[bandie91]

> The main header JS starts using range requests to first load the real HTML, and then it watches requests for resources; the resources have been rewritten to be deliberately broken 404 errors (requesting from localhost, to avoid polluting any server logs)

what if a web server on localhost happens to handle the request? why not request from a guaranteed unaccessable place like http://0.0.0.0/ or http://localhost:0/ (port zero)

[gwern]

If those were guaranteed inaccessible, wouldn't a web browser be within its rights to optimize those away?

[bandie91]

either it optimizes and dont try to connect or does not recognize as never accessible and does try to connect – both are better than accitentally fetching something from a web service running on localhost.