Go started out in C, and was later (post-1.0) incrementally rewritten to be self-hosting. One of the authors (Ken Thompson) is also one of the co-creators of C. I would argue these guys know what they are doing.
Also, the reason for compiler bootstrapping is more of a “beauty thing”, then practicality. It would definitely be faster in a low-level language, but I doubt it would matter as an end user.
EDIT: I meant to write that I think very highly of him as an architect/developer.
I get the desire for generics. I do a lot of C# and have used generics for a very many years. Yet I've been writing Go for around 6 or 7 years and other than in the beginning (when I was new to it) I haven't found myself missing them at all.
In other words, for many people the lack of generics comes across as an oversight. For others, including myself (again, a heavy generics user in C#) that really isn't the case. I write Go in the style of Go and it just hasn't been an issue.
Blanket statements are rarely true. YMMV.
Sure, it is seldom missed as an end user, but as a library user it is essential. That’s why map and the like had to be hard coded into the language, and why concurrent versions couldn’t be implemented for a long time in the language, the same way it was done for Java forever.
A bit contradictory, really, but that's just semantics I suppose.
More importantly even if you could say 99% of devs agree (and you can't because they don't) that still doesn't make it an oversight.
If they'd neglected to add generics because it wasn't considered, that's an oversight. If it was neglected because in the opinion of the creators of the language it wasn't needed for the purposes they created it for, that isn't an oversight but a thought-through engineering decision.
Of course you're free to disagree with that decision, but an oversight it was not.
Whilst I haven't missed it in Go myself, enough other people say they do that its inclusion was inevitable. Which means it probably should have gone in sooner.
After several years and attempts at a good enough proposal, Ian Lance Taylor put one out that was able to cross the finish line, and now we have generics.
You know what, with all the 'discussions' in recent years about whether Go should have generics I'd actually lost track of that amongst all the noise. Which is irritating, as I do now remember the early conversations about it.
Thank you for the reminder.