Cédric Kovacs-Johnson

@ProV1Smile Post-run — distance

Spitballing top 5 beers 1. Free beer 2. First beer of the weekend 3. AM airport beer 4. Campfire beer 5. Shower beer Taking input. Never afraid of constructive criticism.

Very much where the future is heading

@loungeriffs @ericvishria Because it’s one piece of the data estate

@ericvishria you might find my thread here interesting. my view is SOR should win, if they add agentic features why would a CIO switch? they WANT them to be the one that wins. question is how patient will customers be… (and how patient will vendors make them be) x.com/loungeriffs/st…

Last Friday, a few of us were debating which is more durable. I thought SoR, but the counter argument is AI makes migrations so, so much easier.

@burkov This is like that Dropbox comment on HN

I didn't want to comment on OpenClaw. Usually, when there's so much noise in the media, it's some ordinary stuff just hyped well. So I took time to learn how it works thanks to open source. I was right. OpenClaw is 2% of ordinary stuff and 98% of hype. To put it very shortly, in case you were wondering, there are two things in it: 1. You can chat with an LLM via a text messenger. Not anything new. 2. The LLM can use tools that run on your computer. Not anything new either. Most of the "magic" mentioned in the media is about its ability to use the browser. But it's not *its* ability. It's Playwright's ability. Playwright is a library made by Microsoft which allows you to programmatically run a browser. It uses a built-in vision model made by Microsoft that converts the browser's screen into a textual description for LLMs. Again, Microsoft has built Playwright exactly for what OpenClaw is using it. So, OpenClaw's typical workflow: 1. The user types in a text messenger "Buy me a flashlight on Amazon." 2. OpenClaw blindly dispatches this message to an LLM which has access to some tools, including Playwright. 3. The LLM, trained not by OpenClaw folks, decides that Playwright is the right tool (of course it is) and Amazon is the URL to navigate to. 4. Playwright, built not by OpenClaw folks, runs the browser, which navigates to Amazon, and returns the textual description of what Amazon's home page looks like. 5. OpenClaw blindly returns to the LLM this textual description. 6. The LLM (again without any help from OpenClaw) decides that one should type "flashlight" into the search field and press Search, so it calls the Playwright tool with the search parameters. 7. OpenClaw calls Playwright because the LLM told it to and types "flashlight" and then presses Search (it's all part of what Playwright does out of the box). ... In the end of this LLM-controlled scenario, the order is submitted. OpenClaw just listened to what the LLM told it to do via tool calls. I tried hard, and I haven't found anything else worth mentioning in the source code. There's also a part that keeps "memories" about past conversations, but it's all basic stuff. These memories are stored in text files and grep (controlled by LLMs trained to use grep, and trained not by OpenClaw folks) is used to search in them. It's a nice hobby project, just like Cursor or Perplexity are nice hobby projects, but there's nothing there to look for, except for the hype and 2% of unoriginal plumbing code.

@zain_hoda Caching layer = lake house

@jmj This is profound and also super common with exited founders. Who have a huge aperture, resources — and seemingly unable to commit and see any one thing through. When Everything is possible, sometimes nothing Happens

I built a mobile app I’ve been dreaming about for years in a few hours last night. It’s ugly and I probably won't ship it. But I was up and running in Expo within an hour and it worked. The product would have taken me months & probably $50k to do a year ago. We talk a lot about taste as the moat in the next era of company building. I’d add one more: focus. As the great commoditization of software happens, the ability to keep doing one thing every day will become rare. I'm already seeing this amongst people I know who are builders w/ Opus 4.5. Some of them have even raised capital and have their own startups. It's easier to go home and build a bunch of things in Claude Code rather than do one thing. When you can build anything, sticking with one thing is the hard part.

They’re called ‘algorithms’ because Al Gore actually created the internet

@LibreOffice Exactly my point

@cedrickojo That's risky! OpenOffice's last major update was in 2014. There have been rare tiny updates since, but it has "red" security status from Apache and is not recommended for production use: #Security" target="_blank" rel="nofollow noopener">en.wikipedia.org/wiki/Apache_Op… – Far safer to use one of the successor project (like LibreOffice).

Just look up Mattermost or OpenOffice. Free OSS alts, never really worked — outside of niche use cases

@realmadhuguru The moment in time to do this is migrations or data integrations. Otherwise it’s a documentation exercise that goes stale

The best enterprise AI implementations I see: Pair workflow experts + people with great product sense (rare). Understand workflows deeply - tools, manual steps, coordination across systems, hallway conversations. Common gaps to address: - Codify institutional memory (in people's heads) - Codify undocumented processes - Clean existing docs (which of the 3 HR policy docs on the intranet is the latest?) Build: spec workflows, representative evals, systems that auto-improve as models evolve. Critical miss: treating AI like traditional software. Models evolve every 6-8 weeks. Need continuous iteration, not annual roadmaps.

@reidhoffman Context graphs everywhere

Enterprise AI strategy is backwards. Most people are focusing on Chief AI Officers and pilot programs, when the real value is in the unglamorous work where organizations bleed time. More thoughts:

@bentossell The robots also need to write the documents

humans reading docs will go away

@jerryjliu0 I don’t think everyone even agrees what it is yet

I have no idea what a context graph means outside the high-level marketecture Is it just RAG over your activity logs?

8 months in .. I get it

This is literally where I learned how to play backgammon

Enterprises operate on data environments that are effectively unknowable -- massive, drifting systems no one fully understands - and a scale that's hard to intuit The precursor to why context graphs matter @jaminball @JayaGup10

@prukalpa Metadata 2.0

@bhalligan @PaloAltoNtwks @nikesharora The lean startup did some real damage

This might be the most uncomfortable truth for early founders: Listening to customers too early can ruin your company. @PaloAltoNtwks CEO @nikesharora explains why enterprise buyers will always pull you toward “speeds and feeds”—and why that leads to products no one can actually deploy. Customers are great at telling you what’s broken today. They’re terrible at imagining what should exist tomorrow. The best founders earn the right to listen by first having a point of view. Build the vision. Then let the market react to it.

Is backgammon just the best ?