Nah, sitemap leads to canonical versions of all pages in hugo. Alias is just an “extra” page not mentioned in sitemap at all.
To conclude, the current behavior of hugo is not wrong and other search engines do work and do find my blog, it is just Google does not for unknown reason and due to its dominance, I think it is pretty unsafe to have such default behavior.
My bad I thought you meant sitemap to the old. But the point stands that all the signals (backlinks, social signals) would be lost if they just index the new page.
I didn’t leave mine broken long enough to see if it would eventually index the new.
I beg to differ. I think it is wrong. But I have my site fixed for years just by doing custom. I haven’t thought about this issue since the day I fixed it all those years ago, until today.
I understand what you mean and I do not disagree. I think this all depends on how we all interpret the noindex in a custom alias page.
I think it is how Google interprets it. 301 redirects are just there to make Google update the canonical to the new page. That’s why meta refresh redirects work. Google just sees the canonical on the page. If you do not have the page indexable by Google it won’t see it. It won’t index it either if the canonical is there.
I think that’s the key. See:
https://developers.google.com/search/docs/crawling-indexing/consolidate-duplicate-urls
While other search engines might handle meta refresh + noindex differently, I can’t image any search engine would crawl a page where the URL is different than its canonical URL.
We don’t recommend using
noindexto prevent selection of a canonical page within a single site, because it will completely block the page from Search.rel="canonical"linkannotations are the preferred solution.
@jmooring this comes from the link you shared. If I read this correct it says you should not use ‘noindex’ to select the canonical page… which is opposite to what the default template does.
This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.
The way I read that quote, we’re doing exactly what we should be doing. We want to completely block “the page” from Search, because “the page” we’re talking about is the alias, not the canonical content.
Regardless of the (at least to me) unexpected indexing behavior, I am in favor of merging #13951.