FlyingAvatar

It's improved, though still a root exploit in the kernel will almost certainly make possible root access to the host. For a lot of situations, this is an acceptable risk.

The real win in the present for containers is simplifying deployment and improving segmentation of services that already might run on the same machine.

Sending signals via electrons through wires is fundamentally different then sending photons through fibers.

To visualize electrical signals, imagine a tube completely full of balls. When you push a new ball into one end, a different ball comes out the other. The effect of adding the ball is seen very quickly (i.e. a change in the field), but the speed the ball moves through the tube (i.e. how fast electrons move through the conductor) is quite different.

Not endorsing these guys since I have no idea what their availability is in practice, but isn't 3 nines a fairly common SLA? I believe this is what I get at Linode and Digital Ocean.

I've always preferred, option + left/right for fast moving with left/right for fine adjustment so I don't have to leave the keyboard.

The low-end Tempur Queen mattress from this page:

http://www.brookstone.com/tempur-pedic-tempur-simplicity-mat...

47 lbs / 22.22 cu. ft = 2.1 lbs

The weight listed is probably the shipping weight with packaging, so the actual weight of the foam is less. Seems like 8 lbs would be extremely dense.

The difficulty is climbing at an incredible rate because (a) lots of ASIC miners are coming on-line, and (b) a lot more people have started mining.

At some point very soon, mining Bitcoin will become unprofitable overall simply because an essentially fixed number of output Bitcoins are being divided across a larger and larger number of miners.

The only way to mine profitably will be figuring out how to mine both more efficiently and significantly more quickly than the general mining population.

Essentially if you are newly investing in Bitcoin now, you need to believe that (a) the mining solution that you're buying into will be at least an order of magnitude faster than the current generation products, or (b) Bitcoin will increase in value so much that even a small number of Bitcoins will be very valuable.

If this is just another Bitcoin ASIC, it's seems a very risky investment.

I hate the invention of "burstable" RAM on VPSes for this reason. Eventually your IO cache grows until it has most of your guaranteed RAM. Now every new process is playing a game of Russian roulette with your hypervisor to see if it can actually allocate into your "burstable" RAM what the kernel is reporting is freely available.

If you live in the US, this should serve as a warning that you need Long Term Disability insurance. If your employer doesn't offer it, get it yourself. Healthcare is still broken and ObamaCare doesn't fix this situation.

EEPROM -

Electronically Erasable Read Only Memory

It's re-programmable (i.e. by re-flashing it).

edit: I should add that motherboard manufacturers could prevent this type of attack by "locking" the BIOS for flashing unless it was explicitly unlocked by changing a setting in the BIOS menu (some have this already, I believe). The problem at the moment is that the BIOS is writable at all times, even when the OS is running. This makes BIOS updates easier (i.e. you can make a Windows application that can do so, for example), but the problem is that this allows ANY process with Admin access to alter the BIOS as well.

A system's BIOS image is usually written to an EEPROM chip, and they are often removable. So you can either just pop them out, or desolder them in the worst case.

Then you can use an external EEPROM reader that can dump the contents, but is not capable of running the code.

The EEPROM is storage only; it's contents are loaded by the PC at boot. So if it is removed, there is no processing that an occur internally than can mask the data inside of it.

This sounds really fishy.

> Ruiu said he arrived at the theory about badBIOS's high-frequency networking capability after observing encrypted data packets being sent to and from an infected laptop that had no obvious network connection with—but was in close proximity to—another badBIOS-infected computer. The packets were transmitted even when the laptop had its Wi-Fi and Bluetooth cards removed.

Observed how? If not a standard interface (i.e. disk, network, etc), using what? How could he know they were encrypted unless he intercepted a payload?

> With the speakers and mic intact, Ruiu said, the isolated computer seemed to be using the high-frequency connection to maintain the integrity of the badBIOS infection as he worked to dismantle software components the malware relied on.

What would be the point of this communication? It's BIOS would already need to be infected in order to be able to communicate via sound. The situation I can see this being useful is using another infected machine's Internet connection to download an OS specific payload, which makes some sense.

Honestly, I am fine with any IPS display.

I also have taken the somewhat crazy looking step of getting some "prescription" (I have 20/20 vision normally.) glasses made with some light magnification +0.5 to +1.0 and are tinted 40% yellow. That way you don't have to mess with your monitor's color calibration, and it will filter the blues from any outside lighting as well. Also, it's pretty cheap, as you can order them for <$40 from most online eyewear places.

GH/s per watt is everything. I'm pretty sure you're looking at the Sapphire USB miners. These are almost certainly not worth it. If you got 38 of these for free and had free power and started mining tomorrow, you would have perhaps $1500 worth of BTC a year from now.

Assuming you're buying and powering them yourself, you're paying around $1000 and $200 bucks in electricity. It's a fairly risky investment that will likely net you at most $300 on a $1200 outlay.

To summarize your purchase: 132 MH/s/watt for ~$1000

If you want to make a better risky investment, you might look into pre-ordering into one of the newer process ASICs, like Butterfly Labs' Monarch:

http://www.butterflylabs.com/monarch/

This purchase: 1.71GH/s/watt for $4680

The problem is you're pretty late in the game, so it's really hard to predict what the difficulty rate will look like in even a few months.

It's possible that your USB miners will be worthless in a few months due to the mass in flux of newer ASICs. It's also possible that so many people will be purchasing newer ASIC miners that ALL mining will become entirely unprofitable.

And all of this assumes the relative stability of the BTC/USD exchange rate. :)

I'm pretty sure a subsonic rifle shot is still pretty loud. :)

> HTTP header parsing/writing was never near a performance issue.

