[1] https://confluence.atlassian.com/display/BITBUCKET/Bitbucket...
706 karma · joined April 2, 2014
Place where I write about reliability, programming, and other things: https://www.tonylykke.com/
[ my public key: https://keybase.io/roganartu; my proof: https://keybase.io/roganartu/sigs/LyULghKURMpLPigIRAq_rQZWO_NjjkZESA4Mr_TKXYI ]
[1] https://confluence.atlassian.com/display/BITBUCKET/Bitbucket...
As for relative price typically one would be looking at around $300-600 per year, depending on age, driving history, type of car etc. Granted with our universal health care I suspect that our average out-of-pocket medical expenses for injuries from car accidents are lower than the US.
This is a bad analogy (unless you mean to say that this second key would be identical for every door, in which case.. sort of).
A better analogy would be that everyone was forced to install a second entry method to their house that only the government knew how to operate.
The inherent problem with this is that as soon as someone else figures out how to operate it, everything with this entry method installed would be accessible to them until it changes or is fixed.
The central server has access to the entire file, and hence can compute the hash of any arbitrary chunk. Challenges/verifications don't have to happen all that often ([1] indicates they are looking at once a month) so creating a unique challenge for each user shouldn't be too compute intensive.
For each known user:
- Central server chooses random chunk of each file it wishes to verify the user still has. This could be any length from as long as the hash function to the whole file size and could be offset any number of bits.
- Client is asked to provide the hash of the chunk using the file stored locally.
Precomputing hashes for all possible chunk permutations would take up substantially more space than simply storing the file in the first place. A bad actor would need to store the hash for all the possible chunk lengths starting from every possible start location in the file which is in the order of O(n^2) where n is the stored file size (500GB in this case). For reference that would be about one third of the entire 20PB archive for a single 500GB chunk if using a 256 bit hash function.
https://filippo.io/Heartbleed/#tip4commit.com
Don't sign in or register or you may have more to worry about than undesired tax liabilities.
EDIT: It's fixed now.
That being said, I like the idea in general I'm just skeptical of protocols/ideas that rely on things that don't yet exist (unless they go into detail on how they plan to implement them of course).
You only have to look at password handling procedures to see recent examples of things once thought hacker-proof that are no longer. MD5 used to be widely used for hashing passwords but it is now considered cryptographically weak and there are many more examples like it throughout the industry. Most (all?) of these hacker-proof like things tend to be based on computational infeasibility or perfect implementation of certain features. In this case, the protocol seems to rely heavily on two things:
- self-destructing database records - Self destructing data is a hard problem that is more likely to be solved using a time-based destruction rather than a read-based destruction (how can you ensure with 100% atomicity that the data will be 100% deleted when it is read?). A good example of a time-based self-destruction is Vanish [1].
- dynamic procedure that is not accessible from memory - The CPU cannot communicate directly with anything but memory/internal caches [2] so the only way I can think that you'd solve this is by having a custom piece of hardware that does all the computation internally and only returns the result. This is also not an easy problem as it introduces hardware level complexities.
I understand that the author may have just been trying to make it sound more interesting but there are better words that don't leave such a bad taste in the readers mouth ("Secure" is a very commonly used one for example). This is especially important when trying to be taken seriously by the security community who are the most likely to take an irrational dislike to something claiming to be hacker-proof.
[1] http://vanish.cs.washington.edu/pubs/usenixsec09-geambasu.pd...
[2] http://wpcontent.answcdn.com/wikipedia/commons/thumb/b/bd/Mo...
I've used the likes of Palleton [1] and Kubler [2] before but they seem to be geared towards graphic designers and people who actually understand terms like adjacent, triad and compliment in the context of colours.
Hell, I used to work at a pub that had it's staff carry over $100k to the bank every Monday morning, though that amount is arguably grossly irresponsible both from a business perspective and with regards to staff safety.
It has a lot to do with ability to see where you are going and hence predict/be aware of the vehicles movements. To put the extent of this in perspective, I am noticibly more likely to feel motion sick sitting in the back seat compared to sitting in the front seat. This effect is reduced by sitting in the middle, where there are no seats/people obscuring your view. Additionally sitting in any seat except the drivers seat while travelling on windy mountain roads is a surefire way to bring up lunch. Reading while in a car is completely out of the question.
Comfort should be a consideration of any type of transport and the thought of facing backwards in a car makes me uncomfortable. While this shouldn't outweigh safety, it certainly shouldn't be ignored either.