nhf

In all seriousness, I've only run into two groups that (in my personal experience) still use LJ: Eastern Europeans, especially Russians, and pop-culture related "creators" (which is the broad brush that I'm painting fan art, fanfiction, etc with). It's a really weird mix.

This guide is nice for people that are motivated, but a large chunk of users (probably the majority, but I have nothing to back that up except a guess) will only encrypt their emails when it's as frictionless as sending an unencrypted email.

Encryption tends to be successful when it's baked in to the system (e.g., WhatsApp theoretically), a preset default (e.g., Android 5.x+ and iOS), or requires absolutely minimal effort (e.g., the HTTP -> HTTPS "switchover"). This has been a known issue for a pretty long time [1, 2].

[1] http://www.gaudior.net/alma/johnny.pdf

[2] https://cups.cs.cmu.edu/soups/2006/posters/sheng-poster_abst...

Signatures on credit cards are an interesting artifact. Planet Money did a nice podcast on signatures a year or two ago: http://www.npr.org/sections/money/2014/08/29/344034815/episo...

Apparently signatures are still used as a method of determining liability for fraudulent charges, who knew?

From the perspective of a current CMU student: the Software Engineering Institute (SEI) here is the responsible party here. Note that they are separate from the main ("academic") school of CS. They are a federally-funded research and development center, essentially a computing-focused contractor and consultant for the DoD (see more here: http://www.sei.cmu.edu/about/organization/workingwithanFFRDC...).

So, in short, the people at the SEI sleep as well as the people at Raytheon or Lockheed Martin that build the drones, or the folks at Alcatel-Lucent who helped the NSA way back, or maybe even the guys way back at Los Alamos. It's not like some grad students or CS professor got strong-armed into doing this. It's literally their job to do this kind of work for the DoD and other government offices. While the ethics of this line of work are certainly up for debate, they knew exactly what they were getting into.

Scraped from archive.org: http://imgur.com/a/PlCfX

If we're using fashion as an analogy, it's kind of a mixed bag.

There's both this type of fashion: https://s-media-cache-ak0.pinimg.com/236x/c0/66/12/c06612029...

...and this type of fashion: http://img.izismile.com/img/img7/20140521/640/fashion_runway...

out there.

Things can be fashionable, yet crisp and usable. Things can also be crazy, stupid (to most), and groundbreaking (to a few). Put another way, fashion currently dominates, but that would be okay since we don't need to sacrifice good visual design in the name of performance any more.

Unfortunately:

* People get lazy. * People don't user test. * People like to follow trends. * Designers and developers get micromanaged.

...along with all of the other monkey wrenches that most developers and designers have gotten used to. People end up building cool-looking websites that aren't as usable as they should be.

I can definitely agree with both sides of this argument.

People like things to just work—and people also like to be able to customize things. I could see this concept being used as a superpowered versioned of IFTTT if it gets baked into products. People can tinker with building things, but there are also premade recipes that you can load and use.

Yep! There's a lot of research going on in that area. For example, http://dl.acm.org/citation.cfm?id=2702416

Some recent work on fusing machine learning with Mechanical Turk workers to create "sensors".

It wouldn't have the cool art, but it wouldn't be too difficult to produce a first approximation of a similar website using existing frameworks like Bootstrap.

Internet measurement is such a cool subfield. It's really astounding how much data is locked up within the nooks and crannies of digital activity.

To be fair that's a pretty common tactic even here in the US. T-Mobile's cheaper "unlimited data" plans are 5GB of 3G/4G with aggressive traffic shaping afterwards that basically dings you down to 2G-ish speeds unless you're browsing at 3AM or something. Do these companies just not disclose that though?

This presentation (GCHQ) specifically notes use of HDFS and Hadoop to process data: https://theintercept.com/document/2015/09/25/gchq-analytic-c...

Sesame Workshop gets more resources and episodes, PBS gets to stop paying for Sesame Street. I don't think it's all that bad. It's not like it's going off the air on PBS any time soon.

P2P by default except on the enterprise edition, in which case it's local network only. They also claim to disable P2P if they detect your internet connection is metered (I'm not sure how that's done).

A plug for our paper, also at the SOUPS conference. We tackled a similar topic, but with a different method and broader focus (how experts and non-experts in general conceptualize the internet as a system): https://www.usenix.org/system/files/conference/soups2015/sou...

It's great to see a large company like Google focusing on this kind of work though.

> This was an extremely childish move that had the propensity to hurt our industry more than help it. It is incumbent upon us take safety seriously in conducting these experiments. We can't count on level heads from outside the tech industry if we aren't willing to show that we care about people's lives and their safety when we're conducting these experiments.

Agreed. I would have no problems with them doing this on a test track, closed highway, or even a quiet road at low speeds. Even if we take the best possible negative scenario—say, the car is disabled going at 25 miles an hour, the driver can't handle manual steering, and then runs into someone's fence—the insurance companies are going to throw the book at the driver when they learn that they purposefully disabled their car and engaged in dangerous behavior on a public street.

I'm all for pushing the boundaries of security research (there are people in the labs all around me right now doing crazy stuff), but at least we in the academic world get our crazy stuff signed off on by a panel of competent experts.

I highly recommend getting the Pocket Casts app for Android. In my opinion, it's better than iTunes for podcasts (except for the fact that you have to pay separately for the desktop and mobile clients).

Yes, for a lot of stuff, especially for unit conversions and things like that.

This wasn't an accusation of group responsibility. In fact, they specifically state in their announcement that

> Ellen asked me to point out that the sweeping majority of redditors didn’t do this, and many were incredibly supportive. Although the incredible power of the Internet is the amplification of voices, unfortunately sometimes those voices are hateful.

Like any popular site, Reddit attracts everything from enlightened discussions to 4chan-level casual trolls (and worse). Pointing out the issues with threats and indicating that the community could be better without individuals who would publicly threaten the life of another person seems very reasonable to me.

I think this was the right thing to do from a PR perspective. Having Steve back as the new CEO will definitely be good for the community.

I also applaud Reddit's announcement for calling the community out on their childish BS:

> As a closing note, it was sickening to see some of the things redditors wrote about Ellen. [1] The reduction in compassion that happens when we’re all behind computer screens is not good for the world. People are still people even if there is Internet between you. If the reddit community cannot learn to balance authenticity and compassion, it may be a great website but it will never be a truly great community. Steve’s great challenge as CEO [2] will be continuing the work Ellen started to drive this forward.

All in all, a good day I think.

Agreed. However, United still has very glaring security issues with their site that go far beyond the 4 digit PIN. I can't provide concrete details because I still have a pending bug bounty report, but stuff along the lines of bad/missing HTTPS and easy hijacking of account management sessions.

I'm a big proponent of brainstorming while walking. If I hit a mental block while programming, writing, or doing some other sort of critical thinking/creative exercise, I'll just drop everything and go on a 30 minute stroll. I'll either end up solving my problem or getting some exercise, so either outcome is good.

On the other hand, I have a mathematician friend who's a big proponent of shower brainstorming. We still give him crap for the day when he was working on a proof and took a two hour shower.

If I remember correctly from the presentation (I was at W2SP where the associated talk was given) there were unspecified performance benefits over having this outsourced to the plugin architecture. I can't download the paper right now for some reason, but I'm sure it's detailed in there.

I also wonder how ad providers with a greater scope than just advertisements would react to this. For instance, take Facebook—their ecosystem encompasses ads, social features, and login/identity management at this point. If we get into an arms race at some point, I could see this getting ugly if these other features become collateral.