Hacker News Reader

Comment by frankjr | Hacker News Reader

I think the article and the arguments presented don't really make much sense, but I do find interesting that it says the article was published on Oct 5, 2022, yet there are comments going back as far as November 2019.

View on HN

Good catch. This is a SEO trick they probably do every year to convince google that their information is timely.

Edit: Yep. https://web.archive.org/web/20220000000000*/https://www.bicy...

Is Google's search team really this naïve? Google keeps cached versions of sites it crawls. An algorithm to catch these date changes would not be difficult to implement.

It might not necessarily be about tricking Google's algorithm. It could also be to make users stick around. People might be less likely to read an article (and potentially click another link on the site and generate another ad impression) if see that the article is "old".

Most people make cursory updates to the body of the text as well. Like my sibling comment said, updating the date also helps your search engagement by showing a more recent date. That said it’s considered best practice to show the original and updated date.

I think the entire point is that it shouldn't "help search engagement" if it's an older article that's been "updated for present year" just to game search results.

>cursory updates to the body of the text

Still pretty straightforward to algorithmically determine if this has happened.

It’s interesting that a search engine would choose to encode recency bias.

Why would that be surprising? When people search for something, they tend to want more recent info, more recent coverage, more recently made lists, etc.

For some things sure. For others, for example recipes, newer is not necessarily better.

Yep - they're not quite keeping up though. Original 2019: "the National Transportation Safety Board (NTSB) recommended <i>last week</i>"

Then at some point it was <i>last year</i> - and they've just left it like that for new next couple of years.

Only other real change is that the author used to work for "Transportation Alternatives and Families for Safe Streets" - and now doesn't.

The article seems clearly from 2020. It leads with a concern over something from "last year" that is dated 2019.

On its topic, I was expecting to mostly disagree. That said, the case is compelling. To condense, it is basically saying don't turn it into a partisan conflict between riders and police. That plays out in predictably bad ways for communities.

Reply on news.ycombinator.com