Posts by theblackhand 925 publicly visible posts • joined 1 Oct 2009
The problem for Intel (and Microsoft as the two are tied together in this) is price.
The hardware designers producing Windows tablets have to spend more on the licence (US$100) and CPU ($150+) than the competition can deliver a complete product. Moving to a Linux/BSD-based OS would help although you lose some application appeal, and you still end up with a high-performance, high-cost processor against a $10-$30 ARM SoC.
The hardware manufacturers then try to produce a competitive system and its a undesirable $1000 piece of tosh....
OK, I'll debate it.... North America moves the second largest amount of data behind AP and almost 1/3 of the total traffic which looks OK based on traffic volume/population.
http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/VNI_Hyperconnectivity_WP.html
The reason for poor consumer Internet speeds is lack of competition or any sensible alternative (i.e. BT OpenReach in the UK to allow multiple competitors on a common infrastructure) in a large country where the technical limitations (i.e. distance) of consumer technologies such as xDSL or 3/4G make the provision of high speed Internet challenging.
Address the competition issue so that consumers have real choice (not just two shades of the same expensive crap) and things will improve. Please don't tie it in with the net neutrality debate - net neutrality is hard (as a network guy you want to be able to classify traffic to provides different levels of service) while the telco competition issue is clear.
"Hi Bill, don't you have some philantrophic project to manage, you old scumbag?"
Bill's given up the philanthropy game after GQ awarded Tony Blair the 'philanthropist of the year' award.
I think Bill's waiting for the current Middle East "peace" to end so he can try and repeat Tony's success there next.
Limiting login attempts on a cloud service becomes a denial-of-service path - don't like someone? Attempt to log onto to their cloud account with a guessed password. If it works, you get access, if not repeat X times and lock out the account.
On top of that, how do you unlock the account? You probably can't verify the account holder is who they say they are with any great certainty (i.e. e-mail may have been breached, phone may have been stolen, a lot of the default questions in password recovery Q+A's can be be answered from Internet searches if filled in literally (i.e. mothers maiden name, schools, addresses).
As for using client certificates, I would have thought that an app that ties in your cloud sign up (for mobile devices) or licensing for Windows would be fairly straight-forward and maybe this already happens. The problem that I see is that providing an easy way to add more devices to an account or swapping between an old and new device probably voids any benefit from this approach as it would allow either a way of moving certificates or adding new certificates with minimal fuss.
It doesn't appear to have been a iCloud security breach - just good old fashioned social engineering.
There are some interesting points made in the following article s well as steps on how to improve your phone security.
https://www.nikcub.com/posts/notes-on-the-celebrity-data-theft/
It's a little unfair on the minister for the affected department to foot the bill - particularly with many senior departmental civil servants making more than the ministers these days.
As there is a general lack of responsibility, I would propose a Gladiator-style battle between the responsible committee with the members battling it out to pay nothing (lose first round, pay £8k, 2nd round £4k, semi's £2k, losing finalist £1k). Create a TV show with whatever commentators/presenters are available with all profit going to improve security practices in said department.
I don't believe this will address the underlying security culture in many of these environments, but it would make better TV than "Britains Got Dancing on Ice" or what ever the tripe is called...
I suspect it will be Cisco Catalyst 6500/7600 switches with Supervisor 720 engines.
There are 2 models - one supporting up to 256K IPv4 routes and one supporting up to 1M IPv4 routes.
But....
The default configuration of the devices allocates 512K IPv4 routes and 256K IPv6 routes.
Other devices with insufficient memory would also be affected, but I suspect the deployment of Cisco 6500's in telco networks is very common and hence are the key to this issue being so widespread.
The hope is that exposing the devices to a variety of conditions globally will result in a mutant self-powered machine that can then be used for military purposes.
Scientists have built-in controls to prevent these devices becoming self aware. Or at least self aware before April 11, 2019 (there was a typo in the widely known dates when machines were supposed to become aware....)
Regarding the opinions, in the technology evaluations I have read from Gartner, they are fairly generic and can usually be backed up fairly easily. Where technology companies miss out is in publishing dates of reports - "missing major features" can mean your new product release shipped after the report date. "Struggling to deal with technical demands and rising expectations" can mean that you are growing rapidly and engineering/technical/support resources are stretched.
I don't know enough about NetScout and it's competitors to know if these apply.
Are they used? Yes
At a minimum they are used by large companies for creating short lists of vendors to work with and I have worked with one company that choose a global telecoms provider based on a dubious Gartner report (i.e. the answer wasn't one of the two global providers that usually pop up for a global solution and the vendor had almost no presence at all in one of our key geographic regions when they were chosen.... i.e. the regional enterprise account managers dog accounted for 33% of the regional head count).
