Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'd say apart from (1) he has it down for the basics; see http://www.seomoz.org/article/search-ranking-factors for some confirmations.

Why not (1). Well yes (1) too really but: Google use a "game" to get people to describe images and if there's text it can be OCR'ed. Google are quite good at OCR. I doubt they pass all headline images through OCR but I can imagine they might do subject comparison between text and replacement images (eg in pages using Fahrner Image Replacement) to sample whether the text Google is reading matches the images the user sees.

With (2) avoid stuffing.

(4) & (5) <- this is all it takes!



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

Search: