Hacker Newsnew | comments | show | ask | jobs | submit | adsr's comments login

You could keep pointers to both the head and tail in the list object though. Then adding and removing stuff would be O(1) for both ends.

-----


Yep, and that used to be idiomatic, years and years ago when people still wrote their own linked lists. I'd be surprised if it isn't still the standard behavior for that data structure.

-----


Something is odd with that site. If I enter https://www.freakattack.com it reports Safari 5.1.10 as safe from the attack. If I simply enter https://freakattack.com it reports it as vulnerable.

-----


I doubt that would work well in practice. The reason is that timing is extremely important for this to work, if you're even one sample off you'll here a faint sound, and more than a few then it's quite obvious. So if the sound is generated by a separate OS process, you'll never know exactly when the sound is generated and you will not even have control over exactly when your own sound is played back in relation to the other process, think sub ms accuracy.

-----


But it's only commercial in the sense that it permits commercial use, but it also explicitly leave the creator out of any compensation beyond attribution.

-----


I think it's understood that what is done in public is public, however that is not the same as having the same information recorded, analyzed with facial recognition and machine learning technology and stored for the future.

-----


Can't this just be turned off with the Spotlight setting in system preferences though? For browsers it seems to be the same for all that uses the unified search field, it was last time a checked Chrome with tcpdump. I personally preferred to have the URL field separate from the search field for that reason.

-----


> Can't this just be turned off with the Spotlight setting in system preferences though?

Yes. Although note that it will not disable Safari's "spotlight suggestions" which have to be disabled separately via Safari's own preferences.

-----


I don't think it's that clear cut, there are some inherent challenges related to privacy and internet, a persons past that may influence future opportunities in a way that's not really fair. In fact, Eric Schmidt has expressed that same sentiment publicly.

http://www.telegraph.co.uk/technology/google/7951269/Young-w...

-----


It's a bit more, the bool type can only hold two values, 0 and 1. (6.3.1.2 Boolean type #1) "When any scalar value is converted to _Bool, the result is 0 if the value compares equal to 0; otherwise, the result is 1."

-----


> I give Linus a pass. Given the success of the project and the size of the team, I would rather have a foul mouthed Linus than perhaps no kernel and no Linux.

This appears to be a false dilemma. I don't think the success of the kernel and Linux depends on Linus' foul mouth. And I think the point was that he, as a role model may influence and inspire others to adopt the same behavior.

-----


It sounds more like they are using a Cortex-A5 to gain access to TrustZone with an existing x86 core.

-----


And it sounds like Apple is using a separate unspecified ARM processor (probably a Cortex-A5 since that's the cheapest possible one) to gain access to an existing A7 or A8 core.

-----


In Apple's case, they use the ARM ISA but implement their own micro architecture and from vvhn's comment seems to also use a co-processor specifically for the secure enclave. But the link above on TrustZone hardware architecture mentions that this isn't a requirement.

"TrustZone enables a single physical processor core to execute code safely and efficiently from both the Normal world and the Secure world. This removes the need for a dedicated security processor core, saving silicon area and power, and allowing high performance security software to run alongside the Normal world operating environment."

I guess since Apple use the ARM ISA, it's still binary compatible with ARM but with a different implementation. AMD uses an x86/ARM hybrid where the ARM part is an off the shelf Cortex-A5 which already contain TrustZone.

-----


I highly doubt they use their own micro architecture. It'd be a lot cheaper to license Cortex-A5. Using their own micro architecture for the main processor gives them a huge competitive advantage. For the security co-processor, COTS would work fine.

-----

More

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: