Human error? I thought these search engines were crawlers front-ended by algos? What am I missing here?
@Johncdvorak Microsoft has a cloud service (not Github) on Azure that companies use to store IP. One of our engineers mindlessly migrated a bunch of code repositories there. I had them yanked, but probably too late. Until I can trust that Msft discriminates against hiring anyone with a possible tie to the CCP, I'll never allow my stuff up there again. That Tiananmen/Bing story just validated my paranoia.
An IF statement.
Or perhaps every image tagged as "tank man" was removed as a possible result. Though the latter might get rid of pictures like this one
They cook the data to feed the results in a variety of ways.
@Johncdvorak The human in charge of putting in place the rule to censor forgot to put the condition "if ip from china"? Regardless, clear admission of censorship and how once you censor in one constituency you are bound to censor everywhere
@Johncdvorak Theory: Microsoft people build it into the algo in the short term to appease the CCP's political ends in Hong Kong, while the phrase 'human error' is a PR stunt, meant to imply personal (not systemic) accountability in order to appease the critics who noticed?
@Johncdvorak Google/Bing/Yahoo can manually remove search results. They've done it for many years.
For example, if a website were purchasing inbound links to game the algo to achieve higher rankings in the search results (which is against the rules), the search engine can manually remove or demote the website from the index.
This is truly an ignorant question but would these algos have sets of initial parameters or conditions that humans might "accidentally" fuck up? Not sure in these cases (e.g. tech world cases).
@Johncdvorak I'm not sure how the blind iterations of calculations work in terms of initially structure, more what I mean.
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!