Obsidian acts like web browsers do right now so I wouldn’t say it goes against the typical expectation given that most people are familiar with the way web browsers operate.
I can understand where the decision to make links open in the current tab came from based on this logic of trying to emulate something users are familiar with, and I do think that’s generally a good UX design strategy, but in this case I think there’s more that would be useful to add to the analysis.
Using browsers cultivates the expectation in users that internal links will open in the same tab, and external links will open in new tabs. For me, I would consider a link to a different heading within the same file I’m working in to be an “internal link”, and I would indeed expect that to open in the current tab (i.e. just navigate me to that part of the page). But when I’m opening a different file (by clicking a link to a different file, using ctrl + O, using the file explorer, etc), I would consider that to be more akin to an “external link”, which browsers teach users to expect to open in a new tab.
I would also point out that over the last decade or so browsers have been increasingly switching their paradigm to make more actions that open URLs (such as opening a page from one’s recently closed tabs, or one’s bookmarks, or one’s history) open the resulting page in a new tab. They’ve also taken steps to make it easier than it used to be for users to open links in new tabs, including by providing a settings so users can make all links open in new tabs (which is exactly what this feature request is asking for in Obsidian), and by making “Open in new tab” usually the first item in the context menu you get when you right click a link (compare that to Obsidian where it is not). I think those changes (along with the creation of browser tabs in the first place, prior to which you used to have to have multiple browser windows open) are a reflection of the fact that the paradigm of opening links in the same tab has limitations that create friction for users under many circumstances.
Lastly, I’d like to say that I think IDEs are a better model for how Obsidian should work than browsers. Yes, browsers are more commonly used by more people, so they provide a more familiar paradigm to more users. But what you are actually trying to do in an IDE (simultaneously viewing and editing text, usually spread across different files) is IMO a much better match for what I think most users are doing in Obsidian as compared to what most users are usually doing in web browsers, which is often just viewing information - I know I often have a hand on my mouse and no hands on the keyboard while using a browser, which is almost never the case while I’m using Obsidian or an IDE. And there’s a good reason why IDEs work the way they do. It is much easier and quicker to view and edit text across different files when your editor understands how the files are related (Obsidian does this fairly well, with things like backlinks, which I would say are a very IDE-like feature), and also when you can easily open multiple related files at once and switch between them (this works okay in Obsidian, as long as you always ctrl + click when opening new files, but I think the point of this feature request is that it’d be significantly better if holding ctrl wasn’t necessary, as demonstrated by the popularity of the request)