This sounds like a company organizational problem — if the ultimate goal is to get people to use your cloud and provide value, then why would you have metrics that diverge from what the customer sees as the important things? There should be somebody at Google actively talking to customers about what problems they have and communicating that to the engineers so they know what to improve. The organization should be structured so the engineers get the metrics they need or make it part of their job description to make the metrics.
If you / somebody is doing the sand and scan or xray/CT method (which you could pay somebody to do rather than buying a CT machine), then you can create a gerber -- then manually clean it up. Then you have a dangling set of nets that are only separated by layers. You can then infer connectivity from the gerbers on layer to layer manually again to create a reduced set of nets by the shape / visual cues of what the vias look like. That would be far easier than soldering wires to every ball on both sides of a board -- and a netlist doesn't automatically generate a schematic for you, you need to still do a chunk of work to actually create the schematic. To be honest, a netlist is not actually all that useful unless your goal is to attempt to create a full schematic out of the board. For reverse engineering efforts, you would likely focus on one chip and just manually follow each trace for the thing you care about and draw up a schematic manually for that. In most cases you would likely spend like 1 day after you got the scans back building up a schematic for the key chips of interest. For anything that is a bit questionable about if a via actually connects or not, then you would just manually ohm that out. Anyway, I guess if you like soldering and are just doing something for fun, then sure do this method. Otherwise, there are way better approaches than this.
Yea ChromeOS is not an OS for people wanting to move to "linux".. basically if all you want to do is like run a browser and do basic things sure.. but if you want a system to learn about linux and do some fun stuff with your system, pick something like Mint.
Aren't many of the 4090 retail GPU's being built in China for foreign companies -- that are then supposed to be shipped out of the country?
If a factory "overbuys" too many of the 4090 chips, then it would seem to be not too hard to just have a shadow inventory of chips that they just never end up using -- could easily be in the thousands for many suppliers without causing much suspicion due to the normal levels of stock they would need to have on hand for builds.
Somewhat hilariously at the same time we think there must be life given the 14B year old universe, we look for 50 years and say, huh we can’t find anything so something’s wrong.. like sure our initial assumption was that a super powerful alien race would blast a powerful signal that would tell us how to contact them — and this signal they have been powering for millions of years ? — that would be great but seems like that isn’t happening. So likely we need to just keep searching.
I think in many people's goal to "simplify using a computer" it ends up making things that could easily be educational without actively trying to teach you anything -- basically sparks curiosity and informs a bit. (like this great example the author shows here). One example of this (that previously existed in actual computers) is the old trusty red hard drive light telling you that the hard disk is active... if you were like me, you knew the game was going to actually load this time when it showed a particular pattern and you heard the hard drive make a satisfying sequence of fast disk reads. Seems like a nice compromise is to hide the "advanced view" but keep it there for the curious people who likely will be the next generation of computer nerds making the world go 'round.
https://www.youtube.com/watch?app=desktop&v=M64U0XyLA-o&feat...
Palm Pre in 2009 had a magnetic charging / mount thing super similar to what the iPhone only finally got a couple of years ago.. Palm Pre also had the nice card-based UI for killing apps while multitasking -- again something basically iOS only finally got after many years of annoying versions that all sucked compared to the card based mode they use now. webOS used webapps -- which at the time were not great.. but now tons of apps are effectively just webapps. So yea, Palm Pre was uber ahead of its time...
The reality is that in so many countries is that it is too expensive to have more kids than 1 or 2. You have both parents working, living in an expensive place, paying for day care, buying a bigger car, etc.. you end up needing to be well above middle class to actually afford 3-4 kids.. then when you factor in university, then you basically need to win the lottery or your kids end up with tons of debt. And in many cities in China you effectively are an immigrant in your own country trying to become a resident where you can access the cities better healthcare system and schools etc.. so people have a big incentive to want to live in the bigger more expensive cities. (Which ends up making it more likely people will have fewer kids)
One possible solution for this is to encourage employers to install more charging stations at their office -- which would mean a more consistent load throughout the day as tons of electric cars charge during the day. This would also be a huge benefit to employees who could effectively drive their electric car for free and the power company also gets the benefit of people putting less demand on the residential power grid.
Yea and apple likely can write almost whatever terms they want.. basically everybody is actively scraping their content for free, and Apple is like, hey wouldn’t you rather get paid? Super similar to the days when Napster was the way people got music online and then Apple said hey wouldn’t you rather get $1 per song? Then once the precedent is set then they can go around saying look at all these freeloaders.. so only the companies with deep pockets can afford to pay the AI license fees..
For those wondering: this is here:
https://www.google.com/maps/place/42%C2%B041'08.5%22N+87%C2%...
and the land is located adjacent to the land that microsoft took over to build a huge datacenter in the previous location where Foxconn was going to build a giant LCD factory (or at least said they were going to but never did). So the value of this mostly comes from the fact that it is adjacent to an already huge chunk of land that Microsoft owns. Also likely Microsoft is getting some tax breaks for this.
Has there been a writeup on Vancouver BC and their taxes / rules etc they've imposed and outcomes? Would be interesting to know if that has moved the needle at all? It would be also interesting to understand how we could differentiate between "I'm buying a house to live in for a pile of money and hope that at worst it's price follows inflation so that my large pile of money does not devalue.." vs. "I'm an medium to large size investment company that purchases your house with the goal of making it an airbnb or doing some minimal level of work to spruce up the house to then rent out to corporate housing for a local big-tech company down the street for 8k per month"..
