Comment by sillysaurusx
2 years ago
The black bar has intentionally been kept obscure over the decades. (HN is almost two decades old now: https://news.ycombinator.com/item?id=1)
Unfortunately, the answer to your question is very likely "No." There are a few subtle reasons why this is the case, and I'm going to attempt to explain them. I've seen this situation many times, and the outcome is almost always "HN doesn't change." This isn't due to laziness; adding the feature is two lines of Arc. The reason is social.
Social software is hard. In fact, it's one of the hardest types of software ever to be built. Things that might seem like small conveniences or improvements often have counterintuitive effects. These effects are not readily understood by people who aren't running the site, because only the people running the site can see them in detail.
For example, suppose we were to implement the black bar link. Firstly, this means that the black bar now becomes a "superslot", pinned at the very top of HN. It turns what was otherwise a subtle gesture into a feature. It will inevitably raise questions about whether the black bar is really warranted for so-and-so, or whether it's fair that they get the superslot. But criticisms like that can be ignored.
The bigger problem is one that Dan has pointed out many times: it's good to have readers dig a little for information. Only people who are motivated will end up showing up in the thread. And those are exactly the kinds of people who you want showing up in that thread, because the point is to honor whomever died, not to catapult the entire community (and then some) at the thread. After all, every single person who ever visits HN will immediately click the black bar if it was clickable. Are you sure this is the kind of effect that would be a Good Thing?
Then there is the truth that doing nothing leads to the optimal outcome. Suppose the black bar was changed, and it was a mistake. This mistake costs time, because now the moderator has to deal with the consequences. It's not just a matter of reverting the change; when stuff like that gets reverted, people get curious why. So it'd be natural to have to write an explanation, which ends up sparking discussion about very tricky subjects. Again, community software is hard, and explaining subtle reasons for doing X is a delicate process. All of this translates to the potential of wasting some unknown quantity of time; time you won't get back, and time that you won't be doing your duties of running HN.
Then there's the most subtle point: it would break tradition. pg was the originator of the black bar, along with the christmas colors. It might seem cheesy to people who haven't been here since 2006, but there's something magical about seeing HN behave exactly as it was originally written, even when that behavior is sometimes arguably less optimal than it otherwise could be. Because, again, every change has social effects, and these are very hard to predict.
Lastly, the person who died might not want all of the attention. Are you sure you really want to be spotlighted by the entire (tech) world when you pass? It wasn't till I had some uncomfortable moments in the spotlight that I realized that fame is sometimes something that people choose to avoid.
For all of those reasons and then some, the black bar is likely going to stay as it is. If only as a hat tip to the person who originally created the tradition.
I can understand remaining minimal, and not shuttling lookie-loos to an active thread.
But remaining mysterious kinda works against the "memorial" aspect, and making the editorial decision "this passing is worth noting". It creates an "in-the-know"/"not-in-the-know" division that's somewhat against the ethos of being fair & open with those who don't yet know something.
A simple, "RIP, [name] [YYYY]-[YYYY]" would be a kindness to readers, and enhance the respect paid the deceased, while avoiding the thread clickthroughs you're concerned about.
And otherwise, we'll keep having these threads every time the honoree's identity is sufficiently obscure, and any associated stories/remembrances scroll off the front page.
(I personally agree with you, but I just wanted to point out that https://news.ycombinator.com/front is how to see yesterday's front page at any given time. E.g. https://news.ycombinator.com/front?day=2024-01-19 makes it clear who the black bar was about.)
One (rather cold) way of viewing the situation is that if they spend their time adding this feature, it's at the expense of something else they could be doing. There are technical reasons it's not quite as straightforward as just setting a title attribute.
Here's precisely how the black bar works:
https://github.com/wting/hackernews/blob/5a3296417d23d1ecc90...
They uncomment that line of code by removing the semicolon, then deploy it. ("Deploy" is also known as "copy-pasting that function into a REPL," but devs like to overcomplicate it in other languages.)
If it were as simple as writing "Dave Mills [YYYY]-[YYYY]", they might actually do that. But Arc doesn't have keyword arguments, and tdcolor has no way of passing along a "title:" keyword in order to set the html attribute.
It's partly why I updated Arc with keyword arguments.
But, there's still that other aspect: figuring out [YYYY]-[YYYY] takes some amount of time, so it increases the activation energy of the black bar considerably. Right now it's a one-character deletion followed by a deploy. Honoring legendary hackers is a good thing, but the black bar already does that.
I understand tradeoffs & 'technical reasons'. But, a web framework that makes specifying tiny useful HTML edits noticeably harder in code than similar changes would be in plain-text strikes me as defective. That's especially true for a broadly-useful 'global attribute' – supported on all HTML elements – like TITLE.
Looking at the precedents in code near what you've linked, for how possibly-arbitrary STYLE and CLASS and COLSPAN and CELLPADDING and WIDTH attributes are set, it's hard to see why it shouldn't be as easy as something like:
Of course, it might not yet be that easy.
But if that's hard to support, with a couple low-risk broadly-beneficial lines elsewhere, then rumors of Arc's concise expressive power seem exaggerated.