> Is this the kind of use case that is seen as valuable?
I think it could be. Consider an argument like this:
It's valuable to ask ChatGPT questions and receive text responses. Some of the responses are more valuable when they don't just return text, but some markup: bolding, adding visualizations etc. Why can't some responses be more valuable if they return little apps?
One place where I've wanted this myself are with using LLMs for long-running goals I have. For example, I do my blood work about once a year, and I use the results to make changes and track. For a long time I had a long chat thread with ChatGPT. Now I have a little app instead.
An extreme version of this starts to turn responses into more and more fully-fledged apps. I did an experiment recently with creating a personal finance app. I found customizing the app to my specific needs made it much more valuable to me then generic personal finance apps, which have much more effort put it, but aren't tailored to my needs [^1]
The economics don't work unless Starship is doing flights in quantity, and it has met or exceeded its cost targets.
Roughly, a single rack plus solar to power it in the $15m+ range just to launch. (This assumes power dissipation is handled via some means that does not require launch to orbit. Also does not include batteries.) Choose your own hardware for the rack, but call it < $5m.
SpaceX earning $15m every time someone launches a $5m rack would be a great business for SpaceX.
Use your own calculator/LLM, but mine is suggesting that the ~$7B Colossus 1 data center in TFA would be around $50B if launched on Falcon 9 (still ignoring cooling and batteries).
(There are obviously a lot of other asterisks. I'm ignoring power storage and heat dissipation. Maintenance probably doesn't matter given 75% of cost is in the launch. Network bandwidth could be a problem considering how DCs are used. Competition - if Company A spends $100B for $25B of actual AI infra, how competitive will they be against Company B who gets $100B for their $100B by spending it in Canada or Mexico, which they can do right now? Etc.)
None of this works without Starship, which has not set a date for its first LEO insertion test yet. Yet the whole point of orbital DCs is nothing on the ground can move fast enough, hence the rush to orbit...which can't really move at all right now.
Early in my career, people said this about programmers who (weakly) insisted on using assemblers.
Then, about people using high-level languages like C.
Then, about people using C++.
Then, about people using "toy"/"scripting" languages like PHP and Python.
About people who use ORMs instead of writing SQL directly.
About people who use JavaScript ("not a real programming language" was the dis).
People used to argue how it was the mark of a tourist to use anything more visual than Emacs.
This slight won't stick, nobody cares, and it might end up sounding stupid later. You can't usefully insult a professional engineer in 2026 by pointing out that they haven't memorized ASCII or the Arm instruction set.
No hate to any of the PaaSes out there, Vercel included. They truly serve a need.
That said, if you are an engineer planning on working in/around the field, I would strongly suggest developing some competence at basic Linux systems administration. (Also: learn SQL, even though it's out of fashion.)
Linux is probably the single technology where my knowledge has had the longest useful lifespan (SQL is probably second). There are Unix (System V) bits I learned decades ago that are still useful today, on Linux.
Then, you can use a PaaS if you want. But if it's not the right fit, you are in a position to do something else. You might find that designing your application with a modern compute stack (this is not a PaaS) gives you an unfair advantage.
There's now ~2 generations of professional engineers for whom SQL was rarely/never a thing to learn. Between the hard(er) split among front-end/back-end developers, ORMs improving, and the (flawed) idea that NoSQL would make SQL irrelevant, it has become somewhat of a niche skill.
Think about Firebase. One can be full stack on an app built on Firebase and be successful without ever touching SQL. Firebase is very popular, and has been for some time.
Source: I have worked with a set of otherwise solid engineering teams and can say that SQL familiarity has given me a leg up on very smart engineers who nonetheless do not do relational databases.
The only way you can get away with creating an application without touching sql is of you offload the logic to your backend language, and then I don't think you'd be efficient enough to scale.
Also can someone actually understand the logic of joins, indexes, pks, etc enough to create an efficientand scalable db, and not simply have learned sql by proximity?
