Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

It seems like the title length constraint on HN is a bit too limiting and often results in these types of clarifying comments.

Couldn’t we expect more accurate and higher quality titles by relaxing the length constraint? I’m sure it’s been discussed before here, but I’m struggling to think of downsides from such a change.



Having short titles where you can't cram a lot of information is a feature IMO, since it means that you generally have to actually open the link and see what it's actually talking about.

I suspect (without having any evidence) that the longest the titles are, the most likely you are to end up with people commenting without having even bothered to open the link.

Which, hypocritically, is what I just did, but admittedly this comment isn't about the article itself.


It's a tradeoff.

In my view, the primary function of the title is to help a viewer understand whether the topic is of interest to them and therefore it is worthwhile to click the link. A more-informative title will better serve this critical function.

If some HN participants are prone to making off-base comments based solely on a title, let's address that directly (e.g. via guidelines + voting), rather than by nerfing the titles. Otherwise we're just throwing up our hands and saying "this is why we can't have nice things". I'd rather work toward having the nice things.

Incidentally, it's not just (or even primarily) about length; an article's "native" title often makes sense only in local context, and thus does not communicate well when seen out-of-context on the HN front page.


It’s good to make HN users dig for information. HN is designed to gratify intellectual curiosity; it’s the same reason TL;DRs are tolerated but discouraged. Dan has written about this many times over many years — I imagine he might drop by with a few persuasive references.

You’re right, by the way. It would be good if HN was designed solely for growth. It might even have been good in this one case, too. But it’s designed to spark curiosity, which is a much more delicate thing. Most long titles are long because they’re noisy. They don’t usually add precision.


I'm honestly baffled by this argument. Do you click through 100% of the links that reach the front page? More to the point, do you expect/desire that most HN readers should do so? Presumably not; so, we must expect that readers -- no matter how curious -- are exercising some decision algorithm. Why deliberately cripple the input to that algorithm? Shouldn't we trust readers to exercise curiosity even when provided with a good descriptive title?


It's fine to be baffled. For what it's worth, I was also skeptical for a long time.

Mistaken titles should be corrected. Clickbait titles should be reduced. But that's not what's being argued here.

What's being proposed is to change a longstanding community rule. Such things are known to happen, but you have to be careful about doing so. It's almost irresistible to propose changes. I recently proposed one too: that all links in selfposts should be clickable. I still feel that was a decent suggestion.

But we don't have the experience or the information to see all the possible implications of such proposals. Having been on Ye Olde HN for... 2021 minus 2007 years, I think the central question is whether the rule breaks down at scale. Because the title length in place since 2007, and the only reason to change it is that it no longer works, presumably due to HN scaling.

My skepticism alarms go off at such proposals. The clickable links in selfposts are a decent example of a proposal that seems to fit: no one had clickable links in 2007, even for posts from YC. But as HN scaled, that changed. Presently, most posts that make it to the front page get clickable links, so those that don't feel like obvious outliers – the shunned posts. Why not let everyone participate in a fair way?

The title length proposal is different. It's true that it might make some posts more accurate, like this one. But it's also true that a sufficiently creative person can pack a lot of information into 80 characters. Are you sure it's a good idea to change such a longstanding rule, especially when there's no pressing need to do so? Doing nothing is often the best course of action when running something – look how freenode turned out.

The point is, each proposal like this needs to be carefully thought through. It might seem entirely obvious that it's a good idea, much like the selfpost proposal seems like a good idea to me. But we should try to feel skeptical – how much money would you wager that your proposal won't go wrong? Would you place 450k on it? I wouldn't.

But we're asking them to bet far more than $50k on each change like this, because HN is literally the key to YC's power. It always has been. That's why I'm not too bothered if things stay mostly the same – there have been a lot of changes since 2007, but the substance of the site (such as the 30 link limit on the front page) has remained the same.

In fact, one could ask oneself "Why not show more links on the front page? After all, 40 would be more informative than 30." Many of your same arguments would apply. Yet there are subtle but important reasons not to.


Maybe a different limit for links versus "ask HN"s?

External links have some sort of constraint, weak as it may be. The limit more forces people to editorialize rather than focusing their thoughts toward concisitude.


The front page is a treasure. Many redesigns are attempted and they never work (for me, YMMV). Don’t jiggle what obviously works and everyone is very used to, unless you test the heck out of it.

I read the short headline, and the top comment gave me the clarity I need. As is often the case I don’t need to read the article.


It would involve a change to HN.

There's still no mobile friendly stylesheet. I wouldn't hold your breath.


I mostly read HN on my phone and I think the experience is fine. So maybe there’s just not enough demand on that front.


> I think the experience is fine

How do you vote without zooming?

Ever tried to touch "comments" on headline and accidentally clicked "hide"?

Ever tried to touch the "Hacker News" menu item and gone to "new"?

Everyone is used to these things, but they are textbook examples of bad mobile design.


Number of times I’ve logged myself out...


> How do you vote without zooming?

I press the voting button. Most of the time I vote in the intended direction. I think. If I don't, well, there's no way to know.


The text next to the timestamp changes to unvote / undown depending on the direction of your vote


On my android device, it flashes a larger graphic briefly. I look for that to know my vote was the right direction.


You know, I just quickly zoom with a double tap and nail the UX element.

We have better elements to use, but I must also say those tend to come with costs.

Right now, HN is so lean, fast and clean, I will gladly work a little to vote or do some action in return for what is otherwise one of the best "just read the discussion" presentations on mobile. It's a pleasure.


People have been clamoring for mobile friendly blockquote for a while.


People abuse the code formatting for block quote. It looks bad on mobile, but it also looks bad on desktop because it's a fixed width Font, which is not what you want for a quote. The solution is not to change the style – it's for people to stop using code style for quotes! (The style does actually seem to have changed on mobile in the last year or so: I think it wraps now, whereas I think it used to have a horizontal scroll bar.)

To quote something else, manually insert a greater than symbol at the start of every quoted paragraph,

> Like this

Which obviously doesn't indent nicely but is perfectly clear, and works well with HN's low formatting style.


I personally really like the Android app -- it's an excellent reader, and is my preferred way to interact on mobile.


Think for a moment why an app should ever be considered necessary to render a web site. Especially one with little in the way of demanding UX requirements.


> There's still no mobile friendly stylesheet. I wouldn't hold your breath.

Beyond code blocks having line wrap, the HN mobile stylesheet seems fine to me. What issue do you take with it?


Well there isn't a mobile stylesheet. It's the same as the desktop styling (and that code block issue affects desktop too).

There are multiple issues with using the desktop styling, but most are related to Fitt's law[1].

The whole UI is terrible for finger interactions, but the best example is the tiny upvote/downvote buttons immediately above/below each other well within the diameter of a normal finger. It's literally impossible to use that without zooming, and if you try to then there is no way to know if you vote up or down. It should be used in textbooks for how not to do a mobile interaction.

[1] https://www.interaction-design.org/literature/article/fitts-...


I agree the touch targets are ridiculous, but there is feedback - upvotes make the "unvote" text appear and downvotes make the "undown" text appear.


Good point. I forgot the labels are different. They are in tiny grey text, the same length and require you to actually look hard to noticed though.


> It's literally impossible to use that without zooming

Let's keep the discussion honest. I have a high DPI phone screen, I'm at the normal text size for my device, and I regularly use those buttons without zooming with reasonable accuracy.

The hyperbole is not warranted here. While I agree touch targets could be bigger they are certainly not impossible to use reliably.


This is the best thing about HN. Stubborn for change.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: