* time it makes sense.

"What if instead of buying software companies build their own software," is not in any way a dumb question. Tens of thousands of people ask it for themselves and their colleagues every single day. And most of the time it's a fucking disaster. So yeah, if someone can make it easier, we'll take it.

DAMMIT ED, NO Have you TRIED procuring and integrating enterprise software? Because OH BOY are you in for a tragically underestimated and regularly disappointing ride I'm not saying the big apps are all terrible. (They are *mostly* terrible.) But their terribleness isn't even half the problem here

Sorry, paused to have a totally coincidental quick chat with @charity.wtf ON THIS VERY TOPIC, during which we both said, almost simultaneously, "build vs buy isn't a binary choice, it's a continuum" because buying software doesn't meant you don't have to build anything! it's worse than that!

I’ve heard and read several things lately that were leading me to this belief, but I think this is the right framing.

“In retrospect, I’ve come to think that the entire effort was about trying to connect developers with the consequences of their code in production.” 🌍🧑‍🚀🔫🧑‍🚀 Right, but not just at a technical level, the path Charity et al paved was connecting them to *business* outcomes. That was the 💡for me.

What people are struggling with is less about the novel aspects of developing with AI, and more about the table stakes. It's not the AI part that's so hard (turns out systems have never been that deterministic), it's the software part. 100x as much software, 100x as fast as before.

The future of software engineering is SRE

This is a really fascinating view of the future of devops and observability, particularly the perspective on development feedback loops. AI makes us faster at observability, but we will have to be faster at deploying and testing code too.

Another great post from @charity.wtf charitydotwtf.substack.com/p/bring-back... When I ran the first customer 0 cloud at VMware, my Head of Ops would say, “ you can automate all you want, but at the end of the day, someone has to care about the infrastructure.”

So annoyed by all the thinkpiecing going on right now about how "AI-generated code shifts the REAL burden and bottleneck onto code review." So close! but so wrong. (Not linking, because I am not that kind of asshole. (Anymore?))

"A key function of ops is to absorb hard problems and make it easy to write and ship features." 💯💯💯

also, corollary: infra does not ONLY mean backend. there is frontend infra. holy shit do @mcapoz.bsky.social and team do a lot of work to make our UI components and everything else around them scale!

The difference between "dev" and "ops" is not about who writes code. It's 2026, dude: everyone writes code. The difference is a separation of concerns. Are you responsible for writing code to generate revenue, or building the systems for code to run on? charitydotwtf.substack.com/p/bring-back...

Devs have had it too good. We need ops-first tools. And operational loops that don't require coding new programmatic artefacts in order to get systems back up

I don't think I've ever shared this here before, so share I must: www.amazon.com/Subtle-Art-S... (If you open it up, it just says "No no no no no no no no no" for fifty pages or so. Makes a great Christmas gift!)

“Pick your brain” is my most hated expression. Not only is it grotesque, it just says the other person thinks /you/ have more of an obligation to share your wisdom, than /they/ do to spend time formulating questions. 🙄

I thought you'd never ask. 🐺👠🦄💕🥂✨

This book has been one of my favourites to read and contribute to. I’m really looking forward to seeing the entire book in the flesh! (And speaking of forwards… Be sure to read the forward in this book. I might be biased but I think it’ll be great 👀)

I've come to view the entire DevOps movement as a mighty, twenty-year battle to achieve one thing: a single feedback loop connecting devs with prod. It failed, but not for lack of trying. It failed because the technology wasn't good enough. And now it is. www.honeycomb.io/blog/you-had...

Treats from today's LinkedIn mailbox: them: "We're developing a book on 'AIOps'. Because of your expertise, we thought you would be an excellent author for it." me: "why would you think that, when I have only *ever* brought it up as a target of mockery?"

Shout it from the rooftops! I really like this characterization of the types of code we're going to have going forward, from @charity.wtf : www.honeycomb.io/blog/disposa... Disposable vs durable code is a useful split. We always had both, but AI makes it even easier to build the former.

All thirteen of my chapters for the second edition of "Observability Engineering" are now drafted to satisfaction. The first half is already in the hands of tech reviewers, and the second half should be to O'Reilly by Jan 30th. Just doing a read thru and cleanup, interstitials, stick art and stuff.

There is a market for artisanal furniture, but not nearly as big of one as there is for IKEA mass-produced product. And we are going to need to have the reckoning about our industry getting automated out from under us and artisanal techniques no longer being state of the art or economically viable.

The trouble, as @charity.wtf likes to remind me, is that those of us who program for work are not being paid to maximise our fun, we are being paid to produce a product as efficiently as possible. So at some point the people who got into programming for fun, may wind up leaving our field.

Here’s the episode: podcasts.apple.com/us/podcast/t...

Just listened to this episode of The Engineering Room with @charity.wtf and @davefarley77.bsky.social, extremely good! This part on managers being “shit umbrellas” hit home catastrophically for me. So rare (IME) for management to treat their underlings as adults.

While I appreciate the nod to my Molotov-hurling younger self, I've actually moderated a bit on this point. Not about where you _should_ be (point still holds), but on the mitigatory steps you may need to take in order to get there. charitydotwtf.substack.com/p/on-friday-...