Since header lengths are not limited, and a single TCP packet's payload is quite limited, long headers can cause very measurable latency difference. Additionally, while I agree the generation / parsing overhead is probably quite small, saving it for every HTTP request is still a boon.

I'm also curious where you are reading raw HTTP from?

For me it's primarily in two situations, reading a packet capture from WireShark, or in the browser's debugger. In both cases, the tool will end up translating the request for me.

I think that the problem with this thinking is:

(a) Even altering a parameter once in the address could be considered illegal under the current laws.

(b) The access of information in and of itself would not be illegal alone. Say I kept a bunch of people's information in paper files in file cabinets. Then I gave you access to retrieve yours from the file drawer yourself. It is sleazy, but not necessarily illegal to look at other files in the drawer, as I have given you access to their container.

(c) Even if you want to make the action in (b) illegal, the reasonable punishment is almost certainly not a double digit prison sentence.

Enable mod_gzip?

This analysis is flawed in that it assumes that the host with the closest geographical proximity should always provide the lowest network latency.

Very often it does, but especially for smaller countries that do not have the same number of peering points that exist in the US or Europe, latency will appear out of whack when compared to geographical proximity.

For this test to be valid, you would need to measure the latency to the remote host from all four servers to determine if indeed the ideal route was chosen. Even then the latency may vary depending on the time of day or other network conditions.

From a practical standpoint, a simpler and more accurate metric would be to compare page load times in different geographies before and after enabling Route 53.

Wouldn't this make sense only if you weren't using a hashtable, i.e. a list or something else that had to be traversed?

Did your A/B testing lead you to the conclusion that the all caps text in your legal docs was losing signups?

Checking a laptop in your luggage? Forget being stolen, it seems like a great way to get your laptop broken.

Most PCI audits are not mandatory unless you have had a breech that compromises data. Before that "self-certification" reigns.

Three people (including me) from my team were attempting to get tickets. We were all on the page 30 minutes before opening and entered the queue within seconds of opening.

2 of us never got anything other than the looking for tickets page multiple times.

1 of us got to the payment page on one attempt only to have it fail.

It's frustrating that we should each spend 1.5 hours of time to babysit what is essentially a lottery.

If it's going to work like a lottery, then just make it one.

Here's a very simplified example. Note that I am using some made up notation here: "." represents NULL charcaters, and the numbers would actually be stored in hexadecimal.

The contents of 48 byte disk with a 7 byte file allocation table (FAT):

    [................................................]
     FAT----DATA-------------------------------------

Write a file named "blah" with contents "abcdefghi":

    [blah.90abcdefghi................................]
     FAT----DATA-------------------------------------

Notice that the FAT area contains the filename, a "9" representing the length of the file, and a "0" indicating the spot in the data area where the file begins. In fact, the file system might choose the put the data somewhere else, and that's absolutely fine. For example:

    [blah.95.....abcdefghi...........................]
     FAT----DATA-------------------------------------

Represents the exact same contents with the file stored at an offset of 5. So say I want to delete this file now. A typical file system would do something like this:

    [.lah.95.....abcdefghi...........................]
     FAT----DATA-------------------------------------

It writes a single NULL byte to the file system to mark the filename as empty, and the file is gone. None of the rest of the data is cleaned up.

Not to be pedantic, but:

    {users: [{name: 'joe', age: 12}, {name: 'bob', age: 23}]}
    <users><user name='joe' age='12' /><user name='bob' age='23' /></users>

Projects: especially difficult/interesting/field-relevant ones.

Honesty: if I can't immediately tell that you are over-stating your experience, I will find out in the phone screen.

Fit: something that helps me think that you'll be a good fit with my team from a personality stand-point. This may be represented by listing hobbies (though be careful with this) and the types of projects you work on.

Attention to detail: Your resume should look like something that's professionally crafted. If it doesn't, I am going to assume your code looks the same way.

Back when I was applying for jobs several years ago, I would tailor my resume for the job that I was applying for. The more your experiences matches the listing, the better. In my case, being somewhat of a generalist, I have some experience developing web, desktop, cli/server and embedded code, so I would rewrite my highlights at each job where appropriate to best address the skills being looked for.

What is the benefit of using this over simply redirecting / forcing SSL use for form handlers and other sensitive URLs on the application end.

Doesn't this just add more complexity to a problem that is not that hard to solve?

As a "drop out" technical manager, I have hired about 50% with CS/CE degrees, 10% with non applicable degrees and 40% with no degree at all.

I should also mention from experience with a niche app which has a few thousand users, I ended up making $1.50 in revenue over the course of a year.

I released an update this December, and am now charging $1.99 and doing thousands of percent better, and I haven't put any effort into promoting the app yet.

If you start an LLC/LLP/Corp, you will be paying some amount of money in fees and taxes each year to keep it in existence. Where I am in Rhode Island, USA, this is $500 for minimum corporate tax (applies to LLC/LLP as well), and a $75 filing fee for an annual report.

You will also may want to hire an accountant to prepare your corporate / LLC tax return / schedules, especially in your first year.

In short, if you are going the LLC/Corp route, expect to add $1000 per year in overhead, plus at least a dozen hours in time.

I personally would wait (you are already a sole proprietor by default) and then if you start making some decent money look into it again. It is possible to transition an app from a individual to a company account, but there is some footwork involved.

Keep in mind, you will not be reaping any tax benefits of incorporating until you are making actual money. Your decision to incorporate / form an LLC at this point should be driven only by the amount of risk you are personally taking on in potential liability. I would suspect for most free applications, you won't have much to worry about. But if you're sitting on a pile of money and are writing an app that some how might inspire someone to sue you, you might want to think about it.