Hacker News Reader

Comment by dtx1 | Hacker News Reader

[flagged]

How much does Mozilla spend hardening Firefox? How much does Apple spend hardening Safari? I see no reason to believe that Chromium is so far ahead security-wise that the only way to compete is with Rust. Happy to be proven wrong, of course.

What’s a Topah?

Klingon for animal?

What’s a Klingon?

What's Google?

You spelled googol wrong.

https://youtu.be/ibA7noX6rnc?t=63

I think it's like a jabroni.

What’s a Topah?

Klingon slang for Idiot

What is a Tophas?

he's probably cursing in klingon.

HN turns into slashdot on sundays, when all the MBAs are barbecuing.

I LOL'ed at your comment. It seems about right. It's also a reflection of itself.

Why do You think creating a browser in a 'safe' language would prevent security bugs from occurring?

It won't prevent all, but may prevent many.

Quote: If you have a very large (millions of lines of code) codebase, written in a memory-unsafe programming language (such as C or C++), you can expect at least 65% of your security vulnerabilities to be caused by memory unsafety.

The quote is from https://alexgaynor.net/2020/may/27/science-on-memory-unsafet...

It might prevent 70%[0] or so of exploitable bugs. Which is better than C++, which, after almost 40 years of effort, continues to be the source of most of that 70% in the first place.

0 - https://www.zdnet.com/article/microsoft-70-percent-of-all-se...

That's really great that you are quoting a generalized stat, but we are taking about browsers here, which are inherently runtimes allowing anyone to run anycode they want.

Just saying 'c++ bad' isn't really a serious analysis...

Saying that super complex C++ code bases are less exposed to memory safety issues is what is not a serious argument. Have a nice day.

Edit: here's a report from the Chromium team, if that makes you feel any better. It reports similar figures. https://www.chromium.org/Home/chromium-security/memory-safet...

Are you suggesting that decreasing surface area is not worth it?

I am saying that 'crossing someone out' just because they decided not to opt into the tradeoff to decrease that surface is overreacting.

If you think it's worth it, feel free to go and do it. But that's not a reason to astroturf for Rust on a thread about Ladybird.

Because 70% of the security holes in C code bases tend to be things that never happen in a memory-safe language.

source?

From the chromium website

> Around 70% of our high severity security bugs are memory unsafety problems (that is, mistakes with C/C++ pointers). Half of those are use-after-free bugs.

https://www.chromium.org/Home/chromium-security/memory-safet...

I mean, it doesn't get any more clearer that the Chromium team is already considering integrating Rust in Chromium to mitigate these sort of memory corruption vulnerabilities found in C and C++ and that is the main motivation here. [0]

Given that Ladybird is yet another C/C++ browser, it will be riddled with these vulnerabilities waiting to be exploited.

I don't think many here are thinking about the costs involved with security fixes and maintenance, at it is in the hundreds of millions of dollars a year just to keep up with Firefox alone. That is the minimum requirement for any serious competing browser.

There's a clear reason why Chrome, Safari and Firefox need hundreds of millions of dollars of developer maintenance a year.

[0] https://chromium.googlesource.com/chromium/src/+/refs/heads/...

Nice Strawman. My Argument is that it takes likely 10s of millions of dollars to secure a large CPP Browser from the metric shitload of memory safety issues the average C/CPP Programmer leaves behind. If you want to compete with Google, either you use a memory safe language and spend your time securing the other millions of security issues that arise from writing a web browser or you waste time and money to build something no one in their right mind should be using for anything given the state of the modern web.

Chromium is written in C++...

His point is correct. To pretend to make Cpp safe, you need google sized budget (if not in engineering, in marketing, but oh well)

ELI5: as far as I know, „safe“ programming languages are able to avoid memory related issues that are/were prevalent in C(++). Is this actually the class of bug that causes most security related issues in the browser?

Yes, memory safety issues make up a large majority of high-severity security bugs in Chromium. See https://www.chromium.org/Home/chromium-security/memory-safet...

How do you carry water in a sieve? That's the personnel cost analogy to have safe cpp. His point is not about replacing the sieve, it is about having the money to still have some water across even when using sieves.

And Google Spend roughly 100 Times more money on making it safe than ladybird will get in funding in total.

Nonsense, all 4 of the most widely used browsers were written in C/C++. A browser is basically an OS at this point, and these issues can be addressed in the design, just like in any OS kernel, which are also almost all written in C/C++.

I think the demise of Firefox has been greatly exaggerated... I'm still using it as my main browser, and the user experience and performance are excellent.

Reply on news.ycombinator.com