I wrote about this at https://news.ycombinator.com/item?id=11538480 and https://news.ycombinator.com/item?id=11538499.
Better, I'll even tell you why: I use summaries to decide to read the original article or not. Your argument seems to stand on the idea that a short title is enough to decide if an article is worth my time to read. It often is not. The first thing when I see a title that looks interesting is to read the first few top comments to see if other HN reader have found it interesting and get more information about the content.
The extreme version of your position would be that we don't need titles either: HN reader should just trust that top links are intersting since they made it to the top and the title line would just give them incorrect information to discuss, everyone should just read every article that appears in the top 30.
This is obviously ridiculous.
You think titles are enough. I don't, and otehr obviously also disagree.
A summary doesn't tell you that though. The comment ceases to be a comment and instead functions as a sort of mechanical interpretation. As technical people concerned with taxonomy, we can't have "summary comments", that's just crazy talk!
Not only that, but what of the ethical dilemma of posting a "TL;DR" for something that isn't your own writing? That's self-appointed editorial control when nobody agreed to that.
That said, I like summaries, but only when a genuine opinion comes with them. The summary service I see mentioned sounds interesting for sure, definitely a good idea. Although it's not quite clear to me how I would access that.