Hacker News new | past | comments | ask | show | jobs | submit login

Every time I’ve tried it, it falls apart within minutes of using it. You’d think after 20 years of development it might be somewhat stable but I guess not.



It's both a very hard problem, and a very interesting problem.

My impression is that most of the work on the project is motivated by interest in learning about OSs, rather than demand from real use-cases.

If the problem had strong enough demand from real use-cases, I presume it would be stable by now.


IMO, the is a monumental underestimation of the man-hours put into make Windows as compatible as it is, even ignoring the fact that Microsoft's engineers are/were highly skilled.

I like the ReactOS idea, but I think it will take decades to be compatible enough - and I'm even talking about enough.

If the objective was to run obsolete software, I personally had to distribute resources, I'd just do assign them to Wine. It's surely much more productive to make software compatible on a per-case basis, rather than writing a whole operating system with the objective of being compatible.


But I'm not disappointed in ReactOS because it's not fully compatible. I'm disappointed because it's not stable.

(Or was, last time I tried a couple of years ago.)

I think they should go for rock solid stable first. Then people could build on it. It's perfect for embedded, Point-of-Sale etc where the vendor want to keep their codebase untouched but hop off the Microsoft bandwagon.


> I like the ReactOS idea, but I think it will take decades to be compatible enough - and I'm even talking about enough.

It has been under development for decades. That is more than twice the time it took to develop Windows 95. I've tried ReactOS a few times over the years, and it has never worked for me.

I know of two specific examples of ReactOS code that as of a couple years ago were identical with Windows code. There are likely other instances pointing to a non-clean-room implementation.


>I know of two specific examples of ReactOS code that as of a couple years ago were identical with Windows code.

Cool; if they're specific then you ought to be able to cite them for the rest of us.



Odd, here's the commit history for that file: https://github.com/reactos/reactos/commits/master/ntoskrnl/k...

What I'm wondering is if you just chanced upon a submission that wasn't actually accepted into the tree at any point?

Perhaps because it literally says:

>/* This part was copied from Windows source /

> / We should obfuscate this somehow? */

And looks sketchy as hell?

I'm not part of their project (just a fan) and my knowledge of git/github is shaky at best -but that's the way it looks from where I'm standing.

Also, I'm not seeing the original windows source code itself; only a comment.


[dead]


We've banned this account - it's not ok to mimic someone else's username like that, but it's especially not ok to add sockpuppet accounts to a thread you're already in. Doing that will get your main account banned as well, so please don't do that again.


A link to a random blob means nothing if it's not actually in the tree. Here's how that link was faked:

- https://news.ycombinator.com/item?id=10005577

- https://news.ycombinator.com/item?id=21025378


not really a fair comparison, windows 95 had the advantage of not having to be developed from the ground up to retain perfect binary compatibility with blindows 95




Consider applying for YC's W25 batch! Applications are open till Nov 12.

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

Search: