It splits revenue out to 3 categories, "Productivity and Business Processes", "Intelligent Cloud", and "More Personal Computing", with windows as one of several things in the 3rd group. How did you figure it out as a 5th place revenue source?
Tariffs/inflation/everything has raised the unit cost to the point that they're probably close to running a loss again sometimes on the latest gen consoles.
On possibility I've seen raised is that slower GI movement -> slower alcohol uptake -> not getting as much of a "hit" from drinking as the effects come on more slowly.
In my personal experience, I do still get the same hit from drinking–I feel a buzz almost immediately, same as before. Rather, I just don't feel the "urge". I've never been a heavy drinker, but I would occasionally crave a beer or two, particularly at the end of a work week. Also, drinking on a GLP1 (I've been on both Tirzepatide and Semaglutide) absolutely wrecks my GI tract for 24-48 hours. Usually with an onset of maybe 8 hours, I get horrible heartburn, moderate to severe nausea, and even mild diarrhea.
I don't think it's any one thing. People like different kinds of alcohol, for different reasons. For someone who's alcohol cravings are based on the sugar in their preferred alcoholic drink, it isn't surprising then, that a medication that lowers their desire to ingest sugar lowers their desire to drink (their chosen sugary drink). Naturally this doesn't cover all alcohol drinkers, but it can't also be none of them.
It’s definitely not ccache as they cover that under compiler wrapper. This works for Android because a good chunk of the tree is probably dead code for a single build (device drivers and whatnot). It’s unclear how they benchmark - they probably include checkout time of the codebase which artificially inflates the cost of the build (you only checkout once). It’s a virtual filesystem like what Facebook has open sourced although they claim to also do build caching without needing a dedicated build system that is aware of this and that part feels very novel
Re: including checkout, it’s extremely unlikely. source: worked on Android for 7 years, 2 hr build time tracks to build time after checkout on 128 core AMD machine; checkout was O(hour), leaving only an hour for build if that was the case.
Obviously this is the best-case, hyper-optimized scenario and we were careful not to inflate the numbers.
The machine running SourceFS was a c4d-standard-16, and if I remember correctly, the results were very similar on an equivalent 8-vCPU setup.
As mentioned in the blog post, the results were 51 seconds for a full Android 16 checkout (repo init + repo sync) and ~15 minutes for a clean build (make) of the same codebase. Note that this run was mostly replay - over 99 % of the build steps were served from cache.
Do you have any technical blog post how the filesystem is intercepting and caching build steps? This seems like a non-obvious development. The blog alludes to a sandbox step which I’m assuming is for establishing the graph somehow but it’s not obvious to understand where the pitfalls are (eg what if I install some system library - does this interception recognize when system libraries or tools have changed, what if the build description changes slightly, how does the invalidation work etc). Basically, it’s a bold claim to be able to deliver Blaze-like features without requiring any changes to the build system.
One of the famous small canadian mining companies that went under was named something like Bre-X, somehow a lot of members of the general public had shares of it so it was a big scandal on the news when it went under. Also as it was unraveling a whistleblower at the company "fell" from a helicopter in indonesia, I don't recall if anyone was ever charged or convicted for that.
EDIT: Oh damn it was far sketchier than I recalled and he wasn't a whistleblower. From wikipedia:
The fraud began to unravel rapidly beginning on March 19, 1997, when Bre-X geologist Michael de Guzman reportedly died of suicide by jumping from a helicopter in Indonesia.[11][12] A body was found four days later in the jungle, missing the hands and feet, "surgically removed".[13] In addition, the body was reportedly mostly eaten by animals.[14] According to journalist John McBeth, a body had gone missing from the morgue of the town from which the helicopter flew. The remains of "de Guzman" were found only 400 metres from a logging road. No one saw the body except another Filipino geologist who claimed it was de Guzman. One of the five women who considered themselves to be his wife was receiving monetary payments from somebody long after the supposed death of de Guzman.[13]
I was born and raised in Vancouver, and have deep personal ties to the junior mining industry - I worked for two decades as an exploration geologist, and my Dad has been in the industry since the early 1970s.
Bre-X was brutal. I was barely into my teens, but I recall - and have spoken at length with my Dad and many others - about how he was out of work for several years after the scandal. Investment completely dried up. Industry recovery took years, and was accompanied by the implementation [1] of fairly stringent disclosure rules, which define reporting standards to this day. Nonetheless, scams are still commonplace, and pretty much everyone I know has a story or two of a shifty promoter pulling the rug out.
Mineral exploration is a tough business. You can't just sudo apt install a drill rig! The logistics and expense of even small exploration programs are a bit insane. Crews head out to some of the most remote corners the world has to offer, moving hundreds of thousands to millions of dollars of heavy equipment, fuel, food, and camp gear on to site for just a few months of near-constant work, then moving (hopefully most of) it out again. Hundreds of tons of rock and soil samples are collected, by drilling, by walking the ground, by trenching; these samples are shipped out to processing and assay labs. Some properties have the benefit of road access - deep-wilderness, often decommissioned logging roads - but many are accessible by helicopter only. It's an adventure, but it is also very demanding and taxing. There is very little year-to-year consistency, even in bull markets.
Sixty to seventy years ago, the majors - mining companies with actual mines and annual revenue - did the lion's share of exploration work. Over the years, however, the majors have divested almost entirely from risky grassroots exploration, leaving it almost entirely up to junior explorers who must raise their capital from investors.
reply