Improve search engine results for Dorico docs

Hi everyone

I assume I’m not the only one googling for “dorico [problem x]” when I need some help using Dorico. Every time I do this, I realize that the current documentation is almost never part of the search results.

I assume that this is the case because search engines consider the different versions of the documentation as duplicate content and prefer the oldest version of the content. Example: If you search for “dorico beaming”, you will get a result for Dorico v2 (!), while the v5 version is not even in the index.

I suggest improving this situation by adding canonical meta tags to the old versions of the docs whenever a new major version is released.

Search Dorico 5 [problem x] and you are more likely to get the result you want.

Also check the url of the search result, which includes the version number of the documentation.

This topic has been discussed at length if you care to search this forum!

1 Like

Thanks, @Janus. I’m aware that I get better results with better search input, however the duplicate content issue is something not to be solved by the end user but by the content provider. It is something that will only get worse with each major release.

A short comparison:

Google is not just doing some inexplicable magic, but (more or less) behaving by a set of documented rules. As content provider, Steinberg can check for indexing problems and fix them, as explained e.g. by creating canonical tags.

I refer you to previous discussions on this forum.

But I would note that there are still many users of older versions, who continue to seek answers via Google. Would you wish to deny them access to that information?

1 Like

I have read some of them, but they are promoting workarounds or tools to mitigate the problem. My post is about suggestions to address the root of the problem.

No, I’m not suggesting to remove older docs from search indexes. I’m suggesting to use canonical tags to tell search engines which results should be preferred and which pages should be indexed most often.

I’m not sure if the platform used to document Dorico provides this as a feature, but there are platforms that do, e.g. Sphinx and MkDocs.

The manuals team is aware of “version variability” in online search results. Changes are planned in this area, so watch this space for future improvements to how our documentation is delivered.