For a more accurate sentence, try this:
"Salmond insisted that Scottish thieving bastards would be better off ruling Scotland than being subject to the whims of thieving bastards in Westminster whom Scots themselves hadn’t voted for."
If the answer is politicians (the current mob of self-serving, morally dubious sons and daughters of unwed mothers), then maybe people aren't asking the right question...
People will buy either by brand name because they have previously had good experiences with the vendors product, by recommendation, by coolest name or by random. The model numbers are only relevant to the 0.0001% of customers that discover model X has more CPU/RAM/flashing lights that allows it to run third-party firmware and make it a decent product.
Until we have rockets that can be flown into the sun for a reasonable cost, we have to give marketing people jobs where they are unlikely to do to much harm - hence stupid product names and incomprehensible model numbers.
More than one days battery life is ridiculous - how are you supposed to remember to connect you phone to a charger if it isn't needed every day?
I remember the good old days where you had to run to the office in the morning to make sure you reached the charger before your battery died. These big phone batteries will just hasten the obesity crisis.
I believe that the figure is 1.25 - it's just not comparable to a conventional data centre PUE.
Google/FB data centres PUE will generally include all power usage at the facility including all usage due to power transmission/moving coolant to the appropriate locations.
With the pod, I suspect the power usage is just for the pod - any power used to get services to the pod are excluded from PUE calculations as they are outside the vendors control.
Comparing Apples, Oranges, Bananas and Lemons...
This type of secrecy makes me suspect that Mr Blair is hiding WMD's.
I think we should send in a UN team and if Mr Blair fails to comply, get the UN Security Council to vote on action.
Tempting though it is to recommend a tactical strike, I think freezing all assets should be enough to make him squeal....
I don't think there was any application of common sense involved - Gemalto patented a method where all functions were on a single micro controller and Google/Samsung/HTC/Motorola seperate the storage and processing functions.
Common sense would have been for the judge to tell Gemalto that there patents weren't worth the paper they were printed on due to prior art and the broad scope of the patents.
Point 1 is easy to prove, point 2 would take a lot of time and expensive lawyers.
Dear Mr Dabbs
I would like to connect you.
Based on your description of being "sweaty and shouty, slurring my jokes as I swirled my double JDs and leered into my conference colleagues’ faces", I think you would provide the perfect alibi for my public behaviour.
In return, I would be able to provide the same for you.
I look forward to your acceptance
I would expect soldered memory to be significantly more reliable than a mechanical connection as any faults should be detected during manufacturing.
It won't be to every bodies taste, but it's a significant saving for someone wanting a cheaper Mac if they are unlikely to ever open the case.
Given the choice between blowing £600+ on a TV that is marginally better than the one I have to be able to see more detail of my team losing or blowing it (or a small portion anyway - it doesn't take much these days...) on alcohol to dull the pain of watching my team lose, I know which I'll choose...
IANAL but the monitoring appears to be a legal requirement for operating in some (all?) of these countries.
While end-to-end encryption sounds nice, I can understand Vodafone's reluctance to ignore the law in case they were shut down. Or is that just my pragmatic streak?
The e-mail content is encrypted, but the SMTP header will still provide useful metadata about who you are talking to and how frequently.
Add that most of the people will probably use the same password for their encryption keys as for Google or users won't properly secure their tinfoil hats allowing state agencies to read their minds and the encryption doesn't really address the privacy concerns raised about the US based clouds.
So any RADIUS servers using a flawed released of OpenSSL are vulnerable, assuming they haven't already been updated, for TLS-based EAP authentication such as EAP-TLS/EAP-TTLS/EAP-PEAP.
I would imagine that the affected install base for this would be tiny (an OpenSSL 1.x release that hadn't been patched) with most installs either using OpenSSL 0.9.x or earlier for old systems or a patched release for current systems.
WEP/WPA/WPA2 pre-shared key aren't affected - they aren't that secure, but not because of Heartbleed.
All model's will be re-branded but will continue to use the same OS and offer the same features.
Marketing's magic wand will be used to sprinkle special dust over everything to make it better, If you are really lucky, the units will now support larger drives that were previously usable not not on the "validated component" list.
Disclaimer: this is pessimism - any resemblance to reality is coincidental.
Wasn't this an attempt to compete with Sky (and to a lesser extent Virgin)? Sky moving into triple play meant more revenue which meant more premium content moving to the platform.
Mid-2000's it wasn't clear where the competition would come from and I think the BBC saw an opportunity to compete and move from being a broadcaster to a broadcaster/content provider.
As we approach 10 years later, there is competition (Virgin and BT), I don't see anyway the BBC would be allowed to enter the ISP/telecom's space to compete on 3G and one of their partners (BT) looks like being a serious contender in that space.
So what are the BBC's plan's now?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
