🎓 Scholar Proxy

Profile photos, co-author photos and icons are refused before they download, on the pages whose page check below has Skip images ticked (all three built-in ones). The parsers read only the HTML, so nothing they get changes; the browser window on the proxy PC shows those pages without pictures.

Expect a small saving. Nearly all of what a Scholar page costs is its HTML, styles included. The icons and placeholders every page shares come from the cache after the first page, so what the saver really trims is each scholar's own photo, and only once per scholar. Compare the average a page below with the switch on and off.

A CAPTCHA, Google's sign-in and any page no rule covers always load in full, and nothing is skipped while the proxy is stopped on a page, so a challenge is never missing its pictures.

Stylesheets and scripts are left alone. The browser keeps them in its cache (the table shows how many come from it), so after the first page they cost next to nothing, and a page that never runs its scripts looks more like a bot to Google.

Before a page is handed over, the first rule below whose pattern matches the requested URL says what a real answer contains: the element the parser reads. A page without it — Google's sign-in, a consent screen, an error page, a Scholar redesign — stops the proxy exactly like a CAPTCHA: the parser pauses, no data is touched, and you are alerted. Nobody has to have seen that page before. Requests no rule matches are only checked for leaving Scholar. A page with an error status (a 404 for a removed profile) is never stopped; the parsers handle those. CAPTCHAs and Google's sign-in are caught even with this switched off.

If the proxy stops on a page that is actually fine, fix its rule here (or switch the rule off) and that same request goes through: within a few seconds the proxy re-checks the page, finds it passes, and lets the waiting parser carry on. Let it through on the Home page is for a one-off page.