> The only way you can get away with creating an application without touching sql
Please look at app platforms like Firebase[1]. There are absolutely complex Web applications running at scale that do not use SQL anywhere in the stack.
Aside from that, MongoDB and Redis are 17 years old; CouchDB is over 20. NoSQL is well-established at this point. All of the hyperscalers offer proprietary NoSQL databases, and have done so for years. A large number of developers uses those databases in production.
In our API-centric environment, there are a lot of apps that don't do much in the way of managing their own data directly at all, using mixtures APIs for auth and other key application functions.
> can someone actually understand the logic of joins, indexes, pks, etc enough to create an efficientand scalable db
If you are not using a relational database, these concepts do not necessarily apply.
I did some interviewing rounds with PaaS platforms for advisory roles. I loved Heroku and was ruined by Heroku so I thought maybe the industry has something new to offer.
The model is largely "built to be locked in" model. It is not something innovative. The issue is that to what scale of operation the platform considers there customer to be locked in where they can hand them a bill that compensates for the ease of entry model for all the hobby/free tier.
With Vercel I feel like these level is becoming lower and lower. You can within minutes launch a full startup with Vercel and AI assisted coding. And Vercel assumes that as long as you do not recieve any traffic that is good. The moment you recieve even a mild amount of traffic you are considered locked in.
To some people that is a fair trade because they have so little trust in their products in the first place investing hours instead minutes is a fair trade. If the traffic comes they are already in the green. If the traffic does not come any effort they have put in puts them in red. So, you put as little effort as possible to get thing out there.
I like Vercel. They have figured a monetization model for Slop SaaS. The other PaaS needs to catch up. In 2026, PaaS exists as a model to make revenue out of slop.
If everyone could make these machines, there'd be more of these machines.
There are so many examples of this out there, already, that I find this specious "no next generation" argument to be either simply coming from bias, or ignorance.
For sure, we only care about Taiwan because there is one Taiwan. End patents: no more Taiwan problem.
> If everyone could make these machines, there'd be more of these machines.
My post is in violent agreement with this, for this generation of machines.
ASML spends ~$5B annually on R&D with the expectation that they will be able to make ~30% net profit in the future. If you remove patent protection, there will be more competition and obviously profit margins will fall.
I want to rephrase that for emphasis. The point of aa-jv's post was that we would get cheaper chips by invalidating IP. Cheaper chips means lower margins (because you have not lowered input prices). Lower margins was the explicit goal, so to the extent that the changes in IP law work, you will get lower margins for companies like ASML.
At that point, you have a field of companies looking at (say) 10% net returns, still needing to invest billions of new capital into R&D every year. Worse: no patents means that Company A could spend $5B on R&D and Company B could spend $0, and both of them could reap the benefits of that $5B by Company A. So it's not even necessarily clear that the industry would see much net innovation.
Are we even certain there are companies who would enter this capital-intensive business assuming IP was free? Compulsory licensing is a thing, but I am not aware of that even being something that has been requested.
That article doesn't give a timeframe, but most of these use 10 years as a placeholder. I would also imagine it's not a requirement for them to spend it evenly over the 10 years, so could be back-loaded.
OpenAI is a large customer, but this is not making Azure their personal cluster.
> Modern web application is essentially 2 apps (frontend and backend)
This doesn't have to be the case, though. We all use e.g. banking & telecom sites that worked fine before being rebuilt using "modern architectures" and now are memory hogs. There is no reason many/most sites cannot go back to simpler architectures (other than programmers have forgotten how to build those sites, which is in fact a real problem).
I do hear the argument that when you start building, you don't always know whether this simple brochureware site will evolve into a multimedia player platform, so just start with the assumption you will need the capabilities of a complex FE stack. But in the majority of cases this will end up being the wrong choice, and clients/employers will end up spending too much money managing too much complexity.
This is precisely where the publisher has the most control over the user experience. Putting load on the browser makes a user's experience dependent on their hardware & software stack.
