You Shouldn't Need an Account to Use Your Own Computer
Every app asks for a sign-in before it will open a file that is already on your disk. What an account is actually doing, what it costs, and how to tell which ones you can skip.
The last time I opened a plain text file on my Mac, the app asked me to sign in before it would show me the words. Not to sync the file anywhere. Not to share it with anyone. To read it. The file was already on my disk, put there by me, and the app had been installed months earlier.
I closed it and opened the file somewhere else.
That small annoyance has a shape, and once you notice the shape you notice it everywhere. Every install arrives with the same request attached. Create an account. The request lands before you have decided whether you like the tool, before you know whether you need it, and often before the app will perform a single function that has nothing to do with a server. On the getapps.cafe homepage, "Free to keep, no account needed" sits under the free apps as a selling point. That sentence says more about the state of software in 2026 than any feature list I could write.
Four jobs a sign-in can do
Strip away the marketing and a sign-in is doing one of four things.
Billing. The receipt for whoever pays.
Sync. The key to a server that holds a copy of your work so it can appear on your other machines. This is the job people assume they need, and the one most easily replaced by a folder that already lives on both.
Collaboration. How the software knows that the document in front of you is the one your teammate is also editing. Shared links, comments, permissions, who is typing right now.
Licensing and measurement. How the vendor counts you, enforces a seat limit, or records which features get used.
Those are real jobs, and a tool that does one of them well can be worth the sign-in. The trouble is that a single email-and-password field performs all four at once, and nothing on it tells you which one you are doing. You sign in to change a font and become a subscriber, a telemetry row, and a line in a chart that ends up in a pitch deck.
The three costs of the sign-in wall
The first cost is the account you cannot leave. Every service you sign into is a relationship you now have to maintain. It has a password that can leak, a reset flow that arrives at an address you stopped checking, and a billing date you will notice when it charges. Multiply that by the twelve or fifteen tools an ordinary working week actually touches and the admin becomes a job of its own.
The second cost is the failure mode. A signed-in app needs the network to confirm who you are. That works on a train until it doesn't, and it is why a licence check can stop a local file from opening on a plane, in a basement, or during an outage that has nothing to do with you. Offline capability is a design decision, and the sign-in screen is where it usually gets decided against you.
The third cost is the graph. Signing in through a large identity provider proves who you are to one app and hands a platform a record of which products you use and when. That's a business relationship you didn't agree to and can't easily undo.
The honest case for a sign-in
I don't want to argue that a sign-in is always avoidable, because it isn't.
If a tool genuinely stores your work on someone else's computer and keeps it in sync across a laptop, a phone and a tablet, a sign-in is the mechanism. If two people really are editing one document at the same time, a sign-in is what makes the shared state work. If you are paying monthly, a sign-in is the only sane way to run billing. All of that is the honest price of a service.
The question worth asking is narrower. Which of those jobs is this particular login doing for me right now, and would the app still work if I said no?
Four questions to ask any app
Run these before you hand over an email address.
Does it open a local file while signed out? Open the app with the network off. If it refuses, that login exists to protect the vendor's numbers rather than anything of yours.
What actually breaks if you sign out? Sign out and use it for a day. If you lose sync and nothing else, the login was a courier and not a gatekeeper.
Can you export everything, in a format another program reads? A .docx, a folder of .md files, a .csv, a .kdbx. If the export only works while the company is solvent, the login is holding your work rather than storing it.
Does the export need any of that? The strongest answer is a file you can copy to a drive and open on a machine that has never heard of the vendor.
What it looks like when nobody asks
The apps that skip the sign-in entirely tend to share one design choice. The file is the account. There's nothing to log into because there's nothing on a server that needs identifying.
PassBrew opens or creates a .kdbx file from the Cold Brew shelf, which is the same format any KeePass-compatible reader can open. No sign-in, no vault sitting in someone's cloud, and the database travels to another machine with a copy and paste.
SteepJournal keeps a diary in a file locked with a password only you know. The password is the entire security model. There's no reset link, which is the point and also the one thing to think about before you choose it.
NoteCafe writes Markdown notes that stay plain files on your own computer, and CalCafe keeps your events in a local database with no account and no cloud. Both were built by people who looked at the obvious architecture and picked the smaller one.
ChatCafe is the clearest test of the idea. It chats, sends files and plays games with everyone on your Wi-Fi, and it has no server to sign into at all. When the internet drops but the router still works, it keeps working, and most chat apps cannot survive that sentence.
These apps have no sign-in because one wasn't needed, and adding one would have meant inventing a server to justify it.
Connected doesn't mean signed in
There's a second category worth naming, because it trips people up. Some apps genuinely need the network and still never ask you to sign in.
TickerBar reads public market feeds and shows prices in your menu bar with no account and no API key. DemoTrading streams real crypto prices from a public exchange feed and lets you trade them with practice money, and the only file involved is the .trade record you keep yourself. FlightTrack draws live air traffic from a public data source and asks you for nothing in return. All three are on the Frappe shelf, and all three read data without an identity attached.
Reading a public feed is a different act from logging into a service, and the difference comes from how the app was built.
Where this leaves you
I don't think most people can walk away from every service they use, and I'm not suggesting a purge. What I'd like is for the request to be honest. If an app needs a sign-in to sync, say so, and let me use the local half without one. If it needs a sign-in to bill me, take the payment and leave the file alone. If it needs a sign-in because a dashboard somewhere looks better with my email in it, I would rather it didn't ship that screen.
The practical version of all this is small. Prefer the tool that opens the file without asking. When it does ask, work out which of the four jobs is being done before you type a password. And keep the important things in formats that leave with you, because a format you can copy doesn't need permission to travel.
Everything on the getapps.cafe menu works that way. Install it, sign in once to the membership if you want one, and the apps keep your files where they already are. One membership covers all 127 of them, on macOS and Windows, from $4.99 a month billed yearly or $9.99 monthly, with a 7-day free trial that adds a card when it starts and charges nothing until it ends.
If you want the longer version of this argument, it is in Why Your Mac Apps Shouldn't Need the Internet and You Can Only Back Up What You Actually Have.