Become a fan of Slashdot on Facebook


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 internet speed test! ×

Comment Re:WTF!!! (Score 5, Interesting) 513

And I'll bet there's *tons* of aspects of the job that can be done from home, without violating any contractual obligations.
Do the sensitive stuff 8-5 and do the rest at home.

As someone who's been penalized for trying to use their (former) company's work-life balance programs I am going to assume the little guy's side on this until evidence to the contrary is provided.

Comment Re:Changing jobs increases wealth (Score 1) 148

"institutional memory"

Massively underrated IME. Luckily we have a few lifers here but i've lost count of the number of times we have had no idea why something is the way it is or some subtle wrinkle that's non obvious in how something gets processed. People that have years and years of experience often fill in gaps that would otherwise cause big problems.

Comment Re:What does this matter. (Score 1) 60

T-Mo branded accounts are different then.
I get throttled only near busy cells (based on observation such as: at the mall while lots of other people are there).
Near my house I don't see any throttling at all.

While I'm being throttled my phone is still reporting LTE, but I am seeing speeds drop to about 40-70% of normal.

Comment Re:Here is where your logic failed you (Score 1) 447

The ME is a signed binary.
The chipset has key material fused in at manufacture time.
The ROM on the chipset won't load a binary into the ME hardware that isn't signed in such a way as to match what is in the secure fuses.

*if* a chipset sku series is compromised then yes, you can sign and run external FW on that series, but if it is unsigned it won't load on the hardware.

The driver interface to the ME is the "HECI" or "Intel Manageability Engine Interface" depending on version.
It uses a defined interface to pass messages to the FW from the host side.
Anyone can write a driver for this... Intel published the spec and linux source. The firmware treats it as a hostile interface.

I get the "he's a shill" BS from the ACs, but seriously, I worked on this for 6 years, and while I would certainly load a min sku on my machine if I was concerned about access by a state actor, I personally have no worries about my data on an ME enabled machine.
I no longer work for Intel, I have less than 100 shares of stock left, no incentive to help or hurt them.
If you're truly paranoid about the data access then just use a PCI/e NIC and don't use the on-board LAN. Problem solved.

Comment Re:Intel CPU backdoors (Score 1) 447

Well, the way I see it (and from my experience with the firmware) if you *don't* want the active parts of the Intel ME, then just put a min sku (or at least the 1.5 meg version) on your platform. That kills the majority of what people are up in arms about.
That much is easy to do, low risk, and accomplishes what people say they want: disable the active ME content.

I get that there are some that feel that it should *all* be taken off, but seriously, your platform won't be stable without it.
Modern systems run on firmware. The ME aggregates a ton of what used to be little firmwaree pieces and puts them in a common runtime. It's not rocket science.

Things that used to be stand alone (and are all needed for optimum performance, or even boot in some cases):
CPU uCode patch
ACPI support (S3, S4)
and I'm sure I'm forgetting plenty of other stuff that I just didn't even think about on a day to day.

It's the platform firmware version of systemd on linux. (and there are plenty of people that hate it too, so.... )

Slashdot Top Deals

You mean you didn't *know* she was off making lots of little phone companies?