3 ms·
Languages take off broadly because there's something compelling about them (and it isn't necessarily a technical reason). One of most compelling reasons for ado
by criddell 5y ago
Languages take off broadly because there's something compelling about them (and it isn't necessarily a technical reason). One of most compelling reasons for adopting Rust is memory safety and that may not be terribly compelling.
The comment about it being over a decade old was mostly that it wasn't some new thing that people are unsure about where it can be used. It's mature and has been successful in some niches (and keep in mind that niches can be large).
- Ar-Curunir 5y agoHow is memory safety not a compelling argument when we're literally in a thread about memory unsafety leading to security exploits?
- lelanthran 5y ago> How is memory safety not a compelling argument when we're literally in a thread about memory unsafety leading to security exploits? The ratio of { memory-safety-bugs-in-code : bugs-in-code } is too small in many cases to warrant redoing the entire project. My last C (and C++) role: over the course of 3 years, a large C++ project had over 1000 bug reports closed, of which one turned out to be a memory safety issue that would have been prevented in Rust. My previous C position had a similar rate. It's hard to convince the company to throw a 5-person team at a rewrite project for 3 years just to avoid the bug that they got in the previous 3 years, in the process incurring new bugs. If you're the owner of a company, you'd dismiss anyone who tells you that you need to spend a few million redoing work already completed, with additional risk that the redone work will have new errors that were already fixed in the existing work.
- kobebrookskC3 5y agoyour code probably isn't worth attacking
- lelanthran 5y ago> your code probably isn't worth attacking Maybe, maybe not. What bar do you set for "this product is worth attacking"? Because the product owners and the product users definitely thought that their product was valuable[1]. [1]The products in question were 1) Payment acquirer and processor, and 2) Munitions control software.
- kobebrookskC3 5y agoa product can be valuable without being worth attacking, like something that only runs on trusted inputs. the bar for software "worth attacking" for me is that there are people who are paid mainly to find exploits on it, and not just incidentally as a product of development.
- lelanthran 5y ago> a product can be valuable without being worth attacking, like something that only runs on trusted inputs. the bar for software "worth attacking" for me is that there are people who are paid to find exploits on it. I'm in agreement, but then we get back to the fact that memory safety may not be a compelling argument to use a new language. After all, most products aren't "worth attacking" until they are large and successful enough, thus reinforcing the decision not to rewrite a product in a new language. This means that memory safety alone is not a compelling argument for switching languages.
- kobebrookskC3 5y agowhy is the "payment processing" software written in c/c++ and not something like java in the first place? i would imagine not needing to care too much about memory would speed up development velocity. was it before java became popular? i can understand the munitions control software possibly running on a constrained device and not needing very sophisticated memory management (cue the old joke about the "ultimate in garbage collection").
- 5y ago
- Ar-Curunir 5y agoBut the “vulnerability value” of memory-safety bugs is much higher: it’s almost always a security exploit. Also, you don’t have to rewrite your code base immediately, or at all: just build new parts in Rust or another compatible memory-safe language.
- kobebrookskC3 5y agoidk, many large companies report about 70% of their security issues are due to memory unsafety. reducing your security bugs by a factor of 3 sounds pretty compelling to me...
- criddell 5y agoYeah, it's weird. I really don't know why Rust isn't more popular. Edit: I just read another comment you made about somebody's program not being worth attacking. I think you are on to something with that argument. Most software isn't worth attacking.