Wuala

» see wuala.com as it lived, on the wayback machine

~ the obituary ~

wuala's peak years overlap with my time on dropbox, but the swiss encrypted-cloud-storage market never reached me in any practical way. i knew the eth zurich origin from reading about it. i did not move my files. for engineers in turkey doing anything privacy-relevant, the default at the time was either dropbox with a self-managed cryptomator container layered on top, or a vpn-fronted nextcloud on a hetzner box. wuala sat in a niche between those two postures.

the product had an interesting bone structure. dominik grolimund and luzius meisser had built it as a research project at eth zurich, then released it publicly on august 14, 2008. the early architecture was peer-to-peer: you donated local disk space and the system shredded your files into encrypted chunks that lived distributed across other users' machines. that part of the design got dropped before the lacie era. the later wuala was a more conventional encrypted client to a centralized server farm, with the headline feature still being that the encryption keys stayed on your device.

i remember reading about it on hacker news and thinking the model was clean enough to recommend to anyone asking. nobody i knew asked. lacie acquired the company in 2009, then seagate acquired lacie in 2012, which is how a swiss research spinout ended up inside a delaware-incorporated hard-drive maker. for a couple of years the brand kept its identity inside lacie's cloud lineup.

~ the case file ~

~ what it was ~

the client mounted as a folder. you dragged files in. they were encrypted on your machine before any bytes left it. the keys never went to the server. that pitch is exactly what tresorit and proton drive sell today. wuala was selling it in 2008, before the snowden disclosures made the threat model legible to a wider audience. the company also tried groups, shared folders, a public-link layer; the encryption was the differentiator.

the european legal angle mattered too. swiss data residency was the implicit pitch for any business customer worried about the patriot act reach into us-hosted cloud storage. that argument carried weight for a small slice of european enterprise buyers and almost nobody else.

~ the day it died ~

on august 17, 2015 the wuala blog posted that the service was ending. the post named tresorit as a recommended successor. uploads were turned off september 30; the service ran read-only for about six weeks while customers exported their data. on november 15, 2015 the storage backend was wiped. seagate did not explain in detail. the broader read was that consumer cloud storage was not a strategic fit for a hardware company whose margin lived in spinning disks and ssds.

~ the next chapter ~

tresorit picked up the slice of the user base that cared most about client-side encryption. proton drive arrived later from the same swiss-jurisdiction angle. the architectural idea wuala demonstrated, that you can run a usable consumer cloud where the server never sees plaintext, has held up. the broader market still mostly runs on dropbox, google drive, icloud, onedrive, where the provider holds the keys for most users. wuala is one of the references the encrypted-by-default vendors still cite when they explain why their model is worth the friction.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »