harshaw

one thing to point out: with a planet that is 20-30% larger than the earth the rocket equation makes it almost it impossible to get out of the atmosphere with chemical rocket technology. So even if you could visit, we could probably only have drop ships or rovers. coming back out of the gravity well without space elevators or stuff would be very hard. It also means that life on those planets would be hard for them to establish a space faring society.

A bunch of people are commenting that this could lead to grid-death spiral where people opt out of the grid because they can just use solar + batteries. Maybe, but I doubt in Vermont. Vermont is really hard for solar because of tree cover and clouds (according to my conversations with a solar installer). Secondly, in this setup you have buy your own power plant (solar + batteries), which likely for most folks, is a pretty big expense. Green Mountain power subsidizing the cost of buying the Tesla power walls is a big deal for me because I use this setup for a cabin / second home and want to keep my costs relatively low. I can't do solar because I on the side of a mountain and lots of trees.

I couldn't read the article because of paywall issues. I have a Tesla powerwall that I pay $50 / month to lease from green mountain power. I have piece of mind for power in a rural area - the power company get peaker benefits. Seems like a great deal. I wish we had this in Massachusetts. Mad river valley area.

Ex AWS. AWS had one zone specific products. If those were in the affected AZ that data is lost. AWS has multiple multi-AZ products. S3 and EFS are good examples. If you loose an AZ for an extended periodic of time, the dataplane of those services will migrate shards around until you get back to your durability goals.

if you use these things to generate design docs / text, it should be good news if it is actually better at prose as advertised. Some people like sol for prose better the anthropic models.

This is somewhat orthogonal to the article, but the whole bubble on AI data centers seems to presume that the need for compute is so massive that it far exceeds the expected optimizations we would expect with at scale inference (PIM, ASICs, etc). I would expect that there is a set of optimizations like this one (or variations) that would someone negate the buildout. But it's not really discussed.

Some other commentators said this was AI slop. I don't have enough expertise to say if it really is, but it sure has the "claudish" cadence of AI generated text that makes it hard to take it seriously. The authoritative and strong statements, the use of phrasing with colons and dashes, the use of bold, etc.

I've heard it's writing style referred to as "Claudish". I agree that it is very hard to read and creates work for me.

I really wanted to stay in the google chromebook / googlebook echo system. But the hardware was expensive for what you get. Apple announced the macbook neo and I picked one up. Great hardware. can run light weight mac software. I don't run much beyond chrome and wahoo SYSTM (bike trainer app). It's really solid hardware and cost $600 or so.

I use gemini extensively (and claude). But - do I need this integrated in my laptop? Don't quite see it. And it's hard to beat Apple on hardware now.

this post will probably never be read but.. I was on the team that was trying to make the marriage of S3 and EFS work a year ago. it's a pretty hard problem. At one point we proposed this solution (which seems like a caching layer) but it got shot down for a more complex system that would have attempted to rebuild EFS on faster S3 blob storage. I left before this engineering monstrosity made significant progress, and it clearly died at some point.

Looks like they went back to a simpler solution they could deliver but with some obvious warts. good to see something get launched but the sausage making her was brutal.

The reality is that if you read https://www.allthingsdistributed.com/2026/04/s3-files-and-th..., it sounds like the great minds at S3 figured out that a caching layer was the way to go. We (EFS) fucking proposed that years ago. But we had to deal with Seattle and the S3 braintrust who didn't want to do that. I know we wrote a PRFAQ that was close to this concept probably four years ago. The political story is that EFS was taking over by S3 and the EFS folks didn't have the agency or political backing to build a more workable solution. So we wasted a shit ton of time tackling something that was never going to work and many of the tenured EFS engineers left.

somewhat off topic, but I really am not sure that adding chromebooks to every school has made education better. hard to block youtube when they bring these home (I know you can, but the average person can't).

If you don't want to go through your insurance company you can check out an app we built called RoadClub. You get points for safe driving behavior - and you can get the hard brake alerts. Is it a bit annoying? yes. You can't just drive agressively - you need to give space to slow down. I still struggle with it.

Nice research. This is fairly well known in insurance circles. Most auto insurers that do telematics consider hard braking the strongest indicator of risk. One of the things that we do at work (Cambridge Mobile Telematics) is build tools to deal with this risk. We have apps that monitor driving and we play a tone to indicate that a hard braking event was detected. Simply letting people know that they had a hard braking event is an effective mechanism for behavior change (other companies have similar tech)

I am confused about what Tesla is doing. They have effectively two automobile products now with one failed product (cybertruck). reading various articles about this doesn't make it more clear. Do they not want to be a car company?

Full disclosure: I am pretty sour on the current Amazon/AWS leadership as I think, well, they couldn't lead a company out of a paper bag (former manager at AWS). Is there data that Amazon/AWS is still hiring junior devs? I've heard it's very hard to get into student programs these days but I don't have the data. My grumpy position would be Garmin saying one thing and doing another.

Why? I look at this. I want engineers to use the tools on the stuff that AI is good at so we can do more high value work.

