There was a time when local apps written in binary compiled languages were notably faster to run, but notoriously difficult to write and scale across platforms. Both of these have improved but the JS/CSS run times used for local web apps (V8/Blink and JSCore/Webkit) are now fast enough for well written code to not notice any difference for many use cases. I think soon most SaaS apps without much of a deep tech moat (not easy to replicate) will be labelled as rip-off or anti-patterns.
This really matters in map apps when you're on the road. Apple Maps is all very nice with transit planning but without a solid net connection it lies there mutely. Organic Maps (and others) that use downloaded OSM data work well with nothing more than GPS. You can get a transit plan (trains & trams & buses, and the transfers among them) without up-to-the-minute arrival info.
Yes, and Organic Maps and Maps.Me do the same - download entire countries or regions. It is SO tremendously useful when traveling. Have not tried CoMaps yet.
Maps.me, Organic Maps and CoMaps are all the same family.
Maps.me used to be an open source project, when they took it closed source and tried to commercialize it, the last open source version was forked and is now known as Organic Maps.
Then the Organic Maps maintainer started acting sketchy, introduced a paid partnership with Kayak and generally not behaving in a very community minded way. Organic Maps was forked and the new version is now known as CoMaps.
You're not wrong, but this feels like an equivalent to range-anxiety in electric cars.
The area you can download is still pretty large - whole cities, big chunks of countryside. I'd put money on the fact its good enough for most people's uses.
> I'd put money on the fact its good enough for most people's uses.
I don't know anyone (and I mean exactly zero) who ever used the offline feature of Google Maps. Online is good enough for people who almost always have an Internet connection. What happens when they don't is that they don't have access to the map.
People who care about having offline access (because they live in a place where they don't always have Internet or just because they care) don't bother using the Google offline feature. They use proper offline maps instead. Maybe in parallel to Google Maps: it's possible to have Google Maps and CoMaps.
I've used it. What happens is the app mostly keeps working as normal during connection dropouts (in between cell towers while travelling) instead of getting stuck loading or displaying an error. It's the exact same experience, not a separate offline experience.
I really loved the time when I could install shareware on my PC and I was sure that it would just keep working because nothing would update it behind my back; and that I didn't have to pay a monthly subscription fee to keep using it; and of course also that it would not send my data back to the software vendor.
The internet is great for many things, but also has huge downsides.
This is true for me, but I wonder if it’s true for most. An alternative framing is not that apps moved to the cloud because they were faster or more convenient but that apps moved to the cloud because they became “free” by you paying with your attention or data. It’s worth reframing the hypothesis to “The future is (paid) local app” and see if you still think it holds.
I don't mind paying a few bucks for value. But the market for lemons applies. Mostly when you spend money you just get the exact same shitware you can get for free, so I usually suspect that's going to happen.
The whole paying with your data thing should become a payment option, and governments should enforce the availability of multiple payment options with no unreasonable price differences across the board. (And there are ways to find out what your data is worth in dollars.)
This article looks a bit superficial to me. The cloud has definitely advantages, like having the same data on all devices and it provides some kind of backup. The mentioned downsides could be adressed by better software archticture or e2e encryption.
> The cloud has definitely advantages, like having the same data on all devices
That's not an advantage for the user at all, is it?
> and it provides some kind of backup.
You can backup user data in an offline-first app. Better even: you can let the user backup their data themselves!
> The mentioned downsides could be adressed by better software archticture or e2e encryption.
The article is superficial, but they mention at least one thing that cannot be addressed this way: what happens when you are offline or have a very slow connection?
It really boils down to one simple fact, IMHO: offline-first is harder to do. It is better for the users, but it is harder to do for the company doing it, which means it is more expensive. Companies are there for the profit, not to please the users.
> That's not an advantage for the user at all, is it
Of course it is an advantage. I tend to access most of my apps and services from both the PC and the smartphone, having data shared across the two is a deal breaker for the very vast majority of them.
> Better even: you can let the user backup their data themselves!