Part of what makes a good user experience is is working nicely with the users hardware and software stack and that's much easier on the client.
The user would like the website to have native scroll physics, respect their system preferences, react to changing window sizes and with different input methods, screen readers and so on.
If the key to a good user experience was server side control, than the hallmark of a good website would be being an RDP stream and prefers-color-scheme wouldn't be a CSS feature but a HTTP header.
> The user would like the website to have native scroll physics, respect their system preferences, react to changing window sizes and with different input methods, screen readers and so on.
Yes, these are features implemented by every popular browser via HTML & CSS. Fancy front-end work frequently breaks these features.
> key to a good user experience was server side control
I think you're reading my comment from an extremely front-end perspective. I simply mean that where possible it's better to render pages and do logic on the server side versus on the client side. The same HTML + CSS is generated either way, the only question is what % of the work is done by the user's device vs what % is done in a data center.
Since others are not saying it, enforcing this will immediately cause havoc as any number of citizens do not have ready access to any document proving citizenship.
(Non-US people note that this is likely a major difference between the US and your country. The US does not compulsorily provide proof of citizenship to its citizens that can be used at places where one is typically asked to prove one's citizenship.)
Bessent notes here that Real ID would not be considered valid ID for this purpose, which sounds like it will have the same problems as the SAVE act. This could mean debanking anyone who has changed their name and does not have a notarized copy of the name change certificate, and most people who do not drive.
(I am not sure how it would handle minors, who generally do not have any photo ID. Would they have to come in to provide ID when they turn 18?)
The underlying idea is fine, but it creates problems when combined with the reluctance to issue any kind of national ID.
I don't even think it's about partisan tilt anymore. This administration's M.O. is raw chaos, havoc, and just this low-level randomized churn that keeps us all conditioned to believe that nothing in government works deterministically anymore.
> If they don't have documentation, are they citizens?
Yes. As OP said, "anyone who has changed their name and does not have a notarized copy of the name change certificate, and most people who do not drive." Note that the first category includes many married women.
> All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside.
This is part of the US constitution. There's no "if they have proper documentation" qualification.
If the US Constitution is to have force as the core document foundational to the governance of the US, it is important for its clear text to have the force of law.
An executive agency creating new requirements for citizenship has the effect of overriding the Constitution, which brings into question what are the controlling documents for the country.
You are required to prove your citizenship to the government (by proxy of your bank or otherwise). The government lacks a unified document of identity which would by law act as a proof of citizenship, and reserves its right to call any other document it is issuing to be “insufficient”.
Yeah I don't think people are really fully appreciating the scope of this, because it means people would essentially have to have a passport to open a bank account.
It's very dark. I tend to be libertarian about these things and feel like it's none of the government's business. Get a warrant and do your investigations if you want to prove someone is a foreigner up to no good. There is no real problem unless you're xenophobic or racist.
So I don't agree the "underlying idea is fine" at all. This is a step further though, by putting an administrative and financial burden on people to have a bank account.
The fact this is normal in other places in the world doesn't make it ok to me either — two wrongs don't make a right. And in any event many other places are more socialized than the US, so there isn't the same kind of burden on many places as there would be in the US. It would be one thing if the administration were bending over backwards to provide public healthcare, expand education and public research, but they're doing the opposite.
> I don't agree the "underlying idea is fine" at all
I gave you a shout out! :-P
> the reluctance to issue any kind of national ID
Americans have tended to resist this kind of surveillance (when done by the government). Honestly, because it's not necessary. It doesn't make sense to tax 350 million people when DOJ usually doesn't even go after the known big fish. Or when companies can openly violate e.g. money transfer laws at vast scale until they get rich enough to get the laws changed in their favor.
This feels like the kind of thing that will blow up if they implement it and then have to be kicked down the road forever, like RealID. Old people know that the initial RealID deadline was before Barack Obama's election.
Generally will leave as an exercise for the reader.