If I were to summarize how we attacked this when I was on AWS (different team)

formal methods. Some of this started a long time ago so not sure if it was TLA, TLA+, or something else. (I am a useless manager type)

fake clients / servers to make testing possible

strict invariants

A simulator to fuzz/fault the entire system. We didn't get this until later in the life of the service but flushed out race condition bugs that would have taken years to do.

We never got to replaying customer traffic patterns which was a pet idea of mine but probably the juice wasn't worth the squeeze.

i'll bite. Please give me an example of questions that are scientifically validated.

not quite true - there are some regions that have a different set of AWS users / credentials. I can't remember what this is called off the top of my head.

I think this is an astonishingly dumb take. Regardless of what you think of Musk, SpaceX is building a fundamentally important reusable lift technology that can be the underpinning of some many future developments. Who cares if China gets to the moon first? This is how NASA historically gets into a mess with its launch system - political pressure, conflated goals and requirements (see the Space Shuttle - does it launch people?, military payloads?, oh goodie - let's do it all). If anything I wish NASA would do more to make sure we have a decent starship competitor which its hard to see blue origin being anytime soon (but I am not an expert on this topic)

Sounds more like an object system (immutable) with the veneer of a file system for their use cases. I sort of read the doc - sounds like data is replicated and not erasure encoded (so perhaps more expensive?).

I think many people have said this, but "file systems" get a lot easier if you don't have to worry about overwrites, appends, truncates, etc. Anyway, always interesting to see what people come up with for their use cases.

I did Symbian programming back in the day. IIRC the Sony-Erricson P800. It was not developer friendly. The memory management model was hard to program and could crash easily. Also did some work with Nokia on the N60. I was working at Orange at that time and we had hired a contractor to integrate push to talk (because for some reasons Orange though push to talk would take off in Europe, lolol). I got a couple of free trips to Tampere to theoretically help Nokia debug this third party push to talk app. I seem to recall Nokia not being too enthused to work with some pushy American hacker who wanted to open a debugger and fix things. I remember that we had some nice reindeer dinners.

Early mobile is littered with dead operating systems. not really that surprising. PalmOS, Symbian, SaveJE, windows mobile, etc. not worth crying over.

My problem is that there is a reauth FOMA that gets copied with the most trivial of applications. A good example is the Electrify America charging app. It's job is to be available so you can charge. But they log you out frequently - and they want to do second factor with email. Guess what - that doesn't always work. My wife was trying to charge the other day, was logged out, and couldn't get the email verification to go through. So I had her login with my credentials and I answered the email verification and gave her the token over the phone. super annoying.

But more importantly- mobile phones already have good security mechanisms. It's like all these shitty apps copied web based auth mechanisms with timeouts when they could do something better (and probably are built on web technologies with cookies instead of using the trusted store on the phone).

There are precious few apps out there that tell you ahead of time that reauth is happening (Zoom does this - kudos). But even so - I don't think it's necessary most of the time.

This isn't so much about the Author's story but the space. He is correct that marketplaces are hell, and especially auto repair market places. I was one of the several devs that went through trying to make Openbay work (a US based market place for auto repair). We actually did have service providers signed up and we did have a way to acquire customers who wanted auto repair. So you had the illusion of product market fit - but the problem is that it's really really hard to get people to actually click "buy this brake job" and then, more importantly, they have no reason to come back to your app because 1) you don't need brake jobs all that often and 2) they can just go back to the service provider. And many shops are happy to still take phone calls as their default way of booking work.

The reason the company existed is the rich founder was upset that a shop wanted to charge him a fortune to get his BMW M5 repaired. He wanted better quotes. So we built a marketplace to get better quotes. But that's not what real customers want (because most people have Toyotas not M5s). And also we didn't do the customer development / research to understand how repair shops work. You want to know how my repair shop manages their repair schedule? They have a paper calendar and write down your phone number and the job. sure there are better ways to manage the work - but this paper mechanism has worked for them for years and why change it? And you know what - I go back to the shop all the time because I trust them. Ultimately people tend to have a fairly personal relationship with their local mechanic. You can build a leadgen product but the ultimate relationship is between the customer and the repair provider.

TLDR - everyone should understand the lean startup. /working backwards model and relentlessly focus on the customer.

They should make anyone who denies anesthesia for a colonoscopy drink that awful orange flavored drink you take the night before. This is just like... basic care.

Random comment: slide 10 talks about using Twisted and then adding a twisted middle layer. 2010 me (or earlier) liked Twisted a ton but not sure I would have foisted this technology on a development team. As a single dev doing all kinds of crazy shit it (sort of) worked but was pretty hard to debug or understand.

There is room in storage for many kind of products with different sets of tradeoffs. And don't underestimate the ability of a startup to move much faster than AWS.

What company?

As an amazonian, this has come up during the five day RTO discussion. As a manager, I can't imagine a more obvious way to destroy the service that I help run and then really risk losing customer trust.

I we are all cogs at some point but I really have trouble being this cynical.