>Usually when people talk about grand things like changing "the future (business model) of computing," they're full of it. But not this time. Suhail has been working on this for 2 years. There's a good chance it's the new default infrastructure.

What is missing is 'business model' in parenthesis. Then maybe it makes some sense. This kind of managed service is definitely not for me, but it may be the future (business model) of computing.

Ultimately we get the technology we 'deserve'. General purpose computing was a short lived experiment but ultimately nerds could not convince or train enough lay people to appreciate the power it gave them. We're slowly sliding into a future where large corporations control basic access to computing. We're going to have to beg to be given permission to program our own devices. Its already happening on smartphones, tablets. PCs are next.

It feels like we are slowly losing power. Very slowly.

Once that happens, there will be more security and a more unified platform to develop on - no doubt - but I will be shocked if they do not also take the opportunity to lock down what software you can develop, publish, and run on Macs over time.

Subscription services (even for software on previously steady release cycles), serverless infra, cloud computing, etc... the overall business model is shifting back to buying compute time to perform a certain task, or these startups essentially re-selling general purpose compute time as marked-up specific purpose compute time. With some UI slathered on top to make it look sexy.

While a dev board would be cool, a Mac mini with the hardware UART enabled over a USB C cable is getting pretty close!

Trade offs like these arguably could help to keep general purpose computing alive, as they show it is possible to deliver a secure device to customers that can be used however you like, running whatever software you like, without iOS style "no choice available".

Dev boards aren't necessary when an entire functioning laptop only costs $999. An M1 dev board be would way, way more expensive than that and far less functional.

I'm imagining a future where sites become so bloated that you have to stream the rendered result to get anything usable.

In that way Dropbox seems just rsync; and there are more examples like this. Cherry-picking useful functionality from a relatively niche tool and making it significantly more accessible sounds like a big feature. Perhaps this is the future.

"When Silk is run in cloud mode, it can off-load some of the the parts of the browsing and rendering process to EC2. In cloud mode, EC2 acts as a sort of cache plus proxy for browsing, and it also can dynamically take over parts of the page rendering pipeline from the client."

[1] https://www.wired.com/insights/2011/09/amazon-silk/

[2] https://en.wikipedia.org/wiki/Amazon_Silk

Mighty is streaming a browser instance, wholesale, as a video stream. It's very very different both in philosophy and actual implementation.

Once one scratches into the technical details, the similarities end. Interesting, as you describe it, at a very high level, it does sound like Google Stadia or Nvidia GeForce Now but applied to web browsing.

Only time will tell if it was a genius or stupid investment.

> Even if you shutdown or switch to a different computer, your browser will always be on. Exactly how you left it. That means you’ll never lose tabs that remind you of something to do or contain an article you want to read later. You can quickly jump back to work.

I haven't seen that discussed here, but it seems key.

Of course Mighty is not the only way to implement it; if Mighty takes off — or even if it ends up looking like a good idea — watch Chrome natively implement something equivalent.

chrome://history/syncedTabs

Maintaining state across sessions by storing your whole browser on a server somewhere is a cute solution, until you need to access that server while traveling and now your ping to it is 300ms instead of 30.

Not a new idea but an obviously in demand one. As the saying goes, if you want to invent something that will be useful in 30 years, invent something that would have been useful 30 years ago.

Technology moves in sin waves.

Then there is the bloat issue. If websites were designed as lean as news.ycombinator.com, there would be no need for mighty. But they aren't. Instead we have js framework upon js framework, lots of tracking, ads, etc... you can make lean stuff using web technology, but you can't make it using the most popular frameworks. elm had this precise problem. They made their library really fast and really lean and people were like shrug, while flocking to the popular and bloated frameworks.

IDK it seems to me that Chrome and evergreen browsers have led to a stagnation in innovation of the browsing model. I feel like Mighty is a breath of fresh air. That being said, should Mighty ever become powerful, they'll likely benefit from funding maintainers to make js frameworks even more bloated so that even more people use their product... which means that they might do it. On the other hand, large user bases using products like Mighty might be a wake up call to website and browser developers that lean-ness is a real user demand, and maybe fixing it on their end, similar to how people taping up webcams led to builtin mechanical shutters.

If the browser becomes the main platform that people use to ship applications, this could be useful. And why would the browser be the main platform for apps? To bypass the app store fees, to be multiplatform, to be always up to date.

One final advantage to running the browser remotely is that you don't have to deal with a certain browser engine that is lagging behind the others.

But for people at the slow or low spec ends might often not be able to afford that kind of service, that lowest of the low-end might be a real focused niche... say people on airplane Wi-Fi or in remote locations on a satellite link.

but from a purely product marketing and psychological point of view I don't think that a product needs actual technical superiority or real measurable utility to become a big hit. I think it really only needs something that makes people want to use it. mighty could position itself as a sort of luxury upgrade for people with already good specs.

Because from the point of view people who are perhaps already of that successful and wealthy mindset many of them may consider that time is their most valuable asset and the accumulated frustrations and annoyance of waiting for websites to load is something they are prepared to pay a service to get rid of and to provide them and experience which they feel is more in line with their station and their expectations of Life in general.

Forget scraping, forget automated controls, forget installing extensions/plugins/userscripts if not allowed by the cloud overlord.

I can certainly understand the negativity.

As a niche service, why not?

It's bad enough that too much of the internet is in walled gardens. It's bad enough that one company makes most of the browser and search decisions. It's bad enough that phones have controlled app ecosystems. And just when you thought they were running out of controls to take, here's an idea where the browser won't even be on your computer. This is how you get a web that is optimized for a $50/month browser service where you can't even pick your extensions.

> How can we work towards a better world when people just shoot down any idea that they don't agree with instead of trying to expand their perspective?

For you to consider: Don't be so open-minded that your brain falls out.

I don't want it to succeed because it is dangerous.

Many cases of mesothelioma could have been prevented if we had listened to people who thought it was dangerous.

You can't conclude from asbestos that we should shut down new ideas on the first warnings, nor can you conclude from the history of aviation that we should disregard those warnings.

You asked:

> How can we work towards a better world when people just shoot down any idea that they don't agree with instead of trying to expand their perspective?

Going back to the original comment:

> Perhaps we need a new "test", like the "Turing Test", but this time for when humans can no longer tell the difference between new technology and old technology. "Mighty", for example, is just a "dumb terminal" - technology we had in the 1970s.

You want progress. We gain technological progress through a ratcheting action. A ratchet must close over the progress it has already made in order to ensure the mechanism only moves forward.

Likewise, we only progress if the new thing is genuinely better than the old. That requires that you understand the old thing.

Do you think the metric system is better than US standard measurements? Riddle me this: why is a mile 5280 feet? What's special about that number?

There is a reason for that number. There's also a reason it probably doesn't matter in modern usage, but unless we understand why we don't need that special property, we can't be sure that the metric system is actually progressing over the older systems and not regressing.

That's why you get this comment and others pointing out that this is simply tech from the '70s. If it's 50 years old tech, there's no progress going on. They are not failing to expand their perspectives, they are recognizing that this stuff is something they're already familiar with.

In this case I don't think this tech really qualifies as an innovation, it is mostly the wrong patch to fix a real problem, bloatware.