3 ms·
That's about the same as saying that since plenty of people know how to pick locks don't bother locking your door. Did you read the second half of the sentence
by cperciva 18y ago
That's about the same as saying that since plenty of people know how to pick locks don't bother locking your door.
Did you read the second half of the sentence I wrote? You know, where I said "unless they have a reason for needing it"?
I have a reason for locking the front door whenever I leave home, and I do so. Putting a lock onto the back gate would be utterly pointless -- that gate is in the middle of a fence which would be easy for anyone to climb over -- so I don't do that.
Actually it's worse: there aren't any known ssl issues, so you're leaving your door unlocked just in case someone learns how to pick it in the future.
Replace "door" with "back gate in the middle of a pointless fence", and you'd almost be correct. But security vulnerabilities in SSL stacks are not limited to breaking SSL; a buffer overflow could give an attacker far greater access. A closer analogy to blindly using SSL everywhere would involve a special type of lock which, if picked, magically unlocks everything else on the property, including things like wall safes which are much harder to pick.
- ars 18y agoYes, I read the second half and I still didn't see how adding SSL could make things worse. Your magic lock example makes sense, except for the fact that your analogy is broken: ssl doesn't actually work that way. I suppose I could see that adding even a single line of code that you don't need increases your risk of a hack. But that's quite extreme, in reality it's not anywhere near that bad. I just don't agree with your risk/reward ratio of not using ssl. It's: possibility of a bug in ssl that allows remote access to a machine, and is not expolitable without ssl (you have to agree that that a pretty remote possibility), vs the known problems of not having ssl especially in a wifi environment + the possibility someone will not understand you and not enable it even when he should.
- cperciva 18y agopossibility of a bug in ssl that allows remote access to a machine, and is not expolitable without ssl (you have to agree that that a pretty remote possibility) I don't think that's remote at all. The version of OpenSSL on my system is over 200 kLOC. Do you really think that it's a "pretty remote possibility" that one of those 200k lines contains an exploitable buffer overflow?
- jrockway 18y agoIt's more likely that the application you're "protecting" by not using SSL has an exploitable security bug. OpenSSL is a lot of code, but it also has a lot of users who have uncovered subtle-yet-horrible security problems in the past. So it is somewhat trustworthy. Your app has only been seen by you and perhaps a small team. My money says that if someone breaks your system, it's going to be through that vector, not the SSL stack.
- cperciva 18y agoOpenSSL is a lot of code, but it also has a lot of users who have uncovered subtle-yet-horrible security problems in the past. So it is somewhat trustworthy. OpenSSL has a lot of users. But many eyes only makes bugs shallow if the eyes are open -- and 99.999% of people who use OpenSSL have never looked at the code. One of the dirty secrets of open source is that there are a lot of projects which almost everybody uses by very few people actually develop. Your app has only been seen by you and perhaps a small team. My money says that if someone breaks your system, it's going to be through that vector, not the SSL stack. Even if you believe that OpenSSL has had lots of people looking at the code, which is more likely to have bugs -- 20k LOC or 200k LOC?
- thwarted 18y agoThe claim that many eyes makes bugs shallow is based on the number of eyes looking at the code, not the users. It's not a dirty secret that there are relatively few developers for an open source project vs the number of users. This is well known, and documented in CREDITS and Changelog files. The "low" number of people working on open source is often trotted out during a FUD campaign, as if to imply that large software companies have millions of "experts" looking at and developing the code. Give me a handful of guys who _publicly_ worked on a set of code and know it inside and out any day, over a hundred thousand developers who hide behind some corporate name and spin department and who only ever touched one part of it. The chance of code having bugs is more likely directly correlated to the complexity of the code, not its size. The change of finding the bugs in the code to fix them (not finding that they exist) is more likely directly correlated to the amount of code.