3 ms·
This is magical thinking, switching language won't solve the problem. Complexity is the problem, hard-to-read code is the problem, over-engineering is the prob
by stephc_int13 6y ago
This is magical thinking, switching language won't solve the problem.
Complexity is the problem, hard-to-read code is the problem, over-engineering is the problem.
- asperous 6y agoI think its systems thinking. With 40,000 developers, Microsoft is too large to fix the problems you mentioned. They need a broad stroke change to reduce vulnerabilities. The article says they've tried many things and believe moving to rust might work.
- stephc_int13 6y agoYeah, MS clearly has serious maintenance issues with their codebase, for obvious reasons. They tried many things in the past, including C# and .NET, I'm not sure if that really helped or made things worse.
- muldvarp 6y ago> This is magical thinking, switching language won't solve the problem. It won't solve the problem, but it can be an important step towards a solution. > Complexity is the problem, hard-to-read code is the problem, over-engineering is the problem. I actually disagree. Many memory safety issues are not the result of hard-to-read or over-engineered code, but are instead due to careless memory management. Safe Rust code (no matter how hard-to-read or over-engineered) can guarantee memory safety.
- stephc_int13 6y agoIf it was that easy to fix, this problem would have been fixed a long time ago, by many other "safe" languages. My bet is that Rust won't help nearly as much as people tend to think.
- muldvarp 6y ago> If it was that easy to fix, this problem would have been fixed a long time ago, by many other "safe" languages. There are many other memory safe languages. Those usually have a GC though. > My bet is that Rust won't help nearly as much as people tend to think. Why not? Memory safety issues are common and safe Rust code eliminates them.