Whilst that's better for a small audience of techy types (most here, I'd guess), you know that's not true for most users, right?
Like people used to lose work on hard drives and laptops that they hadn't backed up all the time and its much, much rarer now (OK, I don't have data beyond anecdata but still).
I do yes. It's good to have a choice. Many times the reason there is not a proper local export is that it would take more work to implement properly, and it doesn't bring money to the company.
Again: companies exist to maximise profit. When making the product worse (or not making it better) improves their margin, they go for that. They don't care about the users, that's a big lie. A company fundamentally cares about profit. If competition or regulations mean that their profit is correlated to the users satisfaction, then they will have to work for that. But they don't do it "because they care about the users", ever.
The better software architecture is huge - if done right. It's the difference between cloud-only, local-only, and local-first. Not all apps are entirely easy as local-first, but you can often get 80-90% of what you need as that; and leave sync and collisions as rare situations to be handled manually.
slightly off-topic but i dont really know how to put it in any other way: everything under this domain feels fake
i am looking very hard for any hint of humanity behind any of it, but all i see is a vibe-coded website, marketing a huge collection of vibe-coded apps by generating articles like this one. even their social media presence (linked on their blog[1]) feels like agents that have been told to role-play on X.com as caricatures of their position/role in the (made up?) company
and i mean.... come on, a $5 web browser??? that's just a wrapper around the system's webview??? why would anyone think this is something worth spending their time building, economically speaking??? even FREE browsers are struggling in this hyper-competitive market. if you make a paid browser it better be the best browser of all time, not some vibe-coded average of every other browser
i agree with the premise, im all for local-first software. but this really feels like the AI-era equivalent of the countless scams we've seen during the crypto-era...
Well, when the author eventually gives up trying to sell people names of apps and the domain expires, I’m sure it will present good value when it gets bought and repurposed as a porn site; the first two apps under quick utilities are CreamShot and DripCleaner?!
I am using the older version of raspberry pi 3 B+ to host mail and web services locally behind firewalls.
There is a use case here for developing for remote areas that have unreliable internet and power.
Still a way to go but there is something in place.
This site is a beta running from my lounge at home.
https://localboxmail.co.za/software.html
The problem with native apps are the app stores and their requirements. For true cross-platform, you need to develop for 5 platforms (macOS, Windows, Linux, iOS, Android), each with their own technology, favored programming language and ecosystem, deployment, certificates, regulations, requirements, guidelines, arbitrary app review policies, categories, etc. I'm doing that currently and it's madness.
IMHO, it would be best if developers all over the world boycotted proprietary platforms entirely. (Of course, this is never going to happen, I'm just saying.)
> it would be best if developers all over the world boycotted proprietary platforms entirely.
Note that Linux and Android, from your list, are open source.
> For true cross-platform, you need to develop for 5 platforms
I would argue that anyway, you should develop for multiple "kinds of devices". Writing the same code for a desktop computer and a smartphone is fundamentally a mistake: I don't want to slide big buttons with my mouse.
So first, you need to design the UI multiple times: for desktop, for smartphones, and for tablets. If you want a smart watch, you need another design for it. Web apps don't solve that better.
Second, if you can write a webapp that works everywhere, then you can write a native app that works everywhere. I don't see how that would be different. The problem is generally that if you run the exact same code everywhere, then it's not a good experience.
Android is closed-source but with some open-source components, for only some of which they'll actually follow the license and give you the source code.
This is a nice article and I genuinely agree with its premises. Would be nice to reclaim a privacy back. A bit. But this article would be much stringer if it was not generated by Claude
No, the opposite, the only app which is save is the app you never ship (stays in a cloud). No binary or source code available to search for exploits by ai.
The future is you pay $1000 for a "cloud terminal" with 4 GiB of RAM and maybe 128 GiB of storage, and all of your apps live, metered, in the cloud. Arbitrary operating systems, and even arbitrary application binaries, are not allowed, in order to comply with laws and regulations, as well as non-legally-binding policies set by credit card companies in response to pressure groups, on what you may compute and what you may communicate. The government can intercept any communications or interactions you have with online services at any time, and any time you touch a computing device it's traceable back to your real name.
