[Bartweiss]

35 U.S.C. § 102 sets the standards for patentability. Interestingly, the relevant standards are simply statements about the condition of the world:

unless— (1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention

Patent applicants are expected to conduct prior art reviews and attempt to determine novelty, but this is done defensively - for the purposes of not wasting money on an obvious failure, or to head off anticipated claims of prior art by justifying the novel feature.

Meanwhile the patent office is expected to search for prior art and issue a rejection if found. To aid in this, there are also mechanisms for interested outsiders to contact the USPTO with evidence for rejecting a patent. Finally, someone who finds themselves infringing on a patent can file for review of the patent - this is the intended mechanism to ensure that people don't have to constantly watch for all patent applications which might harm them.

In practice, the USPTO is fairly good at rejecting things which have already been patented, or are obvious to a layman. But they often miss domain-specific developments which are either obvious to professionals (many software patents) or entered the market without being patented (apparently, this mid-door design). The ex parte re-examination system seems to have been gaining some teeth lately, but isn't sufficiently reliable to prevent patent trolling.

[monochromatic]

> Patent applicants are expected to conduct prior art reviews and attempt to determine novelty, but this is done defensively - for the purposes of not wasting money on an obvious failure, or to head off anticipated claims of prior art by justifying the novel feature.

If that’s true, patent applicants fail to meet expectations 90+% of the time.