So what's wrong with 1975 programming? (2008) | Hacker News Reader

So what's wrong with 1975 programming? (2008)

varnish-cache.org

37 pointsbaotiao11 comments

An HN evergreen -

https://hn.algolia.com/?query=1975%20programming&sort=byPopu...

HN does need a gold or classic posts tab so we automatically note this reposting. Appearing again and again hints to the fact the article is ^more interesting^ than usual.

Yes and there's a probably a whole other interesting post to be written about the critiques that have come up over the years, including some that claim the original is no longer representative of how varnish works today.

Stackoverflow has "community wiki" questions, maybe HN could have community wiki comments?

OTOH my limited experience suggests that the majority of front-page articles will get reposted several times.

Or someone could publish a "Best of HN" collection, then change all reposts to link to that.

"Best of HN" collection"

that @Mathnerd is a good idea. It would not be hard to do this remotely via a script.

Yeah. It's a little tricky though, because the top posts (https://hn.algolia.com/?query=&sort=byPopularity&prefix&page...) are all one-time events, and the # of people on HN has been increasing so upvotes aren't comparable across time. The criteria should be something like "reached the front page multiple times over several years".

what is the hn.angolia search term for the "past" of a story?

i.e.: for "So what's wrong with 1975 programming?" it is https://hn.algolia.com/?query=So%20what's%20wrong%20with%201...

Having this means you can measure how popular they are.

Most importantly, posts with that label should have links to all previous discussions

Was it all fixed in '76?

yes.

So, I get the frustration expressed here, but PH-K just comes off sounding like an arrogant prick by the end of this. I know I'm supposed to attack the argument and not the author, but it's very hard to read a piece like this when the author makes you feel like he's talking down to you the entire time.

I assume this was posted because of the presence of antirez's 4-years-later rebuttal (well, for some workloads) that's also on the front page today (https://news.ycombinator.com/item?id=13226341). It's interesting to see the difference in writing style: antirez seems to just aim to educate, and reasonably points out how redis's needs and memory access patterns don't allow for good performance if it were to just rely on the OS's VM system. The article about Varnish, however, comes off as someone telling us how stupid we are for not knowing how hardware actually works.

Reply on news.ycombinator.com