4 ms·
Setting up context fail.
by adamkochanowicz 11y ago
Setting up context fail.
- MereInterest 11y agoPrimarily due to the requirement that post titles be identical to article titles. If a post is intended for a small audience who are already aware of what is going on, then little additional context would be needed. If Hacker News allowed modifying titles, a reasonable one would be "Over 50,000 Old Technology Manuals Rescued".
- bmir-alum-007 11y agoPerhaps if there were some semantic way for HN to link an article to a previous one, like threading. For example, a text field with previous HN item ID and a type drop-down with "duplicate"# and "follow-up" fields. # For editors to reduce duplicate posts.
- gus_massa 11y agoThe strict requirement of using the exact title as the article was lifted a few months ago, but it doesn't mean that someone can editorialize freely the submission title. I think that the preferred alternative is one sentence of the subtitle / opening paragraph / abstract of the article. In this case I can't find anything useful. I guess a correct alternative is to use the title of the previous submission to provide context so my proposal is "Saving 25,000 Manual: It Is Done" Anyway, the mods can disagree and change it back to the original, so YMMV. (Note 1: They saved more than 25,000 :).) (Note 2: I think your proposed title is good, but I tried to create one using cut&paste text from the article.)
- DanBC 11y ago> The strict requirement of using the exact title as the article was lifted a few months ago The title thing seems to have caused confusion. dang has said that there hasn't been a change in policy (nor in the way the policy has been applied) and that the policy has always been "the original title is preferred except when it is misleading or linkbait." https://news.ycombinator.com/item?id=8227198#up_8228566 https://news.ycombinator.com/item?id=8227198#up_8228566 But this meme of "you must only ever use the original title, even if it's terrible" is pretty strong. I think people don't understand that HN seems to prefer titles that convey little meaning over titles that create kneejerk reactions. There are plenty of posts on HN from people who are responding to the title, and who haven't read the submitted article. (EDIT: I think I upvoted you. I meant to. I might have accidentally downvoted, and I'm sorry if I did).
- Kluny 11y agoHe's been posting updates from this blog on HackerNews for the past week or so. It's well worth reading the previous posts to get up to speed.
- ersii 11y agoThe author does have a HackerNews account, which is "textfiles" (Note the domain name of the blog). He has not re-published this on HN himself.
- mikeash 11y agoI never heard of any of this but I had no trouble figuring out what they were talking about.
- enraged_camel 11y agoReally? You figured it out just by looking at the title and the domain? I haven't been following any of this stuff, and I wasn't familiar with textfiles before. I clicked the link only by accident. I actually meant to click the submission above it, which was much more descriptive and immediately told me what the submission was about: "How Coursera Cracked the Chinese Market". I know the title rule exists for a reason, but let's not kid ourselves: some titles are written by the author only with a specific context in mind, which is usually their own website or blog. Such titles aren't a good fit for HN.
- mikeash 11y agoWho said anything about just looking at the title and domain?
- InclinedPlane 11y agoThis is literally the second paragraph: "The goal, as you might recall, was to go into a closing warehouse of manuals, take as much as we could, and store it somewhere so that it could be properly “dealt with”, with the rest ending up in the trash."