"deal with the relentless threats of terrorism"
When are our legislators going to declare war on rust? Rust never sleeps, and neither should we. Someone should do something! It's a travesty!
2662 publicly visible posts • joined 8 Nov 2007
A story I heard on the radio in Ireland. A certain station master was responsible for making reports to head office whenever there was a train accident or derailment. Had something of a literary bent, it seems, because his reports ran into several pages of quite descriptive prose about the ins and outs of the incidents and the actions taken to get things back on track.
Head office got a bit pissed off with the length of his reports and asked him to be a bit more to the point. The station master complied, and the next time there was a derailment (in 1921), his report simply stated: "Off again. On again. Gone again. Flanagan"
That was my first thought. My second is to only allow password-less logins via the MP creating a public/private key pair and handing over the public key to the IT guys in a controlled setting.
Works fine for ssh (where I can upload my key and store it in the authorized_keys file, then disable login via a password), so I'm pretty sure that it should work for TLS/SSL as well (and is apparently resistant to MITM, with ssh, at least). You might have a bit more work to do with regard revocation of an authorised login key, but that's par for the course.
But if it's the other side, if it is Trump who has the recordings, then why not release them? He's clearly not one for keeping quiet about anything he thinks might make him look good or might sound "heroic", no matter how much it actually backfires.
Dougal: Actually, Ted, you've done this to me before, so I took the liberty of taping the conversation
Now, we'll just have a listen...
(starts tape) "..."
Dougal: I stand corrected
https://www.youtube.com/watch?v=hPCH4rZU-3Q
Absolutely.
The way to look at this is to calculate the margin of error for this sample.
The sample size is 2692+2128 = 4820
We calculate the 95% margin for error as 1.96 * sqrt(0.5 * (1.0 - 0.5) / 4820) (footnote)
This gives 0.0141157044469341 which says that 95% of the time, the expected number of women will be within 50 +/- 1.412%. This translates to a range of 4820/2 +/- 68 people, or [2342,2482]. The value 2,128 is outside this range so all we can say is that using a 95% confidence interval, the assertion that males and females are equally represented (p=0.5) is not supported by the sample.
Chi-squared is slightly different since it's a measure of fit of a set of individual observations to the expected, but the above is effectively its application to the average case (ie, it ignores the spread of individual samples). Neither provides a measure of how unrealistic/unexpected the result [set] is, as Vaidotas Zemlys has pointed out.
footnote: http://www.dummies.com/education/math/statistics/how-to-calculate-the-margin-of-error-for-a-sample-proportion/
Makes for pretty chilling reading:
https://www.theguardian.com/technology/2017/may/07/the-great-british-brexit-robbery-hijacked-democracy
I notice that any article that mentions Cambridge Analytica over at the Graun has a new note at the top stating:
This article is the subject of separate legal complaints on behalf of Cambridge Analytica LLC [and SCL Elections Limited, and Sophie Schmidt.]
(the "and" bit might vary across articles, but you get the gist)
While spending over campaign limits because Facebook and so on aren't covered by regulations is one thing, the use of targeted ads that use bulk, aggregated data about individuals is what's really scary. The fact that this also seems to be done outside the EU, and so isn't subject to privacy laws should be giving everyone cause for concern.
I set up ipv6 recently, over a Hurricane Electric tunnel. I'm in Ireland and I picked a tunnel endpoint in England. I could have picked Ireland or Holland or anywhere, really. However, whichever one I picked, it would have fucked up a perfectly valid and legal Netflix subscription because they consider me as being someone who's using a proxy to defeat their region locks. Bit of a sledgehammer approach.
I can understand the old processes involved in setting up contracts for regional distribution of films and such, but seriously, in this day and age with so much old content and internet-based delivery, why should we still have geoblocking on so much stuff? The argument about promotion and localisation in each region is bogus so long as the channel provider is still counting eyeballs and paying up as they should. I don't need to see promo material for shit that I watch. I just want to be able to watch shit and have the channel make sure that the content creators get paid. Sort of like a FRAND for consumable digital content. Is that so hard?
Hard to imagine there would be any services out there that would be IPv6 only.
The biggest plus for home users, as far as I can see, is that if you want to run servers from your home network, then going ipv4 means that you need a static ipv4 address which costs more. If you have native ipv6 and a reasonable amount of tech knowledge you can partition your network and transition from an ipv4-style DMZ into the equivalent ipv6 version. So, all those family-shared photos can go straight to your secure ftp/webdav server instead of Facebook or other cloud servers.