It's nice to build decentralized personal software and think you're fighting the good fight, but all the government has to do is pass a law and suddenly that option is, practically and effectively, closed to you. We're headed down that path, and that's before you get to the fact computers powerful enough to support today's more interesting applications are just being priced out of the consumer market altogether.
So why can i still run all my decentralised apps? Where's this magical shutdown the government is going to do?
Are they going saturate the airwaves to destroy mesh networks or hunt every mesh node on the planet?
There was a time when local apps written in binary compiled languages were notably faster to run, but notoriously difficult to write and scale across platforms. Both of these have improved but the JS/CSS run times used for local web apps (V8/Blink and JSCore/Webkit) are now fast enough for well written code to not notice any difference for many use cases. I think soon most SaaS apps without much of a deep tech moat (not easy to replicate) will be labelled as rip-off or anti-patterns.
This really matters in map apps when you're on the road. Apple Maps is all very nice with transit planning but without a solid net connection it lies there mutely. Organic Maps (and others) that use downloaded OSM data work well with nothing more than GPS. You can get a transit plan (trains & trams & buses, and the transfers among them) without up-to-the-minute arrival info.
Google and Apple maps have an offline mode. Just select the area you want downloaded and that's that.
Have you ever tried Organic Maps or CoMaps?
Last time I tried with Google I could only download a relatively small area, and it would only be available temporarily.
With CoMaps, I can download entire countries and keep them forever. That's pretty different.
Yes, and Organic Maps and Maps.Me do the same - download entire countries or regions. It is SO tremendously useful when traveling. Have not tried CoMaps yet.
Maps.me, Organic Maps and CoMaps are all the same family.
Maps.me used to be an open source project, when they took it closed source and tried to commercialize it, the last open source version was forked and is now known as Organic Maps.
Then the Organic Maps maintainer started acting sketchy, introduced a paid partnership with Kayak and generally not behaving in a very community minded way. Organic Maps was forked and the new version is now known as CoMaps.
We'll see how long this one lasts.
Yep, that's the beauty of open source: "what do we do if we end up disagreeing with the maintainers?" -> just fork away :-).
You're not wrong, but this feels like an equivalent to range-anxiety in electric cars.
The area you can download is still pretty large - whole cities, big chunks of countryside. I'd put money on the fact its good enough for most people's uses.
> I'd put money on the fact its good enough for most people's uses.
I don't know anyone (and I mean exactly zero) who ever used the offline feature of Google Maps. Online is good enough for people who almost always have an Internet connection. What happens when they don't is that they don't have access to the map.
People who care about having offline access (because they live in a place where they don't always have Internet or just because they care) don't bother using the Google offline feature. They use proper offline maps instead. Maybe in parallel to Google Maps: it's possible to have Google Maps and CoMaps.
I've used it. What happens is the app mostly keeps working as normal during connection dropouts (in between cell towers while travelling) instead of getting stuck loading or displaying an error. It's the exact same experience, not a separate offline experience.
Google and apple maps work the same with offline mode.
I think some people who grew up with web apps didn't know local native apps preceded them.
I really loved the time when I could install shareware on my PC and I was sure that it would just keep working because nothing would update it behind my back; and that I didn't have to pay a monthly subscription fee to keep using it; and of course also that it would not send my data back to the software vendor.
The internet is great for many things, but also has huge downsides.
Well, thanks to evergreen and auto updating systems, there's no guarantee that X app will be the same today and tomorrow.
It's probably what I hate most about Android- apps shouldn't need an update as often as they do.
It is really like some vendor breaking into your house and changing your furniture because "it looks better now". Should be illegal.
This is true for me, but I wonder if it’s true for most. An alternative framing is not that apps moved to the cloud because they were faster or more convenient but that apps moved to the cloud because they became “free” by you paying with your attention or data. It’s worth reframing the hypothesis to “The future is (paid) local app” and see if you still think it holds.
