Product @CultureAmp. Previously @Atlassian. Future of work, mashups, home brewing, and automation. Opinions my own. he/him. May also use @mindsocket@aus.social

Sydney
Joined June 2009
Roger Barnes retweeted
"Let’s plan the roadmap for next quarter” is the modern day equivalent of “let’s fax everyone the strategy.” We're going into q4 planning and it makes my brain hurt. Roadmaps existed because software was expensive to build. You needed a bunch of people. Meetings. Someone moving little boxes around Jira for three months from now. It makes no sense and I'm trying to work out what we call the replacement. Maybe just "Continuous Compilation", ideas just compile to working product. I want a list of ideas that we want to continuously compile. That's the roadmap now. The cheapest way to figure out whether something is a good idea is often to just build the fucking thing. Here’s how I work now. I've done this weekly over the last few months and it works. My current napkin math is that I can usually build 6 months of old-school roadmap in a week. Sometimes more. I've also hit a wall sometimes. It happens. When we have an idea, I usually 10x the ambition and then write a prompt, not more than a page long, that basically says: - here’s the ambitious goal - here are a few use cases - here’s roughly how I think we should build it in terms of where it fits into code base, design considerations, etc.. - here’s how we’ll know if it actually works - now go fucking surprise me Agents build all night. Not even with a crazy factory or complex agent or agents. The base models are just too good for that now. In the morning I look at it. Ok, they usually build in a few hours. Usually some of it is great. Some of it is stupid. Some of it exposes that my original idea was wrong. And almost every time it gives me 5 new ideas I couldn’t have planned upfront. So we throw shit out. Who cares? We learned something. Then we change direction and send them off again. It usually takes 3 rounds before the shape appears and we've learned about the gremlins and possibilities. 95% of the time I end up building something with 10x the scope of what somebody had carefully spread across the next quarter in Jira. “What should we put on the roadmap for next quarter?” My brain now translates that to: “Oh, you're stuck in the 1500s”
34
19
8
345
40,550
Replying to @mattpocockuk
Plan driven development? Your skills are not quite the same as plan mode in Claude Code, but more similar than different.
31
Was it "bring your kids to work" day at Microsoft recently?
50
You urgently need a different project name, or to at least stop actively avoiding mentioning who you aren't in important places like your web site and bio.
24
I'm finding the single wiki config per instance limiting. I have multiple wikis for quite different topics. Config should be at agent and/or directory level. I already map telegram topics to context directories separately.
276
I like it! I'm finding the (expected) differences in interpretation of @karpathy 's guidance interesting. I've built flexible but deterministic schema validation with llm skills. I'm working on integrating these as complementary to other skills. github.com/mindsocket/struct…
1
247
Replying to @amix3k
I genuinely don't know what I'd use an activity log for, so not sure what to ask for. Most of the things I need relate to clarifying intent, prioritising in context & getting out of the weeds - all more related to forward view than backward
40
Replying to @levi @bcherny
+1. Multiple times now I've had to ask to see the plan first, which breaks the question asking flow
35
Replying to @sussanley
"Our plan is clear" is usually accompanied by some kind of plan. Did you forget an attachment?
8
.@DJKanban Inc just hired its first 3 employees. What could go wrong? (Narrator: Everything) #vibecode
2
1
218
Naming things is hard, just ask @UnderArmour
48
Voting in Bradfield and undecided? My frustration with politics has turned to hope since meeting @Nicolette_Boele and volunteering. A true community independent. See for yourself: youtube.com/watch?v=X2c7SDCk…
173
Replying to @mindsocket @amix3k
Examples of conversational things (I've started dabbling with MCP for these myself): * Reschedule anything due that looks work related until Monday * Suggest tasks that to be moved into existing sections in the work project * Tag priority tasks that I keep rescheduling for review
1
20
Replying to @amix3k
2 things: * It depends what it can do. I'm interested in things I can't already do from the app that's already on my device. * Assuming it does net new useful things, I'd be happy for the conversational interface to simply be in the todoist app, at least to start with
1
1
81
I'm done wringing my hands about the state of #auspol parties. This was my first time doorknocking alongside hundreds of volunteers from all walks of life. It was fun and rewarding. It's time for a positive, representative change in Bradfield.
Just had our biggest doorknocking weekend. Thank you to the 127 doorknockers (yep - you read that right) who invited people back into their democracy this weekend. This is people-powered politics, and it’s an absolute privilege to be a part of 🌟.
138
Happy clamshell packaging injury and fresh battery hunting day, to those who celebrate
49
Eminem v LLM His coffee's ready, 𝚙 is weak, cloud bill's heavy There's vectors in his pinecone already, similarity He's nervous, but on the servers, he's got json ready To drop prompts, ...
1
80
"This is either genius or bollocks" is almost always how genius emerges. You've gotta be ok with that. Keep going, keep creating.
“A lot of what I do is like ’This is either genius or b****s’” Part 7 of Behind The Tune is about my new single “Bus Stop Please” with #DanielSteinberg Watch the full clip on my YouTube youtube.com/playlist?list=PL…
1
97
Replying to @sqs
You don't have to change code if you only ever make new messaging apps.
46
Roger Barnes retweeted
Goals are for people who care about winning once. Systems are for people who care about winning repeatedly.
247
3,438
180
18,099
805,057