@JeffFerland plus, they can be used to confuse someone else, causing them to run slower, absolving you of the need to actually run faster than the lion.
I mean, say you were writing professional grade software that would involve sensitive client information. (Take this in the context of me being an amateur programmer.)
Would you use hlib and hmac? Are they good enough to secure data? Or would you write something fancier by hand?
Edit: In contex...
"Addressing" that law, strictly speaking, means suing God for creating mankind with limited brains. There is little to do here, except to be aware of the difficulty of assessing the security of a security system: namely, that you cannot assess the security of your own creation. One way to say it ...
I just stumbled upon this answer from a year back. That first sentence is gold.
I got to watch the process of designing some of the bits and pieces of DTN. It was pretty interesting to see the thought process that goes into trying to make it, you know, work.
Since it is, largely, a store and forward system I think they did take some lessons from our decades of experience with email.
By the time I started my grad program here the networks research group had already been working on it for a bit, and while I was friends with several of the guys I never worked on the projects.
Still though, based on overhearing things and 2 beer conversations I think the best description I can come up with is messy.
I tried to design a secure email encryption system and couldn't come up with a solution I really liked, so I certainly see how even more difficult offline messaging systems get messy.
@CodesInChaos It is an interesting one, precisely because while the TLS standard says how to renegotiate, it never says what renegotiation actually achieves.
The security issue comes from servers which assumed that it achieved what they wanted it to do (and it did not).
But since nobody bothered to write down what they want except in the most vague terms ("we want it to be secure")...
@RoryAlsop This reminds me of a joke: how do you recognize a polar bear from a grizzly ?
You run towards the nearest tree. If the bear catches you before you reach the tree, that's a polar bear (they are very fast runners). If the bear climbs the tree, it's a grizzly.
Bad tags are created frequently, especially on beta sites where the barrier to tag creation is relatively low. There are a few mechanisms already existing that are meant to help remove bad or useless tags:
The list of new tags in the 10k tools at /tools
An automatic script that deletes tags tha...
I mean, say you were writing professional grade software that would involve sensitive client information. (Take this in the context of me being an amateur programmer.)
Would you use hlib and hmac? Are they good enough to secure data? Or would you write something fancier by hand?
Edit: In contex...
