Why the future is local app
getapps.cafe
getapps.cafe
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.
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.
I'd see if an LLM can help you get the data, build you a migration tool, and help you migrate it.
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 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.
Have lived in Alaska. Used offline Google maps. Worked perfectly for years and years.
I've used it plenty for things like driving trip to remote places with poor signal (e.g. where you'll want to be able to get directions back to civilisation) and foreign cities (when you're on a phone contract with bad overseas data prices).
Worse, without a net connection you can’t even get the phone to start a route.
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.
Go home with your nihilism.
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.
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.)
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.
GrapheneOS is based on Android precisely because Android is open source. They couldn't do that with iOS, or Windows, or any proprietary OS. Same for LineageOS.
Honestly it seems like you don't really know what you are talking about here?
Also you can totally fork iOS, in binary form, on the dark web, and run it somewhere that isn't an iPhone.
What runs in your phone is AOSP, with some firmware blobs (many of which are proprietary), with some system apps that may be proprietary (like Play Services and Play Store or whatever the OEM adds). If you need to convince yourself, try to build AOSP and run it.
The OP is just trying to derail the discussion. Would you say that Linux is not open source because the Linux you run on your computer most likely has to run some proprietary blobs? Or that Linux is not an OS because Linux is a kernel? What would it bring to the question of "are there open source OSes out there?", really?
> it would be best if developers all over the world boycotted proprietary platforms entirely.
I noted that Linux and Android are open source.
To which I got the answer:
> 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.
Now please explain me how "they only follow the licence and give you the source code for some open-source components of Android" is obvious, and how I choose to be wrong when I say that the full goddamn OS is open source, not because Google follows the licence (because they don't have to, they own the codebase) but because Google chooses to publish it. There are many reasons for that and I hate Google as much as the next person, but the fact remains that the Android OS is open source.
So back to the original point where "developers should boycott proprietary platforms", I don't think that Linux and Android should be boycotted. Developers should use and support open source Android systems like GrapheneOS and LineageOS (which they generally have no idea about) instead of boycotting them.
The best open source mobile OS you can get nowadays is based on AOSP, where the A means "Android". Sure, there are people who make proprietary products out of that. But it doesn't mean that Android is not an open source OS. The OS part, again, is open source. Even the Android SDK is open source.
It's closer to the difference between "Linux" and "GNU/Linux", which is often used by trolls like you to derail discussions. And more often than not, those who derail the discussions by saying "GNU/Linux" don't actually understand that there are mainstream Linux distributions that don't run a GNU userland, and that "GNU/Linux" is actually incorrect for whatever they are trying to say.
I also know that natural languages are ambiguous, and that when people say "I use Linux" they mean "I use a Linux distribution". When it gets ambiguous, I know to ask "are you talking about Linux the kernel or Linux the family of OSes?", but I don't remember it happening in a constructive discussion.
My experience with people derailing the discussion to "I cannot understand when you talk about Linux as an OS, because Linux is a kernel" is that they actually do understand, but they somehow choose to derail the discussion. Interestingly, those who do that often believe that "GNU/Linux" is the right way to talk about Linux-based OSes, which I find amusing.
Many people genuinely don't understand that the Android OS (AOSP) is open source, but that the Stock Android image they receive with their phone is not. And therefore they could genuinely wonder why developers don't revolt and choose to support "an open source mobile OS". But the truth is that the best open source mobile OS is AOSP-based, it is part of the "Android" family when we use natural languages constructively.
And for some reason, there are people like you who seem to love talking endlessly about whether or not is it correct to say "Android" for "Android-based" or "Android Open Source Project", and probably also love to annoy people saying "GNU/Linux" even if the OS in question is actually not using the GNU userland.
Android is an open source OS: you can take the sources, compile them, run them on your phone and you will have a fully functional OS.
Because it is open source does not mean that nobody can make a proprietary product out of it.
The internet is great for many things, but also has huge downsides.
It's probably what I hate most about Android- apps shouldn't need an update as often as they do.
"Resolve login issues by improving forgotten password messaging"
Hold up a whole queue in a pharmacy to change some fucking text lol.
No proof yet that they will be maintained but they seem to be pretty simple. Single use software, sold as a bundle.
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...
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.
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.
Sure, sync between devices is a nice feature. But it can totally be implemented with offline-first apps.
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.
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.