Remind HN: Please Submit Original Articles and Titles and Canonical URLs

Some recent threads here on HN reminded me to review the Hacker News guidelines linked to from the bottom of the main page and other pages here. There are two guidelines that are good reminders of how to resolve some of the issues that have received meta-discussion here. Let's discuss the guidelines and how to apply them. The headings I've written below emphasize practices that I think are helpful:

1. PREFER ORIGINAL, JOURNALISTIC TITLES

"In Submissions

"Please don't do things to make titles stand out, like using uppercase or exclamation points, or adding a parenthetical remark saying how great an article is. It's implicit in submitting something that you think it's important."

A few paragraphs below that, the guidelines say, "You can make up a new title if you want, but if you put gratuitous editorial spin on it, the editors may rewrite it."

One user benefit of original titles is remembering which articles have already been submitted, perhaps with no discussion (the most frequent case for any new submission) or perhaps with much discussion and placement on the front page of HN just yesterday. Another user benefit of original titles on new submissions is that other things being equal, most titles devised by journalists actually grab attention better (and with less spin on the discussion) than most titles made up by HN participants. New submissions on the Hacker News new page are now so frequent that any article is at risk of being missed, no matter how good it is, but I've seen really, really, really good new articles fail to gain the discussion I think they deserve mostly because they are submitted with a modified title, obscuring what they are about or making the articles sound more dull than the original title did. (Some good articles have an article title that heads the article and a cross-reference title used from the publication main page, and sometimes the cross-reference titles are better than the main article titles at BRIEFLY describing the article. We all get to try to use our best judgment when given a choice of titles.)

2. PREFER ORIGINAL SOURCES WITH CANONICAL URLs

The Hacker News guidelines say "Please submit the original source. If a blog post reports on something they found on another site, submit the latter." Blogspam is highly disfavored here on HN, as I have observed repeatedly in comments after 1305 days here. Everyone prefers submissions of sources that actually do some fact-checking and research before writing, rather than just copying other people's writing. In particular, ScienceDaily and PhysOrg are not original sources and are often not reliable sources. There is a better source for any article you find in either ScienceDaily or PhysOrg, and many Hacker News readers have learned to disregard articles from either of those press release digest companies.

A trickier issue is what URL to submit from sources that give articles multiple URLs. For a while the online site of The Economist (a very good source appreciated by a lot of HN readers) was quite inconsistent in how it assigned URLs to articles, giving many articles two or three different URLs depending on how they were linked internally on the site. Other sites distinguish print versions of articles or mobile-viewable versions of articles from the main view of an article with different URLs. I've seen a lot of discussion of preferred URLs of articles here on HN over the years. The suggestion I pick up from other users here is to prefer the "canonical URL" (a Web search for that exact phrase will find sources to explain that concept) for each and every article submitted here. Hacker News runs an automated detector of duplicate submissions that works best (upvoting the first submission of an article when a duplicate comes in) if articles are submitted with their canonical URLs. If you read the article first in a news aggregator (e.g., Google News) or in an RSS feed or as a link out on someone's blog, as you submit the URL, chop off the extra characters at the end that reveal the source of the link. Some readers like to read mobile versions of articles, and some readers like to read print versions of articles, and some readers like to read one-page versions of articles. The canonical URL from many of the better sites on the Web should either autodetect a reader's Web client or format to a preestablished user preference. The other URLs may be what you like, but they have unpredictable behavior when submitted to HN and read on many other user's devices--with an especially annoying behavior being starting a print job as soon as the page is accessed. Just submit the canonical URL, and let the user decide what to do from there.

Similarly, just submit a good new article, and let users decide from there what to do about it. A lot of good new articles get ignored. No one has time to read all of them. HN users decide for themselves what to read. If some good article was submitted a few days ago, and no one noticed it, take the site title "Hacker NEWS" as a suggestion to look for something new, something from today's news, and try submitting that for the discussion it deserves. There are always plenty of new articles to read. (In the occasional case of resubmitting a "classic" article on a recurrent issue that has come up again in HN discussion, which is apparently allowed by the duplicate detector eventually lapsing after a set time, then please include the article date as a parenthical comment on the title, e.g. "How to Disagree" (March 2008).)

Following the habits of other HN users who began here even earlier than I did, if I see a duplicate submission that escaped the duplicate detector (usually because one submission or the other wasn't a submission of the canonical URL), I try to note the earlier submission, with a link, in a comment on the later submission(s). In such cases, I am not expressing any opinion about what other HN readers should do about the observation that the link has been submitted twice (or more). I just note that so that people like me who wonder "Haven't I seen that before?" can get a reality check on what they remember. Where (and whether, and how) people decide to discuss articles is up to each reader here. Over the years, I've appreciated comments like that from the previous users who have that habit, so I upvote their comments and have begun emulating them. I don't think the previous users who have that habit mean any offense by that, and I certainly don't.

I'd appreciate comments here from other HN readers about what submission practices are most helpful to the community (even if you disagree with me). Best wishes for a great reading and discussion experience here.