The document exists. Somewhere. In the folder system from 2019, in a mailbox, in the binder labelled Miscellaneous. paless reads every page you put in, printed or scanned, so finding something means asking for what it says, not reconstructing where a past version of you would have filed it.
scan0043.pdf, final_v2_NEW.pdf, IMG_2210.jpg. The names describe the moment of saving, not the content. Search by name and you search someone's mood in 2021.
Mailbox, desktop, cloud drive, chat downloads, the paper binder. Every search is really five searches, and the fifth one is standing up.
The folder structure made sense to whoever built it. Three colleagues and two years later, everything recent lives in a folder called Sort Later.
In paless, search does not skim file names and hope. OCR has already read every page, table and figure of every document you put in, and Pal reads on top of that, so you can ask a question in plain words and get the answer with its source attached.
Notice periods, invoice totals, serial numbers, the one sentence in the appendix: if it is printed on any page, it is findable, even in scans from the binder era.
Ask Pal what the warranty covers or when the lease ends. The answer comes back in plain language, with the documents it came from cited underneath.
paless speaks your language and the languages of your documents, so a question in English finds the answer inside a French contract.
“What is the notice period for the storage unit?”
Two months to the end of a quarter. Clause 7 of the rental agreement with Lagerbox Süd; the 2025 amendment did not change it. Next possible exit: September 30.
Sources
Full text search and cited answers from Pal are part of every plan. See what's included
Six ways search in paless beats the folder expedition.
OCR turns scans, tables, charts and even formulas into text, so paper documents search exactly like born digital ones.
Results land you on the passage that matters, not on a list of PDFs to open one by one and skim.
Narrow by correspondent, period or type: everything from one supplier, every policy from 2023, all delivery notes from March.
Pal never answers from thin air: every response cites the documents it drew from, one click from the original.
Email attachments, phone scans and synced folders all land in the same searchable place, so one search really is one search.
The moment you find it, share it: an expiring link replaces the download, rename, attach, apologize routine.
They ask about their December order. You have the delivery note and the signed acceptance on screen before they finish the sentence, and the call ends with an answer instead of a promise to call back.
found in 6 s · call: 3 min
The old cabinet went through the scanner into European servers. Now the 2009 binder answers questions it never answered as paper, encrypted and stored as three replicas plus an isolated backup.
18 years imported · all searchable
The whole decision hangs on one sentence about liability. Search lands on it directly; nobody reads forty pages, and the meeting keeps its schedule.
asked 15:04 · cited 15:04
An archive this easy to search must be hard to reach. paless runs on isolated European infrastructure with AES 256 encryption at rest and TLS 1.3 in transit: no Big Tech, no US CLOUD Act, processing under European law only. And a structured export compatible with the open source paperless-ngx project keeps the exit door open.
Yes. Every document is OCRed on arrival, including tables, figures and old scans. Once it is in paless, a 1998 fax and a fresh PDF search identically.
Search finds passages; Pal reads them and answers the actual question, citing the documents it used. Use search when you know the words, and Pal when you know the question.
Yes. paless speaks your language and those of your documents, so questions and documents do not need to match languages.
No. Full text search works from the content itself. Tags, which Pal largely suggests on its own, just add convenient filters on top.
The campaign launches on Kickstarter, with time limited Founder's Pricing for early backers.
Keep reading: All use cases Frequently asked questions