I'm pretty sure that a big factor in ISPs not supporting ipv6 is that they realise that it means that they won't be able to charge extra for static IPs. That, and not wanting to spend money on staff training.
Less massive planets would have a harder time holding onto lighter elements thanks to solar winds and the like. The lighter elements will obviously be higher up in the atmosphere, making them more likely to be stripped away over time. More massive planets are better at keeping hold of these because gravitational forces are higher, but there are no doubt several other reasons as well (like distance from the star).
It's called "key escrow". The device used by the consumer has a secret key that can be used (along with other information, such as the device ID) to recover the session key used to encrypt the communication. The device is supposed to be tamper-resistant, so users aren't able to access the escrow key. A copy of the that key is also stored by law enforcement, allowing them to decrypt the communication whenever they want.
The other way to implement it is to present users with a new encryption scheme that's supposedly secure, but has a flaw that is known to your mathematicians, but (supposedly) not anyone else. This gives them an advantage when it comes to decrypting stuff because it becomes feasible to use some short-cut to brute-forcing the message.
With both sorts of secret (escrow key or "back door"), the security of everything is dependent on how secure that secret is. As we've seen from NSA leaks (giving rise to this weekend's botnet that hit the NHS among others), plus the existence of plenty of hardware and maths wizzes outside of the NSA (or whoever) who can, with enough time, effort and money, crack that secret, rendering the encryption completely irrelevant.
Unfortunately the EFF's signing on to the bogus ...
That's not what the article said. It said that it's worried about "the independence of the office and its ability to conduct fair investigations".
Personally, I don't think that an investigation would come up with enough to tie Trump to the Russians directly, though I suspect that there are others in his entourage who were compromised. Still, if he has nothing to hide, then why should he fear the probe? A normal, sane individual would allow this to run its course. Instead, Trump uses bluster and now, it seems, direct interference in the workings of the investigation. That doesn't project an image of him being free of taint.
the partisan interests of a few wealthy donors.
Surely you're not serious here, or are you back to talking about Trump and Russian donors again?
If it's looked at as a copyright statement, then the default state when you put the appropriate (c) mark on the document is that it is your [the author's] property and should fall completely under copyright laws. If that's all you do, then the position is clear: you [someone other than the author] can't go and copy the material except under certain fair use conditions.
When you add the GPL statement, you are granting certain extra rights (but, crucially reserving certain other rights, such as not tampering with the rights granted, or modifying the document and re-releasing it without continuing to honour the conditions set out under the derived works sections) to anyone who might happen to have or receive a copy of the document. It shouldn't be looked at under contract law. In particular, it shouldn't be necessary for both parties (the author and the person who has a copy) to enter into a signed arrangement.
The question of how the person receives the GPL-copyrighted document should also be irrelevant. It's like the question of whether you buy a book from the publisher, a bookseller or you get it second-hand, somehow. The delivery mechanism or how you came by the copy is irrelevant since copyright resides within the copy itself.
It's humbling to see such a devastating and wide-ranging attack appear as if out of nowhere. Indiscriminate, uncaring and just plain nasty in it effects. If I were a normal person (well, actually, I am, more or less) and not some puffed up politician, this would leave me speechless and basically in awe of the fact that I am basically a zero when it comes to the new normal elemental forces at play on the Internet.
If you can streamline the installation of a secure VPN and get caching of push data when the link is down, then the convenience factor could be worth it.
However, this is really nothing that a moderately tech-savvy person couldn't do in an afternoon. At least the secure VPN/DMZ part, anyway. The store and forward part will depend on the particular IoT device. Most of them won't admit to this sort of configuration, although all of them should by right allow you to configure exactly where the data will be sent to, and over which network link, rather than being hard-coded to only send to a fixed server or using a proprietary protocol (making me notice that this particular offering has a whiff of embrace/extend/extinguish about it).
Apropos of nothing, I recently lost the drive attached to the Pi that I'd been using as a music/radio player. Nothing lost since it was an old drive that I'd expected to fail. I had also been using the machine's wireless card to provide fail-over Internet access so that if my broadband went down, I could just turn on tethering on my phone and I'd be back online again. I decided to replace the Pi with an ODROID (simple) and then idly wondered about doing the fail-over on my OpenWRT router. Turns out that my wireless card can be used in both client and AP mode at the same time, so once I had that insight it took about an hour to migrate the fail-over completely onto the router. No doubt setting up a VLAN/DMZ would only take a similar amount of time.
Now if only my ISP would support IPv6 in some way.... though I guess that would take a bit more than an afternoon to fully explore :)