Why Write This Down
I hold one granted patent: TR 2025 014583 B, granted in 2025, for an energy-loss monitoring and intervention system for public buildings. That is the extent of what I will say about my own file here. What I want to set out instead is the shape of the process itself, because almost everything I read before starting was either a law-firm advertisement or a forum post from someone who had not finished.
The process is not mysterious. It is slow, procedural and unusually strict about language. Understanding that early changes how you work on the invention itself.
Prior Art Comes Before Everything
The first real work is not building. It is searching.
A patent is granted for something new and inventive relative to everything already public anywhere in the world, in any language, in any form. That body of material is called the prior art, and it includes earlier patents and applications, academic papers, product manuals, conference posters and public demonstrations. It also includes your own disclosures. Publishing or demonstrating your idea before filing can destroy its novelty.
Searching prior art properly is a research skill in its own right. You are not looking for something with the same name as your invention; you are looking for anything that already does what your invention does, however it is described. The vocabulary is the hard part. The same mechanism appears in different documents under completely different terms, so you search by function, by classification code, by the problem being solved, and you keep widening until you stop finding surprises.
Most ideas do not survive this stage intact. That is the point. Finding a close document early is cheap; finding it after filing is not. In practice the search does not kill the invention so much as sharpen it — you learn what is genuinely yours and what is background.
Claims, Not Adjectives
The part that surprises most first-time applicants is that a patent is not an essay about how good an idea is. The legally operative part is the claims: a numbered set of sentences that define, in precise structural terms, exactly what is protected.
Claims do not contain “efficient”, “innovative” or “smart”. Those words have no boundary, and a boundary is the only thing a claim exists to draw. A claim names elements and the relationships between them. An independent claim states the invention in its broadest defensible form. Dependent claims add narrower features that fall back if the broad claim is challenged.
The rest of the application — the description and drawings — exists to support the claims. The description must disclose the invention completely enough that a person skilled in the field could carry it out without inventing anything further. This is the trade at the centre of the whole system: you disclose how it works, and in exchange you get a period of exclusivity.
Learning to write in claim language changed how I think about engineering generally. It forces you to say what a thing is in terms of parts and connections, not what you hope it achieves. Any sentence you cannot phrase that way is usually a sentence you have not yet understood.
Filing, Examination and the Wait
In Türkiye, applications are filed with Türk Patent ve Marka Kurumu, the national patent and trademark office, under the industrial property legislation currently in force. The route is broadly the same one used by national offices elsewhere: the application is filed with a description, claims, drawings and an abstract; a search report is prepared against the prior art; the application is published; and substantive examination follows.
Examination is a written conversation. The office issues opinions on novelty, inventive step and industrial applicability, and the applicant responds — usually by narrowing claims, clarifying language or arguing why a cited document does not teach what the examiner reads into it. Several rounds are normal. Nothing about the pace is negotiable, and the correspondence is the file: what you write becomes part of the permanent record and can be used later to interpret the scope of your own claims.
If the requirements are met, the patent is granted and published. If they are not, the application is refused or withdrawn, and the disclosure is already public either way.
Twenty Years, and What That Means
A granted patent runs for a maximum of twenty years from the filing date, not from the grant date. It is not automatic: annual fees keep it alive, and missing them lets it lapse. At the end of the term the invention enters the public domain permanently.
It is also worth being clear about what a patent is not. It is a right to stop others from using the claimed invention in the territory where it was granted. It is not a product, not a market, not proof that anyone wants the thing, and not protection outside that territory. National rights are national; international coverage means further applications, further costs and further deadlines.
Knowing all of that before you start is the difference between treating a patent as a milestone and treating it as an instrument.
Note: this entry is a structural placeholder. The competition record is accurate; the technical write-up still needs the owner’s own detail.