Hacker News new | past | comments | ask | show | jobs | submit login

The most important single thing to read in a patent is the claims, especially the independent (stand-alone) claims.

Infringement analysis amounts to attempting to map the claim's list of elements onto the accused method or structure; if the map function returns FALSE, then that particular claim very likely isn't infringed by that particular accused method or structure.

(That is: If any claim element isn't present—possibly in the form of an "equivalent"—then the accused method or structure doesn't infringe that particular claim.)

Prior-art analysis works backwards: If a claim returns TRUE when mapped onto a piece of prior art—or onto something that would have been "obvious" at the time the invention was made when taking into account all of the relevant prior art [0]—then that claim is probably unpatentable.

In this patent, claim 1 seems to be representative. I've broken up the paragraphing and added bracketed lettering.

(Usual disclaimer: This shouldn't be relied on as a substitute for legal advice; I'm not acting as anyone's lawyer; etc.)


1. A method for positioning a window on a user's display, the method comprising:

[A] providing, in a window-based computing environment, an image [i] representing an entire display area of a user's computer screen and [ii] comprising a plurality of selectable regions,

each selectable region [x] representing a sub-area within the display area and [y] having a shape and orientation similar to the corresponding sub-area,

wherein the selectable regions are arranged in the same way the corresponding sub-areas are arranged in the display area;

[B] enabling a user to simultaneously select more than one selectable region from the plurality of selectable regions; and

[C] automatically moving a window in the window-based computing environment to a sub-area corresponding to the selected more than one selectable region.


[0] Obviousness analysis is really tricky because it necessarily requires hindsight. The test is, in essence: Given everything that was known in the relevant field(s) at the time, would a hypothetical person of "ordinary skill" (in that field or fields) have regarded the claimed subject as obvious?

My personal tl;dr for obviousness analysis is this: Suppose that a team of competent-but-not-necessarily-stellar colleagues were shown or told about the claimed invention. If their collective reaction was to raise their eyebrows and murmur, "hmm; that's interesting," then the claimed invention might well have been non-obvious. On the other hand, if their collective reaction was to shrug their shoulders and say, "um, yeah, and?" then the claimed invention might well have been obvious.

In the real world, obviousness analysis is supported by "objective evidence" of nonobviousness, if available. For example, commercial success that's shown to be due to the technical merits of the claimed invention (as opposed to being due mainly to, e.g., marketing) can weigh heavily in favor of nonobviousness. Admiration of experts, ditto.

(I once heard a story, very likely apocryphal, about a patent examiner who rejected a patent application on grounds of obviousness; the patent attorney overcame the rejection by pointing out that the claimed invention had received the Nobel Prize. Personally I'm skeptical because the timing wouldn't work, but the story illustrates the concept.)

Came here to say this, and I wish more non-lawyers understood how to read a patent. The scope of coverage is not the title. The scope of coverage is not the drawings. It's the claims.

If folks want to explain why this shouldn't have issued, identify a single document from before 2008 that describes every single feature of the claims.

It's worth noting that this issued just before a 2014 Supreme Court case (Alice v. CLS Bank) that probably can be used to kill the patent regardless of whether it's novel and non-obvious.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact