Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN admins: please stop editing submission titles
398 points by gnosis on Jan 31, 2013 | hide | past | web | favorite | 73 comments
The widespread practice of admins changing submission titles is incredibly annoying and obnoxious.

I would be more forgiving if the edited titles were actually clearer and more descriptive than the titles the submissions originally had. But much of the time they are worse.

Sometimes the titles will be changed to the title the article originally had. This would be fine if the title was a good one. But often the original article title is not very descriptive, and the submitted title is better. In this case, changing the title to the original article title helps no one. It makes the submission look more vague, obscure, and generic.

If the title changing was actually improving the site, I'd be all for it.

But you guys are incompetent, and are making HN worse. Please stop it!

You're probably wasting your time here. A previous thread about this got a lot of attention, and I got a nastygram from pg as a result, basically saying "quit wasting everyone's time with this issue." I don't think they're likely to change this policy anytime soon. Good luck, though.

His nastygrams are awesome:

Subject: "stop"

Message: "Please don't rewrite headlines to change the meaning."

I agree! How about this one: http://i.imgur.com/QwuX5aA.png

Edit: I'm talking about 'awesomeness' of his comments, not 'nastyness', don't take it out of the context pls.

Nasty is insulting your mom, the size of your penis, and your race. It's just short and straight to the point without a buncha fluff.

That definitely fits the nasty criteria.

Personally, this leans on the polite side to me.

It was more about 'awesome' rather than 'nasty'.

That's not nasty, it's frank and polite.

Can we settle on "brusque"?

I wonder if it caused as much trouble as his whole "new, revolutionary language" debacle.

I'm curious, what was the subject of your remark that caused that response?

Do you happen to remember which comment he's referring to?

What did you do?

This doesn't seem too bad to me. This message is straightforward, not nasty.

FWIW, I'm only using the term "nastygram" as a stock figure of speech. I didn't find the note I got from pg to be particularly nasty, although I did - and still do - disagree on this particular issue. shrug

What issue?

Admins changing the titles of submitted links. There have been a few cases where it happened and the "fixed" title was worse than the one provided by the submitter to a pretty ridiculous degree. I said something about that on one such story, my comment was the most upvoted comment by far, and sparked a huge thread on the subject. Hence the "nastygram" from pg.

No biggie, I understand that they don't want link titles full of gratuitous spin and editorializing, but sometimes the original title just isn't useful. But what can ya do... pg's site, pg's rules. :-)

There are nastygrams? Better than hell banning!

Very often the original article title assumes the reader is already familiar with a particular context, based on the fact he or she is already visiting the site it is hosted on. This of course does not apply to readers of Hacker News who see the article title totally out of context, surrounded by other articles.

I can see the value in editing titles that excessively call attention to themselves, include a comment by the poster, or inaccurately reflect the content, No one wants the front page of Hacker News to look like a page from Craigslist. The title should accurately describe the content of the article so readers can quickly decide whether to open it or move to the next one. The question asked by moderators should be whether the title fulfills this purpose, not whether it conforms to some bureaucratic guideline. Life is already too much distorted by slavish devotion to bureaucratic guidelines without bringing that mindset to Hacker News.

Too often the original article's title just plain sucks, even if the article content is good.

Many authors are good at writing interesting, informative, and useful content, but simply suck at choosing good titles.

I think some of them don't realize that on news aggregation sites like HN, articles often sink or swim by their title. At the very least, HN users will often not bother to look at something with a boring or generic looking title, and never know that the content was worth it.

Of course, if the submitted title was clearly abusive, like some ebay auction titles, using visual garbage like "* * * * L@@K * * * *", then those those submissions should just be deleted and their submitter warned and then banned if they persist.

But changing titles (by their submitters) to something more informative and interesting than the original article title should be allowed and even encouraged, as long as the new title is not misleading.

If HN admins did this themselves, that would be great. But way too often they change the titles to be more boring and generic (which is sometimes closer to the original article title, and sometimes not). This helps no one. They're effectively making HN less interesting, because fewer people will look at articles with boring and generic titles.

Actually, writers rarely choose the titles to their articles. That's what editors do. So it's the editors of publications that suck at choosing good titles, not the writers.

Many HN submissions are of articles written by authors who don't have any editors. Those articles are self-published either on blogs, personal websites, or forums.

Anyway, regardless of whether an author or editor is responsible for the title, way too many titles to articles with good content suck. Improving on those titles while staying faithful to the article's content should be encouraged, not discouraged or tampered with.

I find their changing of titles to be accurate, usually. It's nice that they change linkbait titles to accurate ones.

Nothing is worse than when someone posts a story then twists the title to fit their own interpretation of it.

Thank you oh wise and objective admins.

It's nice that they change linkbait titles to accurate ones.

