As I said, I understand your reasoning (your description is as I'd thought it would be). I disagree with it.
Arguing over whether or not HN is a publication is awfully semantic. At the very least, it's a publication of links and the commentary of them. So we disagree there as well, though that's not particularly material.
There's a deeper issue of reputation, credibility, consequences, and more, that matter. This ties strongly into my "Big Questions" re-asking (submitted an hour or so back), and a suggestion from Ted Lemon in 2015: that a very large part of the addressible problem we're facing is the noosphere -- a superset of media, all of human thought. There's something very broken (and some terrifying consequences) to how ideas are created, propagated, (or propagandised), distorted, mistated, etc.
I'm not sure how to fix this. I'm not sure that it's fixable.
When I first started studying science -- in primary school -- I found the mentions of scientists names and such to be distractions. Shouldn't the ideas matter more than the person who had them? I've completely reversed my thinking, because it seems to me impossible to consider an idea without understanding the context within it, and context matters. The person themselves matters.
I've been thinking a lot about reputation, identity, and community, particularly regards my recent experiences with Imzy, and constrasting that to other spaces. In particular, identity IS reputation in a whole lot of ways -- starting with the etymologies of "fame", "reputation", "fame", "honor". Fully anonymity is also full impunity, and at least my experiences elsewhere suggest that works poorly. I've contrasted that strongly to HN's practices and results. Even where I don't always agree, or find elements frustrating, one thing HN does is, generally, allow tensions to dissipate fairly quickly, through UI/UX (including some of the annoying or missing elements of it), moderation, examples set, and more.
I realise that offering authors names risks a cult of personality. At the same time, some good (and bad) reputations are well deserved. Electronic communications generally strip away many of the normal social queues. Stripping away the few that are left, most especially authorship, strikes me as heading the wrong way. I'm a fan of Jakob Nielsen's microcontent guidance.
I'm also increasingly convinced that not offering users tools to quantify and qualify the content they access online is a tremendous failing of present tools. I've a limited capacity to really process information and messages, various sources (Stephen Wolfram, Walt Mossberg, NY Times moderation desk) suggest ~150 - 300 emails, ~800 comment moderations, are about the upper limit of what someone can sustainably absorb or process per day, over the long term. And that is if this is their full-time job. If you're looking at assimilating quality and complex information, the count's likely far lower.
As such: cues to quality, including reputation and identity, matter.