I don't mind paying a few bucks for value. But the market for lemons applies. Mostly when you spend money you just get the exact same shitware you can get for free, so I usually suspect that's going to happen.
The whole paying with your data thing should become a payment option, and governments should enforce the availability of multiple payment options with no unreasonable price differences across the board. (And there are ways to find out what your data is worth in dollars.)
Meta has the option if I'm not mistaken, but my guess is that it is very, very rarely used.
Because it's a decoy option. They really want your data, not your money, so they set the price high.
This article looks a bit superficial to me. The cloud has definitely advantages, like having the same data on all devices and it provides some kind of backup. The mentioned downsides could be adressed by better software archticture or e2e encryption.
Obsidian is just one example of how powerful local apps can be with a generic sync behind them (icloud sync, gdrive, obsidian sync, etc)
> This article looks a bit superficial to me.
Yes it is, I totally agree.
> The cloud has definitely advantages, like having the same data on all devices
That's not an advantage for the user at all, is it?
> and it provides some kind of backup.
You can backup user data in an offline-first app. Better even: you can let the user backup their data themselves!
> The mentioned downsides could be adressed by better software archticture or e2e encryption.
The article is superficial, but they mention at least one thing that cannot be addressed this way: what happens when you are offline or have a very slow connection?
It really boils down to one simple fact, IMHO: offline-first is harder to do. It is better for the users, but it is harder to do for the company doing it, which means it is more expensive. Companies are there for the profit, not to please the users.
> That's not an advantage for the user at all, is it
Of course it is an advantage. I tend to access most of my apps and services from both the PC and the smartphone, having data shared across the two is a deal breaker for the very vast majority of them.
oh sorry, I understood "running the exact same code everywhere".
Sure, sync between devices is a nice feature. But it can totally be implemented with offline-first apps.
> Better even: you can let the user backup their data themselves!
Whilst that's better for a small audience of techy types (most here, I'd guess), you know that's not true for most users, right?
Like people used to lose work on hard drives and laptops that they hadn't backed up all the time and its much, much rarer now (OK, I don't have data beyond anecdata but still).
> you know that's not true for most users, right?
I do yes. It's good to have a choice. Many times the reason there is not a proper local export is that it would take more work to implement properly, and it doesn't bring money to the company.
Again: companies exist to maximise profit. When making the product worse (or not making it better) improves their margin, they go for that. They don't care about the users, that's a big lie. A company fundamentally cares about profit. If competition or regulations mean that their profit is correlated to the users satisfaction, then they will have to work for that. But they don't do it "because they care about the users", ever.
The better software architecture is huge - if done right. It's the difference between cloud-only, local-only, and local-first. Not all apps are entirely easy as local-first, but you can often get 80-90% of what you need as that; and leave sync and collisions as rare situations to be handled manually.
slightly off-topic but i dont really know how to put it in any other way: everything under this domain feels fake
i am looking very hard for any hint of humanity behind any of it, but all i see is a vibe-coded website, marketing a huge collection of vibe-coded apps by generating articles like this one. even their social media presence (linked on their blog[1]) feels like agents that have been told to role-play on X.com as caricatures of their position/role in the (made up?) company
and i mean.... come on, a $5 web browser??? that's just a wrapper around the system's webview??? why would anyone think this is something worth spending their time building, economically speaking??? even FREE browsers are struggling in this hyper-competitive market. if you make a paid browser it better be the best browser of all time, not some vibe-coded average of every other browser
i agree with the premise, im all for local-first software. but this really feels like the AI-era equivalent of the countless scams we've seen during the crypto-era...
[1]: https://getapps.cafe/blog/a-new-name-behind-the-counter-welc...
Yeah, I read the subheader, saw Claude, and closed the tab.
Well, when the author eventually gives up trying to sell people names of apps and the domain expires, I’m sure it will present good value when it gets bought and repurposed as a porn site; the first two apps under quick utilities are CreamShot and DripCleaner?!