But immediately one can say that most minors will not have the requisite picture ID because they do not drive and we are not required to carry picture ID (this rollout would be touch more people than the requirement that drivers carry ID). So as of right now, most minors in the US cannot prove citizenship under the criteria Bessent is suggesting (yes, the country should be debating this).
Let's call it all the people under 15 so we don't get the "akshully learner's permit" folks objecting. The US has ~60 million people in the 0-14 age bracket, apply whatever ratio you want to that for citizens/noncitizens and you are still going to end up with a lot, likely millions, of people.
Minors aren’t allowed to open bank accounts without their parents being on the account. Unless their parents are also in that age bracket, but that’s biologically unlikely
You asked who doesn't have the required documentation, I am telling you it's minors. Saying parents are also on the account does not change that as of this moment, those documents do not exist and therefore have to be secured.
> Unless their parents are also in that age bracket
This is irrelevant because the point was to identify a broad population that currently does not have the relevant documentation. That's people 0-14.
> Since others are not saying it, enforcing this will immediately cause havoc as any number of citizens do not have ready access to any document proving citizenship.
I didnt have all the documents available for my Real ID which has quite the requirements. In the limit, at least as many as any other citizenship proofing task. We can assume the greatest difficulty would be for the homeless.
It took me ~15 minutes on the social security admin website to get a card ordered to me because mine is lost somewhere in a safe. I had it sent to my house, a PO box, homeless shelter, or any other location would work too. Can be done via a library if you're homeless. Zero excuse.
It took me ~20 minutes to figure out which hospital I was born at and get a copy of my birth certificate shipped to me. See above. Likely marginally more difficult for a homeless person. Not terrible difficult though if you're not so cracked out you don't remember even the state in which you were born. Again, zero excuse.
It took me ~30 seconds to find a document to prove my current residency. Trivial for a homeless person as well. Zero excuse.
Again, in the limit, the government should provide an easier way to do this. But the pearl clutching over the difficulty is to vastly overstated.
This is simply a fantastic excuse to not require citizenship for yet another thing. Something absolutely unheard of in other western countries. I'm beginning to think all of this avoiding proof of citizenship has an ulterior motive.
The point is that there are hundreds of millions of consumer bank accounts in the US, and it's not clear that Treasury appreciates the turmoil they are proposing. The country has not had a debate over this, it sounds like it might just drop out of the sky one day and create unnecessary chaos.
We can use the rollout of Real ID itself as a gauge. Executives of both parties, and several Congresses, landed on 20 years as an appropriate rollout time to do so smoothly. And that's basically only needed for air travel, which most Americans do not do in a given year.
It's not crazy to ask that a more disruptive change be subject to more scrutiny and deliberation about its rollout.
In your case, everything was straightforward, you already have a license, and your bank is local so you can walk in and show your ID, awesome for you. But over hundreds of millions of people, every edge case will present. (Is it okay for banks to freeze assets of people in hospitals who are unable to perform the necessary steps and present themselves at a bank? Inmates? How are joint accounts handled? What counts as bank account? What happens to money currently held legally here by foreign nationals?)
The one that might affect the most people here: if you have to show ID, presumably the bank has to be able to authenticate it against your person. Which means an in-person visit. This would be bad if you are one of the tens of millions of Americans whose primary bank does not have any branches in their state of residence. I bank at my alma mater's credit union, even though I have not lived in that state for decades. Would I need to travel there to show my ID or have my account frozen?
Again, a bipartisan set of Congresses and Presidents landed on 20 years to rollout when the only real penalty would be some people would not be able to board a plane when they wanted to, without extra scrutiny.
A botched rollout of this could lead to unpredictable financial calamities as rents and other bills go unpaid, etc.
There is simply not an emergency here, we don't have to upend our financial system pretending there is. The ulterior motive here is to preserve the stability of our financial system while making changes.
Is this the kind of use case that is seen as valuable?
I joked a while back that LLM-brain was going to have people building bespoke apps on each HTTP request, and people thought I was exaggerating!
reply