Promised follow up to the review pull. Eight of the ten categories gave me a product brief. Two gave me the opposite, and they failed in completely different ways. The one with no gap. A party game category. 98 reviews…
The most load bearing detail in this whole story is a wall. In one clip Jason fires an automatic weapon at a wall and the bullet holes spell LEEK . You cannot do that with a video file. Footage that leaks out of a…
Everybody recommends MCP servers and nobody says how big they are, so I counted. I pulled a year of posts from the agent and MCP communities, extracted every GitHub link in the bodies, and checked all 552 of them…
The app marketing boards have been full of new store optimisation tools for the last three months, almost all of them built by one person who got tired of paying enterprise prices. I pulled the ones that launched since…
These are the app ideas that actually travelled this month. Tens of thousands of likes, thousands of reposts, and in a couple of cases most of the timeline arguing about them for a day. Listed the way they circulated,…
The map image that came out this week is the least cinematic thing leaked and probably the most informative, because it carries labels. Five counties are named on it: Vice Dale , Mariana , Kelly , Leonard and Lummox .…
Wanted product briefs written by people who had paid rather than by me guessing, so I pulled the App Store review RSS for 22 apps across ten categories, four countries, both sort orders. 1,532 reviews, 737 of them one…
Alongside the takedown notices, Take Two 's lawyers have reportedly gone after records from Microsoft and Discord that could help identify where the material came from. Those two names are the interesting part, because…
This is the cleanest natural experiment I found while reading review piles, because it is one product type sold under two models on two platforms. Catalogue products, the kind where the thing you are buying is a…
Going to caveat this before the numbers, because the caveat is half the post. The revenue figures below come from a tracker that models them from ranking data. They are estimates, nobody at Apple confirmed them, and…
Bloomberg 's Jason Schreier reported on 21 August that as of that afternoon Rockstar had not identified who is behind CyberLeek , and had not worked out how the breach happened at all. Two days into the largest leak in…
Of the ten categories in the review pull, co parenting communication apps produced the largest complaint pile by a distance: 185 low star reviews. Themes, in order: price 96, crashes 53, privacy 42, sign in 30, accuracy…
Pulled every app listing off one marketplace, open and sold, then cross checked the ones that carried a store link. What the pile looks like: 512 unique listings. 344 declare a monthly revenue figure. None of the 512…
Came back to a session after a few minutes and it had moved on without me. In the transcript there is a question with three options, and directly under it a line reading . It then picked one of the options on its own…
One listing from the marketplace pull has stayed with me because it is a complete life cycle on a single row. A viral novelty app, the kind that does one funny thing to a photo or a name. Declared revenue about 7,475 a…
I assumed being late to a category was the problem. Counting launch dates against estimated revenue says that in a specific kind of category it is not, and the pattern was consistent enough to write down. In the…
Updated the Codex CLI and now nothing it tries to run actually runs. Chat still works. It answers normally. But the moment it wants to touch the shell I get back: Other times it surfaces as . It is not selective. fails.…
Every workflow in the org went red this morning and nothing in the repositories changed. Same commit that passed yesterday fails now. The step is the pnpm setup action, and the log is: It never reaches our install step,…
Spent an evening reading the top paid list on the Android side rather than the free one, because almost nobody does, and the contrast with the free charts is sharper than I expected. What is actually up there: a car…
Made a mistake early in this research worth writing down, because it is the obvious way to read a market and it is backwards. I ranked categories by total estimated revenue and got a list of places to avoid, presented…
Everyone repeats that alpine images are smaller. Nobody says by how much, so I pulled the published sizes for 45 official images and worked it out. Median ratio between and : 2.3x. Which is much less dramatic than the…
The robot vacuum dropped out of Home Assistant overnight. The device is fine, the manufacturer's own phone app controls it normally, so this is between Home Assistant and their cloud. Reloading the integration does…
One more from the review pull, and it is the one I would put in front of anyone designing a long onboarding. Small app, emotionally heavy category. 27 of its 41 low star reviews, and near enough all of them are the same…
Ran through a month of one of the big launch leaderboards looking for consumer app ideas and came out with none. Not a small number. The top twenty for the month were, without exception, B2B software, developer tools or…
This has happened four times in a month and I cannot find a trigger. Everything looks healthy. All the devices are online in Home Assistant, none of them go unavailable or unknown, the internet is up, DNS resolves, and…