The domain was registered on 15 May 2026 so these are probably vibe coded apps, thats why they are sold so cheap. The naming is probably AI too.
No proof yet that they will be maintained but they seem to be pretty simple. Single use software, sold as a bundle.
I am using the older version of raspberry pi 3 B+ to host mail and web services locally behind firewalls. There is a use case here for developing for remote areas that have unreliable internet and power. Still a way to go but there is something in place. This site is a beta running from my lounge at home. https://localboxmail.co.za/software.html
I always wanted to do local and offline first app with optional sync, but the difference in amount of work is just stagering.
Now with LLM to do the boiler plate of event logs, healing, retries and testing all that, it's finally reasonable for a side projet.
And the result is great, the UI is reactive, it works in malls with bad signal, you get a no account experience almost for free...
I do all my apps like that now.
The problem with native apps are the app stores and their requirements. For true cross-platform, you need to develop for 5 platforms (macOS, Windows, Linux, iOS, Android), each with their own technology, favored programming language and ecosystem, deployment, certificates, regulations, requirements, guidelines, arbitrary app review policies, categories, etc. I'm doing that currently and it's madness.
IMHO, it would be best if developers all over the world boycotted proprietary platforms entirely. (Of course, this is never going to happen, I'm just saying.)
> it would be best if developers all over the world boycotted proprietary platforms entirely.
Note that Linux and Android, from your list, are open source.
> For true cross-platform, you need to develop for 5 platforms
I would argue that anyway, you should develop for multiple "kinds of devices". Writing the same code for a desktop computer and a smartphone is fundamentally a mistake: I don't want to slide big buttons with my mouse.
So first, you need to design the UI multiple times: for desktop, for smartphones, and for tablets. If you want a smart watch, you need another design for it. Web apps don't solve that better.
Second, if you can write a webapp that works everywhere, then you can write a native app that works everywhere. I don't see how that would be different. The problem is generally that if you run the exact same code everywhere, then it's not a good experience.
Android is closed-source but with some open-source components, for only some of which they'll actually follow the license and give you the source code.
Yeah for my personal stuff I prefer PWAs. Just harder to get good offline mode working on those.
This is a nice article and I genuinely agree with its premises. Would be nice to reclaim a privacy back. A bit. But this article would be much stringer if it was not generated by Claude
It’s a garbage article with no real content aside from being a superficial advertisement
No, the opposite, the only app which is save is the app you never ship (stays in a cloud). No binary or source code available to search for exploits by ai.
> No binary or source code available to search for exploits by ai.
That's an advantage for the company, not for the user.
The best user experience is when the client app is open source, so that I can adapt it to my needs.
There is not much choice when even your phone OS isn't a local app.
That makes no sense. How would the OS be an app? By definition, the OS is not an app.
The future is you pay $1000 for a "cloud terminal" with 4 GiB of RAM and maybe 128 GiB of storage, and all of your apps live, metered, in the cloud. Arbitrary operating systems, and even arbitrary application binaries, are not allowed, in order to comply with laws and regulations, as well as non-legally-binding policies set by credit card companies in response to pressure groups, on what you may compute and what you may communicate. The government can intercept any communications or interactions you have with online services at any time, and any time you touch a computing device it's traceable back to your real name.
It's nice to build decentralized personal software and think you're fighting the good fight, but all the government has to do is pass a law and suddenly that option is, practically and effectively, closed to you. We're headed down that path, and that's before you get to the fact computers powerful enough to support today's more interesting applications are just being priced out of the consumer market altogether.
So why can i still run all my decentralised apps? Where's this magical shutdown the government is going to do? Are they going saturate the airwaves to destroy mesh networks or hunt every mesh node on the planet?
Go home with your nihilism.
Simply making something illegal will reduce it by 99%.
Didn't we say a thousand of times that local apps are not "winning" because they can't be monetized (a bit like web games)?
Yes except the hardware you running these apps on is getting astronomically higher