Yes, and if that's all that happened, I doubt most people would complain. The problem is when more accurate / descriptive titles get changed to ones that are less representative of what the linked content is actually about. And yes, this absolutely does happen.

From what I've seen, it's usually the opposite. A fairly neutral, descriptive title is changed to the linkbait article title.

It is a bit annoying re-read or click the same comments on what seems a different or tangential story, get confused, and realize it was a title change.

I'm actually quite happy to see this too, I never have a problem with titles here like I do over at Reddit.

I have a feeling neither of you have had the titles to your own submissions changed.

"Can a 32-bit OS machine use up all 8GB RAM + 20GB page file?"

One of my submission titles was changed to the above.

I cried for days after.

Out of curiosity, what title did you originally submit? I noticed your submission was this URL, and the changed title matches the submission's actual title.


I reworded it to something like "Can a 32 bit machine use more than 4GB of ram"

Meh, it's dumb to call them incompetent over this. Have you seen how good they are at stopping spam, for one? Sometimes the changes make sense and sometimes they don't (depending on your point of view). I've seen it go both ways. This is hardly making HN worse. My suggestion: pick your fights, people.

I don't mind clarified headlines, but editorialized headlines bug the heck out of me and they can color the discussion. Today I complained when the title was "Are Placebos Really Sugar Pills? Or Something worse? instead of the original "Are Placebos Really Sugar Pills?" because it added innuendo.

I think I like the fact that the admins edit titles, but I wish the original title were viewable somewhere in the UI, especially when it gets referenced in a comment.

The guidelines of Hacker News


were actually edited to the current version quite recently, after an earlier discussion of this issue. The current guideline language is, among other details about titles,

" . . . .

"Otherwise please use the original title, unless it is misleading or linkbait."

That plainly says that original titles are preferred.

The guidelines also say, "Please submit the original source. If a blog post reports on something they found on another site, submit the latter." So blogspam is plainly disfavored. Personally, I think that it is a rare blog that has many posts worth submitting here, and I'm always on the lookout for better sources from which to submit articles.


By the way, as a rhetorical matter, I'm surprised after reading through all the other comments before posting mine that I don't see a lot of examples of before-and-after titles to see if the submitters who don't like the curators preserving original article titles (or shortening original article titles in a way different from the submitter) are really coming up with titles that are much better than the original article titles. What are the most important recent examples you have in mind?

My own experience on HN is that when I see a cool article from a good source, and I submit it with the original, professionally crafted title, I sometimes submit an article that was earlier submitted by someone else, and sank into oblivion because it was retitled in some way that made the article look dumb. Few participants on HN have much professional experience in headline writing, and I'd rather have most submissions be submitted with their original titles based on my observations. (If you have convincing counterexamples, that is examples of HN-user-made titles that were really good and more helpful than original article titles, I'd be glad to consider those examples.)

I have two examples where I'm of the opinion my title was clearly better suited for HN.

Goodwill Hunting (http://epicureandealmaker.blogspot.nl/2012/11/goodwill-hunti...) which I changed to something like "An investment bankers take on the Autonomy take over. This blog is well written and I've read every post on it but in most cases the subject of a post will only become clear when reading the post. Thus it is a bad title for use on a news aggregator.

GNU Typist (http://www.gnu.org/software/gtypist/) which I changed to "GNU Typist: Universal typing tutor" which is a variation on the first sentence of the page. My title clearly describes what one can expect, is not overly long or linkbaity.

Interested in the opinion of other HN'ers on my edits.

I don't have an opinion on the general matter, but in these two cases I agree with you - the edited titles are better for HN.

I think your edits were both good. Were they changed by HN editors?

Yes, and to add:

I willingly and knowingly broke the rules for the greater good. I expected that this would be better received here.


The new title is "Branch"

The original title was

"Obvious Corp's Branch now open to public"

If we're going to be limited to the title of the original piece why not enforce this, in the code? Fetching the page and parsing out a title tag really shouldn't be too many LOC - and would stop the debate.

And it would spark another debate "Please take that out of the code again".

Because this thread seems, sadly, devoid of examples (when there are plenty), I wanted to point out one that's active right now.

The front page's #2 link at the moment (http://news.ycombinator.com/item?id=5144325) is currently titled "Cache Rules Everything Around Me". That's better than the article's actual title, "Cache Money Hoes". I suspect that if the article's original title were used, far fewer people would have bothered to check it out, to have learned from it, to upvote it, or to comment in the thread about it.

I don't know if an editor will change it to the original title, but I certainly hope they leave it alone.

