Posts by libove

Acer confirms server intrusion after miscreant offers 160GB cache of stolen files

Maybe if Acer didn't refuse to release technical manuals to, you know, mere end-user customers (the way that Dell, Lenovo, and basically all other major manufacturers do), there wouldn't have been such an incentive to hack a "technical documents" server. Not that I'm condoning hacking, of course, but Acer is a little bit evil, so I don't feel particularly bad that this data was stolen, and I do hope that technical repair manuals for Acer products end up broadly released as a result.

Internet overseer continues wall-punching legal campaign

FUD argument is wrong. Action Fraud has some points. Read on ...

The article's argument that ICANN's argument amounts to FUD - that criminality will increase in the absence of good, accessible WHOIS data - is incorrect.

Action Fraud's comment that it takes a consumer both sufficiently aggrieved as well as capable to even open a police case, followed then by an adequately resourced and motivated police force, in the same jurisdiction, to get WHOIS data to begin to track down malfeasants, is absolutely correct.

A large amount of small fraud is gotten away with all the time today due to the ease with which wholesale (fake) copies of e-commerce websites are set up and then SEO'd into high-end positions in search results that people looking for an article to purchase may find themself on such a fake site. Absent readily accessible WHOIS data, they (were they so skilled and inclined to check first) could not get the information needed to take into account in judging whether the site is likely legitimate, or (in the more common case of the consumer who lacks the wherewithal to do such analysis themself) other site rating services would not have enough information to form an opinion, because the sites churn quickly.

The result is that many consumer just take the risk of buying from the unknown site, and are defrauded (which, under credit card fraud rules in the US, is low risk to the consumer, but for the rest of us in the world, where we have far less legal protection against credit card fraud, is largely at our own risk), or, the more cautious consumer simply doesn't buy (which hurts the legitimate small business which forever whatever reason, either of their own potentially misguided commercial choice, or because of the defaults of their possibly-hidden-behind-two-service-provider-layers-Registrar has their registration data private).

There are US-EU dick-size contests going on here. But there also truly is a problem with the current towards-privacy-everywhere argument in WHOIS data.

My proposal is that WHOIS data should be made accessible in most cases (call this two different levels of WHOIS Privacy) to anyone who files a legally sworn statement as to the reason and validity of their request, with no need for police or court process, to be granted the requested access; if that access turned out to be fraudulent then that person will have committed an easily prosecutable crime. The second (higher) level of WHOIS Privacy would be accessible only upon presentation of a court or police finding that the registrant is under unusual threat, and would then enjoy the level of privacy that current one-size-doesn't-fit-all WHOIS Privacy provides.

Google's 'encrypted-by-default' Android is NOT encrypting by default

Encryption in Software Slow on modern devices?

I find it hard to believe that having to perform disk (on-device Flash-based storage) encryption in software (on the main CPU) would make a modern mobile device noticeably slow. The article quotes Google as saying it ... but can someone give specifics/details?

For example, it is very common today to buy moderately priced mobile devices with quad core 1.2GHz+ CPUs. Given the relatively slow performance of the Flash storage on mobile devices (which limits the amount of data which the CPU core would need to encrypt every unit of time), and the efficiency of modern software encryption algorithms, is the performance impact of software-based encryption really noticeable to the human user?