A CTO at a small Zurich sensor company has just finished a prototype. The notebook pages are full of sketches, the board is still on the bench, and the drawings are clean enough to share. The company will not file a patent, but the CTO fears a larger rival could patent the same idea next year. One option is defensive publication: describe the idea in public so it becomes prior art. This article explains where a defensive publication prior art timestamp fits, and what it does not do.
The short answer
Under patent law, an invention must be new, and anything the public could already access before the filing date counts against it. If you publish a clear description of your idea and can later show the date, your own publication can be used as evidence that the idea was not new when someone else filed. A timestamp on the exact published file helps with the "can show the date" half. It does not decide whether your description is detailed enough to count, and it does not stop anyone from filing. None of this is legal advice.
What the rule says
At the European Patent Office, Article 54 of the European Patent Convention says an invention is novel if it does not form part of the state of the art. The state of the art is everything made available to the public by written or oral description, by use, or in any other way, before the date of filing.
Swiss law uses the same structure. Article 7 of the Patent Act (PatG) says an invention is new if it is not part of the state of the art, and the state of the art is everything made accessible to the public before the filing or priority date by written or oral description, by use, or in any other way. The Swiss Federal Institute of Intellectual Property says on its blog for startups that form and source of a publication do not matter, and that a blog or a video can be enough.
Two consequences follow. First, a public description can work against later patents by others. Second, it works against you too: once you publish, you generally lose the option to patent the same thing yourself. The Swiss act has only narrow six-month exceptions, for example for evident abuse against the applicant or for certain official exhibitions. If a patent might still be wanted, speak to a patent attorney before publishing anything.
Where the date becomes the weak point
The rule turns on a date: what was public before the other party filed. Your evidence of that date is often weaker than people assume. A blog post can be edited. A file's "modified" field is written by the computer that saved it. A page you uploaded last spring may carry no date a third party trusts.
Two things have to be shown. The content must have been publicly available by that date, and the content must be exactly what you say it was. A timestamp helps with the second point and supports the other. Reaching the public still depends on you putting the file where people can find it, at an address you record.
What a timestamp on the exact file does
Back to the bench. The CTO exports the notebook scans and the drawing set into one PDF, with a plain-language description of how the device works. That PDF is sealed. Sealing computes a fingerprint of the file, and the seal carries a trusted timestamp under RFC 3161 from a third party, so the date does not come from the CTO's own clock. The how it works page describes the fingerprint, the timestamp and the public verification link. Under Article 41(2) of the eIDAS Regulation, a qualified electronic time stamp enjoys the presumption of the accuracy of the date and time it indicates and of the integrity of the data to which it is bound.
The same PDF is then posted on a public page, and the page address is saved with the sealed original. Later, anyone can compare the document at that address with the sealed file. If the fingerprint matches, it is the same document the seal refers to. An edited copy has a different fingerprint, and the validator checks the signature and timestamp on a signed PDF.
What it does not do
The limits matter more than the mechanism, because the other side will find them.
The seal shows that this exact file existed at that time. It does not show that the public could see it at that time. That depends on where and when you posted it, so keep a record of the posting too.
It does not decide whether the description is enough. How much a publication counts against a later patent depends on what it discloses, and a vague note counts for less than a clear description with drawings. A patent attorney can judge that.
It does not show who made the drawings, and it does not stop anyone from filing. A patent office or court decides what happens to a filing. A seal also does not show the file was unchanged before it was sealed. The original sealed file stays with you, so store it in more than one place.
A sensible order of steps
First decide, with advice, that you do not want a patent. Second, write the description so a competent engineer could follow it, and include the drawings. Third, seal the final PDF and keep the sealed original. Fourth, publish that same file at a stable public address and note the date and address. Fifth, repeat for each version you change, since a later version is a different file with a different fingerprint.
Back at the bench
A year from now, the CTO may open a letter from a competitor's counsel. With the sealed PDF, the public address and the verification link, the CTO can show what was published and give a date a third party issued. Whether that defeats the claim is a question for a patent attorney and ultimately an examiner or a court. What the CTO avoids is arguing from a notebook photo and a file date that anyone could have set.
Seal the exact file you plan to publish before it goes online, and keep the sealed original next to the public address.
