我在HN上的哪些评论被点赞了?

2作者: sema4hacker2 个月前原帖
我花了太多时间在HN上发表评论。我从来无法预测哪些评论会被点赞,所以我将它们整理成表格,以查看是否有任何趋势: ``` 68分 1条评论 关于马斯克与OpenAI 39分 1条评论 关于私募股权 26分 1条评论 关于Teamshares 24分 1条评论 关于监控 23分 1条评论 关于项目提案 19分 1条评论 关于复杂网页 16分 1条评论 关于消防车效率 15分 2条评论 关于SQLite和电梯费用 13分 2条评论 关于AI编码错误和平衡立方体 12分 2条评论 关于家庭垃圾和我的薪资历史 11分 4条评论 关于研究对象的年龄、拉斯维加斯球体、我的技术历史(2次) 10分 4条评论 关于Waymo、害虫承包商、YouTube十进制、窃贼 9分 11条评论 关于生产力、一个糟糕的评论、监控滥用、TLDR风力涡轮机、机器人捡垃圾、人类记忆、中本聪、AI著作权、糟糕的商业、条形码安全、我的技术历史、Windows Cortana 8分 5条评论 关于SQL、POE交换机、写作质量、节食、我的技术教育 7分 5条评论 关于格式、拟人化AI、AI泡沫、技术就业与收入、Pascal 6分 7条评论 关于能力与可用时间、影响与操控、糟糕的客户服务、Lisa GUI、满意度、推荐书籍、我的薪资历史 5分 6条评论 关于网络提供商、接受信用卡、网页代码、旧文件、机器人捡垃圾、放弃Macintosh 4分 14条评论 关于:项目估算、面向对象编程、AI泡沫、糟糕的摩天大楼、调试、UCSD Pascal、拟人化AI、便宜的卡车、阿司匹林、Unix历史、原始文章年龄、足球中心、服务员、我的技术历史 点数为3或更少的评论太多,无法整理。 ``` 我没有发现主题、点赞数、写作质量等之间有明显的趋势或关联。
查看原文
I spend too much time on HN posting comments. I can never predict which will be upvoted, so I tabulated them to see if there was any trend:<p><pre><code> 68 points on 1 comment about Musk vs. OpenAI 39 points on 1 comment about private equity 26 points on 1 comment about Teamshares 24 points on 1 comment about surveillance 23 points on 1 comment about project pitch 19 points on 1 comment about complicated web pages 16 points on 1 comment about fire truck efficiency 15 points on 2 comments about SQLite, elevator cost 13 points on 2 comments about AI coding errors, balancing cube 12 points on 2 comments about domestic garbage, my salary history 11 points on 4 comments about age of study subjects, Las Vegas Sphere, my tech history (2x) 10 points on 4 comments about Waymo, pest contractors, YouTube Dewey decimal, burglars 9 points on 11 comments about productivity, a bad comment, surveillance misuse, TLDR wind turbines, robot litter pick up, human memory, Satoshi, AI authorship, bad businesses, barcode security, my tech history, Windows Cortana 8 points on 5 comments about SQL, POE Switches, writing quality, dieting, my tech education 7 points on 5 comments about formatting, anthropomorphizing AI, AI bubble, tech employment vs. income, Pascal 6 points on 7 comments about ability and available time, influence vs. manipulation, bad customer service, Lisa GUI, satisfaction, recommended books, my salary history 5 points on 6 comments about network providers, accepting credit cards, web page code, old files, robot litter pick up, abandoning Macintosh 4 points on 14 comments about: estimating projects, OOP, AI bubble, bad skyscrapers, debugging, UCSD Pascal, anthropomorphizing AI, cheap trucks, aspirin, Unix history, original article age, football centers, waiters, my tech history Comments with 3 or fewer points are too many to tabulate. </code></pre> I see no discernible trend or correlation between subject matter, upvotes, writing quality, whatever.