It's a standard platform, but the platform itself is not standardized, which is the point I think GP was making. The problem isn't that people don't think the web is a standard platform, but that it's hard to quantify what it means to support the web as a platform because of conflicting implementations. Standardization is ostensibly supposed to help with that, but it obviously doesn't perfectly reflect what browsers will actually support. Ultimately it the question comes down to whether people can actually use the browser with the websites they want to connect to, but that's not something that can be quantified easily due to it depending on what features websites are using and which features a browser has implemented, and network effects can make it hard to pin down what's expected at any given time, which is where all of these "imperfect" metrics like calculating some score based on the standards is coming from.