It usually works well on Android. I use it on a few forums.

I’m seeing some complaints about delivery times though not being immediate.

That’s the same with every Android app, though. When you’re not actively using your phone, some apps are put to sleep, which also stops their notifications. Very common on Samsung phones especially. You can usually add an app (or installed PWA) to a list of apps that you don’t want to sleep.

What’s the downside to having one’s phone number in the public directory?

The difference is that an IP of a VPS doesn’t directly connect back to you. It’s in the provider’s name. Some providers let you change your IP address to a different one for a small fee.

The reason I mentioned the dot com bubble is because a lot of the companies back then failed because they couldn’t figure out a sustainable business model. It was mostly hype-driven with the idea of getting users first, then figuring out monetization later.

That’s why we have ad-supported sites today. It was the main business model that was the most sustainable.

There were a lot of small sites, sure, but a lot of them were hosted on services with no real business model. Even back then, not a lot of people self-hosted.

you lose access to push notifications

Web apps have supported push notifications for a long time now. I think even Safari supports them now.

Twitter had some great outcomes when they rolled out their PWA: https://web.dev/case-studies/twitter

Twitter Lite is now the fastest, least expensive, and most reliable way to use Twitter. The web app rivals the performance of our native apps but requires less than 3% of the device storage space compared to Twitter for Android.

65% increase in pages per session
75% increase in Tweets sent
20% decrease in bounce rate

yet they kept pushing their native apps, probably because they can collect more data through them. The web is way more sandboxed than regular apps.

Just give me a mobile web page if you’re going to do that shit.

There’s some apps that just load a site, but the site refuses to load if you load it in a regular browser? Why?? Spoofing the user-agent would probably work around that, but I haven’t tried.

Document? Sprinkler app. Web Page? Sprinkler app. Installing from a source other than Google? Oh you better believe the sprinkler app can do that.

Android apps tell the system which URLs they can open. If you click a Google Maps link, it can prompt you to open it in the Google Maps app. It sounds like whoever created the sprinkler app misconfigured the app and it’s saying that it can open all URLs, not just the URLs it cares about. They probably read a tutorial about how to make a webview in Android and didn’t know what they were doing :)

This is why everything apps are so popular in many parts of the world

What’s wrong with a web browser? I know it’s not as seamless, but it’s far less limiting and literally any company can create a site, regardless of their size. There’s systems like Google Pay that avoid you having to enter your credit card details on every site.

it seems pretty sub-optimal for a personal site to be publicly associated with even a permanent IP address

What’s the downside you see from having a static IP address?

I don’t see any way to achieve this without a CDN, unfortunately.

I think you’re looking for a reverse proxy. CDNs are essentially reverse proxies with edge caching (their main feature is that they cache files on servers that are closer to a user), but it sounds like you don’t really care about the caching for your use case?

I don’t know if any companies provide reverse proxies without a CDN though.

Would you prefer Mozilla to not exist? They’re trying to find revenue streams other than the money they get from Google.

the original ublock.

You mean the original uBlock Origin. The original uBlock has been gone for a long time.

Sounds like you’re forgetting about the dot com bubble. The internet wasn’t fine abck then because nobody really had a sustainable business model.

That’s not Cloudflare-specific; you can use any CDN that supports origin pull in the same way :)

It’s not ideal because… Cloudflare… but at least you’re using standard web tools. To ditch Cloudflare you just unplug them at the domain and you still have a website.

Definitely agree with this! That’s one of the pain points of “cloud” services - they really try to lock you in, making it impossible to swotch.

without having to deal with LetsEncrypt.

You still need encryption between your CDN and your origin, ideally using a proper certificate. Let’s Encrypt (and other ACME services like ZeroSSL) are pretty easy to use, and can be fully automated. I’m using Let’s Encrypt even for internal servers on my network, using a DNS challenge for verification instead of a HTTP one.

Perhaps its irrational but I’m bothered by how many people seem to think that Github Pages is the only way to host a static website

It’s strange because out of all the possible options, Github Pages is the most basic. You have to store your generated files in a Git repo (which is kinda gross) and it barely supports any features. For example, it doesn’t support server logs or redirects.

I guess it’s popular because people already use Github and don’t want to look for other services?

You seem to recommend a VPS but then suggest a bunch of page-hosting platforms.

Other comments were talking about pros and cons of self-hosting, so I tried to give advice for both approaches. I probably could have been clearer about thay in my comment though. I edited the comment a bit to try and clarify.

I have some static sites that I just rsync to my VPS and serve using Nginx. That’s definitely a good option.

If you want to make it faster by using a CDN and don’t want it to be too hard to set up, you’re going to have to use a CDN service.

Self-hosted CDN is doable, but way more effort. Anycast approach is to get your own IPv4 and IPv6 range, and get VPSes in multiple countries through a provider that allows BGP sessions (Vultr and HostHatch support this for example). Then you can have one IP that goes to the server that’s closest to the viewer. Easier approach is to use Geo DNS where your DNS server returns a different IP depending on the visitor’s location. You can self-host that using something like PowerDNS.

Definitely possible, but I think WordPad in Windows 95 was written from scratch.

You don’t have to use a third-party short URL service. It can be hosted on your own site.

A lot of people are already using a third-party short URL service like qrco.de because they don’t realise you don’t actually need a service like that to make a QR code.

WordPad didn’t exist until Windows 95. You might be thinking of Microsoft Write, which predated it.

WordPad in Windows 95 was a demonstration of how to use the rich-text editing component built into Windows. Its C++ source code came bundled with MFC (Microsoft Foundation Classes - programming library for making Windows apps using C++) as a sample.

The fact that it was a useful tool for end users was essentially just a side effect.

or the url has a ton of tracking params appended to it for some reason

Ideally you should use a short URL that redirects to the full URL. The tracking parameters should be on the long URL, not the short one.

Nintendo could try make up something like “it’s not computer software since the Switch is a console, not a computer” or something like that. Not a great argument, but they have good lawyers and could probably convince a court that it’s true.