Stories
Slash Boxes
Comments

SoylentNews is people

SoylentNews is powered by your submissions, so send in your scoop. Only 18 submissions in the queue.
posted by Fnord666 on Monday March 20 2017, @04:56AM   Printer-friendly
from the is-it-assault-or-battery? dept.

Three months after a journalist reported being attacked by a troll who posted a seizure-inducing image on Twitter, a suspect has been arrested:

A man accused of triggering an epileptic seizure of senior Newsweek writer Kurt Eichenwald through a tweet was arrested by the FBI on Friday morning. An FBI spokesman said the name of the suspect has not been released but confirmed that an arrest was made, Dallas News reported.

The arrest comes three months after Eichenwald said he suffered a seizure after someone sent him a video clip of a flashing strobe light in an intentional effort to trigger his epilepsy. A Twitter account called @jew_goldstein — which has since been suspended — responded to Eichenwald with a gif of flashing strobe lights and a message: "You deserve a seizure for that post." Shortly after, Eichenwald's account tweeted: "This is his wife, you caused a seizure. I have your information and have called the police to report the assault."

From the Dallas News article:

The agency announced that John Rayne Rivello, 29, of Salisbury, Md., was arrested Friday morning in Maryland on a cyberstalking charge.

[...] Eichenwald's attorney, Steven Liberman, told Newsweek that "What Mr. Rivello did with his Twitter message was no different from someone sending a bomb in the mail or sending an envelope filled with anthrax spores."

[...] According to a criminal complaint, messages sent from Rivello's Twitter account mentioned Eichenwald, saying "I know he has epilepsy," "I hope this sends him into a seizure" and "let's see if he dies."

Authorities also found an screenshot of Eichenwald's Wikipedia page on Rivello's iCloud account, the criminal complaint said, altered to list his date of death as Dec. 16, 2016. Other files on the iCloud account include a list of things that trigger epileptic seizures and a screenshot of a Dallas Observer article about Eichenwald's attempts to find the person who tweeted at him.

[...] On Friday, Eichenwald said that more than 40 people sent him strobes once they found out that they could trigger seizures.


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 0) by Anonymous Coward on Monday March 20 2017, @06:25PM

    by Anonymous Coward on Monday March 20 2017, @06:25PM (#481655)

    (This post is UK-centric because that's what's easily found on the web. The rules probably differ under, say, the US's FCC, or equivalent bodies elsewhere.)
    There are appliances that do just that, known as "Harding boxes"; a list of them is in this PDF [amazonaws.com], and you're required to have a "pass" result from one of them before BBC will accept your program. Note that Professor Graham Harding's name is used to refer to a specific algorithm originally used for checking PSE safety, to a particular series of appliances and applications implementing that algorithm, and more loosely for any PSE checking device, whether it uses the original Harding algorithm or a different one.

    The rules to be implemented by such a check are laid out in Annex 1 (from p.15 on) of this PDF [ofcom.org.uk].

    There are essentially two concerns:

    • flashing areas of the screen
      • if neither color is "saturated red" (Is #ff0101 okay? #010000? not clear what "saturated" means here...), then:
        • no problem if minimum brightness over 160 cd/m^2
        • no problem if maximum and minimum brightness differ by less than 20 cd/m^2
      • no problem if total flashing area no more than 25% of screen area
      • no problem if no more than three flashes per second

      While specs are given in cd/m^s, which varies wildly by screen, they do give a curve for converting between cd/m^2 and video levels. This looks pretty straightforward to implement, subject to some questions about "saturated red". (I'm sure further reading would turn up the answers, just not sure why the OFCOM guidance isn't more clear.)

    • regular patterns, such as stripes -- too complicated to explain, and also pretty complicated to check for; go read the PDF if you care.

    Since flashing is the preferred tactic of PSE griefers on the web, I'd be inclined to check only for flashing, and ignore stripes.

    Note that any flashing that is less than 25% of screen area (e.g. 50% height by 50% width = 36% = no problem) is okay. That makes it even simpler -- block all animated gifs over 25% of screen size. For larger gifs, there's also the crude, though more invasive, expedient of enforcing a minimum delay of 167ms between gif frames; as a flash takes two frames, this enforces 3 flashes per second maximum. These are also usable, in conjunction with a classification algorithm, as mitigation strategies when playing back "bad" gifs; I see good gifs playing as normal, bad gifs less than 25% of screen are click-to-play, but then play normally, while large bad gifs are click-to-play with an enforced maximum framerate.