Back to News
Advertisement
Advertisement

⚑ Community Insights

Discussion Sentiment

75% Positive

Analyzed from 580 words in the discussion.

Trending Topics

#llm#blog#writing#off#postgres#slop#post#discussion#topic#technical

Discussion (13 Comments)Read Original on HackerNews

weitendorfβ€’about 1 hour ago
Gotta say this site is killing it in the LLM-written technical blog SEO game the last couple years, they're consistently able to dominate (and actual deliver in spite of the LLM-isms) SERP for a set of technologies that very closely align with the ones that a typical backend/infrastructure engineer would work on.

I checked their list of blog articles from the last year and they haven't even written that many, which makes it quite impressive because they really model the set that I've been working with (Go, Postgres, grpc, otel) well by I guess defining some kind of customer archetype and writing really detailed guides about what they'd be interested in learning more about. At this point I've encountered their site "organically" like 5x in the last year and recognize the name/style of content so they're doing something right in the marketing department for sure.

On one hand you could just dismiss it as spam but articles like this actually represent a pretty significant LLM spend/human review element that delivers real value to me as a technical end user looking for info on google or in technical blogs on HN (ie it would take me a long time to generate something like this myself and I wouldn't do it proactively, only when-needed). So it actually does help me quite a bit that they do so before I think to ask about it.

krelianβ€’17 minutes ago
I first encountered them in various stackoverflow answers, this is the next step. IIRC they did promote themselves but it was subtle and the answers were good.
sysashiβ€’37 minutes ago
`on conflict do select` and eager aggregation are the most notable for me, I use upserts and joins like that all the time. Annual postgres upgrades are always exciting (features) and terrifying (ops) :D (not that something can go wrong, I just feel anxiety hehe)
nextaccounticβ€’25 minutes ago
Does Postgres 19 support full temporal tables, or what's landing is a subset of that?
exceptioneβ€’about 2 hours ago
The features look exciting, but the article is exceptionally unclear to me. I am not sure whether that is my fault, the author's, or the LLM's who wrote it.

The headline feature (property graphs) has a better example here in the official docs: https://www.postgresql.org/docs/19/ddl-property-graphs.html

zdc1β€’about 2 hours ago
Tangential, but the Claude-isms make it a less rewarding read
rvzβ€’about 2 hours ago
It is getting quite nauseating to see Claude's fingerprints all over the page and at this point I just expect the slop to continue.
andrubyβ€’about 1 hour ago
But the payoff here is that you get a sneak peak into PG19

It's off-putting to me too. I wonder if the AI enthusiasts that stopped manually writing also find it off-putting

throwatdem12311β€’about 1 hour ago
No you see when I generate a blog post it’s not slop. You see I reviewed the output and declared everything is fine. So it’s not slop.

It’s only when you generate a blog post that I get the ick from slop.

Because I clearly put effort into my blog post and you did not.

sbarreβ€’about 1 hour ago
I came to the comments to see the discussion and almost all of them are just complaining about LLM-assisted writing.

It's exhausting and tedious, like people nitpicking grammar.

It's also off-topic. I'd love it if HN banned (or strongly discouraged) this kind of repetitive low-value off-topic meta-discussion.

We get it, LLMs are used to write now. Some people don't like it. No need to broadcast that dislike every time, it adds noise to the discussion. :-(

(yes I realize I am also adding off-topic noise but this thread was nothing but that, so I guess I am on-topic until actual Postgres discussion starts).

exceptioneβ€’22 minutes ago

  > complaining about LLM-assisted writing.
I urge you to read the article. "assisted" is doing a lot of load-bearing work here. The problem is not just the writing style, but rather that it doesn't make sense at all.

Take the example for PGQ. "Two ordinary tables, one graph on top of them:" The example makes zero sense, because the fucking tables aren't even there. You can't be lazier than writing a prompt: "please make a blog post about the new PostgreSQL release, and give code examples for each new feature", then ctrl-c, ctrl-v the output.

You can write articles with the most horrible style if you want, but this level of disdain for readers is beyond belief.

garenpβ€’31 minutes ago
I must say I was immensely entertained by reading "IBM stepped up with buildfarm animals". :)