Rohit Kumar @trycloudless

Client-side encryption means your files are encrypted on your device before upload. The cloud only stores unreadable data, not your original files. For private backups, that matters: recovery without giving up privacy. youtu.be/HdRKxtsUMy4

@BleepinComputer Incidents like this are a reminder that endpoint compromise eventually becomes a data recovery problem too. Detection matters, but so does having clean, versioned backups that were not silently overwritten by the compromised machine.

@mddanishyusuf Simple pricing with clear limits is underrated. For indie tools, I think users mostly want to know: what do I get, what happens when I grow, and will the deal still make sense later.

@TomZarebczan Shipping the toggle based on actual feedback is the right kind of product iteration. I like changes like this because they reduce user uncertainty without needing a big redesign. Small controls often matter more than they look from the outside.

@TheHackersNews This is why "where is the data stored?" is only one privacy question. The other question is who can read it once it is there. For sensitive backups, client-side encryption before upload changes the blast radius if a storage account or workflow is compromised.

@TheHackersNews File-altering malware is a good reminder that prevention is only half the plan. You also need recovery from a known-good point: versioned backups, offline/offsite copies where possible, and restore tests that prove clean files can come back.

@backblaze "Convenience can become constraint" applies to backup too. Bundled platforms are easier to start with, but modular setups give teams more control over storage cost, retention, encryption, and recovery workflows. The hard part is keeping the UX simple.

@ProtonDrive Speed matters a lot for cloud storage UX. For important files, I still think users should separate storage/sync from backup: fast access is one job, versioned recovery is another. Both matter, but they protect against different problems.

@felixycc Future-you being part of the engineering team is a good way to put it. I feel this especially with reliability code. The clever version is tempting, but the boring version is the one you can debug months later when something fails.

@felixycc This is very true for backup software too. If a backup is running but the user cannot tell what is happening, whether files are protected, or how restore will work, it feels broken even if the engine is technically doing the right thing.

@FilecoinTLDR "Restore confidence gap" is a useful phrase. Storage durability is only one part of the problem. The real test is whether you can reconstruct the exact expected files later, verify them, and make that process understandable when something has already gone wrong.

@myworkdrive Agree. I would add one more layer to that thinking: collaboration tools are for collaboration, access tools are for access, and backup tools are for recovery. Trying to make one sync layer do all three is usually where the painful edge cases show up.

@myworkdrive This is the distinction people miss: sync, file access, collaboration, and backup are different jobs. OneDrive/Google Drive-style tools are useful, but once a bad delete, conflict, or ransomware-encrypted file syncs everywhere, you need a separate versioned recovery path.

Sync is useful, but it is not a backup. Google Drive, Dropbox, iCloud Drive, OneDrive, etc. keep files available across devices. But if you delete or corrupt a file, that bad change can sync everywhere too. Backup needs versions and restore. Short video: youtube.com/watch?v=QPF4sY…

Sync is for access. Backup is for recovery. The difference matters because sync can copy mistakes everywhere: deletes, corrupted files, overwritten files. A real backup needs versions, deleted-file retention, and a restore path you have tested. #backup #cloudstorage

Nice. One thing I would be careful about is whether it is sync or backup. If the local profile gets corrupted and the script uploads that state, Google Drive may only preserve the bad version unless you have retention/versioning covered. I am building CloudLess around this backup-not-sync problem: local encryption, versioned file backup, and your own storage backend. Beta for Mac/Linux/Windows now. Disclosure: I am the developer.

@cyber_razz Seeing backup files is inventory, not validation. The only backup status I really trust is a restore that was tested somewhere else. Even a small scheduled sample restore catches a whole class of problems that “ls /backups” will never show.

@felixycc Tiny bug, huge lesson. I keep coming back to this while building backup flows: “backup completed” is not the finish line. The product has to make restore verification boring and visible, otherwise users only discover the truth during the worst possible moment.