AI safety allows AI companies to tell governments and worried people to not worry too much while they plow ahead doing what they want. That’s not to say there is nobody at these companies actually being responsible but having to open up your company to the equivalent of nuclear watchdogs is not something companies would be excited about. Just like nuclear weapons and nuclear energy we benefit from the tech as well as get greatly exposed to a huge disaster.. so let’s just call it as it is.. somebody is going to open Pandora’s box, we should be thinking super hard about how to control it and if it’s possible.. otherwise trying to ban it—is that even a possibility?
The acid test for this is, if its OK for the FBI to spy on you as a citizen, then cool, we need the right to spy on the FBI agent and his family members -- why does he get to spy on me and my family members when there is zero reason for them to do that? I don't have anything to hide, and neither should he, so its all kosher right?
I really wish these articles would figure out a better way to say “this year will be the warmest year on record and our records go back 125000 years”, not that 125k years ago it was warmer than now..
For the mac revenue big drop over the past year -- wouldn't the idea be something like "raise prices, lower revenue, but higher margin -- and therefore higher profit total than previously?" I wonder if that is reflected anywhere...?
Yea the costs for sure will vary a bunch -- but just searching around a bit I found this:
https://pedestrianobservations.com/2012/09/25/relative-costs...
That cites some Danish study about elevated vs. non-elevated costing roughly 2-2.5X. Really my point to highlight is that building on the ground ends up in reality being far more complicated than piling up ballast and sleepers on the ground and mounting tracks to that and assuming you don't have many roads. Of course if you are crossing vast areas with virtually no roads or infrastructure, yes use the cheapest option, but in the areas where its just one complicated crossing after another, then you end up with tons of "mitigations" as they call it costing tons of money -- and when that is properly factored in, the cost difference is far less than expected.
That's great that France has the TGV mostly on the ground as they exit the suburbs of every city -- whats the typical configuration of the crossings for roads? Does the road basically go in a small tunnel under the crossing or does the track become elevated temporarily? Or do they actually allow cars crossing the tracks at grade?
I have observed high speed rail lines in Asia and Europe and there is one commonality — a very large percentage of the rail lines are built either on elevated tracks near populated areas, or underground for sections. While this certainly adds cost, it seems to me (would love this to be confirmed by somebody who knows) that the costs are more likely to be closer to the projection. It appears in the US many of these plans rely upon lower costs where the train is built on the ground. So they end up building tons of small bridges, regrading roads, moving wires, moving drainage, communication equipment etc as the train interacts with existing infrastructure. All of the unknowns of hundreds of miles of that ends up costing tons of money and risk. So you have some consultant saying “well elevated costs 2x what doing it on the ground does” or something and in the end they build it on the ground for 4x the original cost due to having to handle zillions of little motivations all over the place. My take — just elevate it and simplify the situation. It ends up being basically the best of all worlds in the end.
"If you’re building an AI native product, your primary goal is getting off of OpenAI as soon as you possibly can."
Right now you essentially have:
Customer -> business -> open-ai -> microsoft_azure
Where-as many companies don't have this extra middle-man like this now -- they are more used to this type of situation:
Customer -> business -> microsoft_azure
Hmm.. :-) seems like microsoft just absorbing the tech of open_ai into a set of models that you can use / train on etc into azure proper and forget about the middle man is likely the end game here.
“Microsoft's next console after the Xbox Series X and S would represent not just a new device but a new "ecosystem generation,"…”
“According to the presentation, Microsoft had already selected ARM64 architecture for the core device's CPU. The company expected to deliver development kits to game creators by 2027, ahead of a 2028 launch.”
If one thing comes from ADU related wins being that there is a decoupling of mortgage cost to rent then that should be massively encouraged everywhere. In fact give tax breaks to people who build them even. Basically the idea that in a time when mortgage rates are 7% plus on a 1M house, and somehow you can find a rental for less than whatever that would cost is so unlikely that you have tons of people fighting for the rare case when that happens. But with an ADU it’s just a random additional space that could easily house a couple of people and charging a low rent and low utilities for that seems super likely. In particular imagine the amount of older people who have a large enough back yard and might benefit a bunch from additional income and having a neighbor close by to look in on them occasionally.. seems like with the aging population in US it’s a win win.
Good point -- one thing that would be interesting would be a postmortem at some point from a group of CEO's or somebody "in the know" who is able to get to the root of the issue so we can go beyond speculation. These people are typically in power positions that don't require explaining so often we get phrases like "I've thought hard about it and I've made a decision", where the real reasoning is left for people to guess at. Probably the main data we will eventually get will be when there are 2 companies that are in similar areas where one is full remote and the other has big office buildings in a downtown someplace.
Seems most of these top down driven mandates could be explained by some simple logic—if you don’t like your job very much as CEO, then you are thinking “man I wish I didn’t have to be here”.. and then it starts resentment toward people who also aren’t suffering along with you.. basically these mandates seem likely to be driven by stress at the top of the management, not based upon actual data. If you are asking for data that means you aren’t in charge sadly.
No matter what happens with riscV having any amount of control or a board seat or large amounts of votes on what ARM does is just smart given that there is a big risk that ARM when public decides “ya know, going forward we want to change terms for our license to start making some real money..”.. basically that one choice could dramatically impact any ARM licensee company with big exposure to license fees. So it’s a smart idea to have involvement in ARM. My hope is enough companies do this that they decide to keep it private and make this an open consortium that benefits all licensees.