Refresh your inbox, you're going to be receiving a nasty email from PG soon... Having said that, I haven't seen many instances where a renamed title wasn't accurate. However, I would love to see a little label below the new title informing everyone the title is edited from the original. All it needs to say besides the timestamp beneath is: [title edited] or even just: [edited]

This has been discussed before. https://news.ycombinator.com/item?id=4102013

My main issue with changed titles is that it makes the submissions incredibly difficult to find later. Often times I'll mark them in my mind as "interesting" to come back to later, but when the titles change I find myself having to scan through every single title.

MyTitle: Another discussion of something on HN that will not change.

If you don't use the original title it'll probably get changed. There's nothing you can do about that.

If you care, add a comment after submitting, and put the title you would have preferred there, as I did at the top of this comment. In fact you could do that even if you're not the submitter.

I'd be interested in hearing the rationale for changing the titles.

Usually they'll change submission titles to match the original article's title.

I actually prefer it this way. This way, HN is neutral - by presenting the article title as-is, they aren't asserting any opinions on the subject (usually).

Er, no, they're asserting the opinion of the author of the article. That's not neutral at all.

That's the point. The article title should reflect the author's opinion, not the submitter's, nor HN's.

The goal is not to reflect anyone's opinion. It's to make HN a more useful and interesting site, which isn't going to happen when the tendency is to make article titles more boring and generic -- resulting in most users not even bothering to read the articles.

Why is the author's opinion more relevant or acceptable than the submitter's or HN's? You state it like it's obvious, but I don't see why the author's opinion is particularly special.

The way I see it, when I click a link, I'm signalling that I'm interested in reading the author's opinion, not the submitter's interpretation of the subject - that's what the comment section is for.

When I put it that way, it only makes sense that the title should reflect the original article's title.

When I'm reading through a list of titles on HN deciding which to click on, I'm more interested in why the submitter thinks it's relevant to HN and why they think I should click it. If I do click through, I will read the title.

We could present both the original title and the updated one within the comments thread, and allow people to vote on which title they like more.

The thing is that the attention a submission gets is almost completely determined by the title.

A crappy title will ruin the chances of most people even bothering to look at the comments.

So I don't think this is a workable idea. Just leave the submitted title as is, and let the HN readers vote it up if they think it's worth it.

I wonder whether a simple post-downvote would be more effective than the existing flagging mechanism

This is not reddit. Stop worrying about such things.

I look forward to some social news site offering some version of this; it could help create more informative and time-saving titles. Alas, I don't expect it here, given the HN minimalism.

Roughly how it might work: in addition to the 'comments' page, there'd be a 'headline' page. (Borrowing terminology from journalism, this might also be called the 'hed' page.) Posts here would be proposed alternate titles; the most upvoted would be the default on lists (like top stories).

Perhaps, list views could show the top 2-3 titles (maybe as a user-specific option, or only when there's controversy).

Each submission could also get a 'subhead' ('dek' in journalist lingo), for more exposition/teasing of content. It'd be voted on the same way.

Perhaps, even, a similar approach could be used for competitive summaries, but I suspect that's likely to result in more controversy and less consensus/convergence.

The only problem I have is that they tend to "Engineerize" the titles, if you will. The new titles are usually more bland. Maybe it's selection bias, but when I notice it's usually a "well, that was a lot more intriguing the first time I clicked this submission."

I dunno' - I can see the pros and cons of having more emotional or attention-getting titles, just as I can see the same for the "this is the information in the article" kind of title too. It's a tough line to balance on. Personally, I dig the emotion side more, but I also know I don't speak for everybody.

I'm personally fine with the policy. In several cases, where the title was too long, their edit of the title was better than mine.

I think the submission system may auto-change some titles. For instance, I submitted an article awhile back with the title something like "5 Reasons Technology is changing in 2013" and it posted as something like "Some Reasons Technology is changing in 2013."

Not really what you're referring to, but I thought it was interesting.

It was almost certainly previously submitted if it changed immediately like that. (It's not listed in your submissions page)

It was the "Common Tech Myths That Cost You Money" article. I was incorrect that it changed from a number to "some;" but it did purge the number.

I stand corrected.

x-post from the consumerist thread:


--"Please change the link title"

Worse is the articles they completely censor because they have some vested interest in the article not becoming popular.

i.e. http://news.ycombinator.com/item?id=5098218 had 200 points when it vanished from the front page.

Dear alumni and editors:

Please continue changing bad titles, changing links to original content, hiding bad stories, and banning abusive troll users. You make HN a lovely place, despite your somewhat behind-the-curtains wizardly appearance.



Could not disagree more.

Hell, I wish they'd hellban question-mark titles.

Seeing both would be nice - something along the lines of:

"Original Title [A brief description if the original title is insufficiently detailed]"

I prefer that the headlines be rewritten. If I want what you're describing, I'd go to reddit.

So you prefer Slashdot?

Applications are open for YC Summer 2019

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