Hacker News new | past | comments | ask | show | jobs | submit | tkrunning's comments login

Shhh... ;)


Good point! I'll update it :)

I just threw the site together in a few minutes before publishing the post (experimenting with Gatbsy), so literally just added a max-width and changed the font styles.


This is definitely true. What I'm referring to in the article is cases when you've done a proper update of the article to make sure it's up to date, added new information, etc.

I actually hope Google will start comparing versions of articles before/after date updates, and penalize those that update the date without updating the content.


And while I appreciate good updated content, I still insist that the _publish_ date should not change. In your case, the publish date ought to be many years ago. The _updated_ date, however, may be re-written with those refreshes. Content should include both (when applicable) and indexes should reflect this.


Medium has quietly removed the support article describing this, but it used to live here: https://help.medium.com/hc/en-us/articles/227017408-Set-cano...

Sadly, it's not archived anywhere I could find, but it did mention that you could send them a CSV file of up to 50 URLs to update the canonical links for.


That doesn't contradict anything the agent said. I believe you that they could previously set canonical links. But it's highly likely that upper management changed that policy (I'm surprised they allowed it to begin with).


IMO, it's not too surprising that they tried to be a bit more writer friendly initially (to attract more quality writers and larger publications). I guess it no longer fit with their new business objectives, which is why they quietly removed it this fall.


I'm not surprised and as I say in the article "A heads-up would have been nice, but still, that's their prerogative."

I do understand why they would want to move in the direction they are, but I also think authors in particular should be aware of the trends and make choices that will benefit them, rather than Medium, in the long run.

What's a bit sneaky here is that they have been selling authors on the fact that they own the content, and the fact that you can move it elsewhere later (including updating canonical links). It's a bit like a SaaS offering an export function, but after you put a lot of your data there they quietly remove it. For bloggers, you can't claim to give them proper data portability without a way of updating the canonical links.


It's not only a problem with the search engines' own rankings, it's also that people are less likely to click an older article on the SERP.


That’s definitely true. I often find myself clicking first on a more recent search result and then backing out and clicking on older content if the initial choice didn’t live up to expectations. Since I probably still had decent time on page, etc for the first link, I’m guessing I’m making Google think that content is more relevant than it is.


Feel free to take a look for yourself: https://cl.ly/93ce51946717

I just removed the support agent's name (no need to name names), otherwise the emails are complete.


I don't think Medium is able to plug the hole that easily, but they might tell me to stop doing it or plug the hole just for me. I don't believe I'm violating their ToS, but you never know how they would interpret it.


I don't believe I'm violating their ToS, but you never know how they would interpret it. That's why I'm not sharing it publicly. But feel free to email me on mediumtrick at nomad dot email.


I only set up this blog for this specific article, so haven't given following much thought, but yes it does have RSS (https://bts.nomadgate.com/rss.xml). And people do still use that.

My main blog (https://nomadgate.com) offers more options: RSS, web push, email newsletter, community forum, etc.


Alright, I must've missed that somehow, thanks!


Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: