HNHacker News
TopNewBestAskShowJobs

mlb_hn

726 karma · joined December 29, 2017

playing with little space aliens at julius.ai
submissionscomments
mlb_hn··on Former reporter creates ‘Rate my Professor’ for newsrooms
I'm doing my graduate work on that. It's hard.

MIT has a cool project looking at how individual terms are covered though: https://www.media.mit.edu/projects/media-cloud/overview/

The relationships when modeling from the author/publication get really big really quick. I suspect as NLP gets better it'll get easier to model though.

mlb_hn··on Coverage of a Disinformation Operation Against 2019 EU Parliamentary Elections
Wasn't that thorough. ODNI report from 2017 was broader in scope (https://www.dni.gov/files/documents/ICA_2017_01.pdf). The 2019 Mueller report looks restricted to areas where there was possible collusion (just some of the social media stuff that was retweeted by Trump campaign people/stuff related to the DNC hacks/etc.).
mlb_hn··on U.S. military to analyze 350B social media posts to understand popular movements
I'm not sure there's been any evidence that the military has had success utilizing current computational approaches to data. E.g. the DCGS project was a failure and I don't think that project tried anything beyond what you'd get out of Microsoft Excel (https://nypost.com/2014/10/27/army-spent-5b-on-failed-techno...). DoD plans for projects that last for decades and the algorithms people are building for commercial purposes today are developing way faster than DoD's procurement can keep up I'd guess,
mlb_hn··on U.S. military to analyze 350B social media posts to understand popular movements
I'd guess there have been significant advances in network theory and NLP since Sun Tzu's time but I'm not an expert.
mlb_hn··on U.S. military to analyze 350B social media posts to understand popular movements
US Government's been great at collecting tons of data. Analyzing it... not so much.
mlb_hn··on Can Facebook Be Fixed? Should It Be?
The full friend data is the only substantial difference I'm seeing between the two (which is quite substantial). The current Facebook API doesn't allow an app to even see friend names unless the friends also authenticate the app.

Basically, with the CA quiz Facebook allowed you to export your data to another site or app so you wouldn't need to put it into that site as you're saying. If you were to instead just upload the entire Facebook archive that seems like you're opening up even more data about yourself to potential abusers as well as some friends data that's in there (I think at least name and phone number?).

mlb_hn··on Can Facebook Be Fixed? Should It Be?
Uh, isn't the ability to export your data what got Facebook into the whole Cambridge Analytica scandal in the first place?
mlb_hn··on Audit suggests Google favors a small number of major news outlets
Apologies on the terminology - by invalid traffic I'm referring to bots as well as click farms and other issues as used in the Media Rating Council's definition (they divide it into general and sophisticated invalid traffic both of which have a lot of types of traffic, http://mediaratingcouncil.org/101515_IVT%20Addendum%20FINAL%...).

I'm just a bit concerned that the Russian malware dudes were re-purposing their click fraud for astroturfing way back in 2015 and they had no problem just sitting and building dwell time instead of bouncing (https://www.trustwave.com/en-us/resources/blogs/spiderlabs-b...). I haven't been able to find anything indicating that US media companies have any kind of tracking to defend against or even identify a similar strategy being used to hit their article analytics to influence article production/placement, especially when it's now known that a Russian information campaign against the US was going on at the time.

mlb_hn··on Audit suggests Google favors a small number of major news outlets
Thanks for posting here pixelmonkey! I had a question; I asked the Parse.ly help/marketing people a while ago if you guys were tracking the rates of invalid traffic on articles and they said you guys don't track it. Is that accurate? Or are there any estimates on how much noise/dirt there is in the data?
mlb_hn··on Senate Testimony on Privacy Rights and Data Collection in a Digital Economy
At ~53:00 or so Mr. Brown asked about the impact of analytics on newspapers. You pointed out that the newspapers are basing editorial decisions on the analytics (article metrics).

As we all know, click-fraud's a huge problem on the internet. Have you come across any recent numbers on click-fraud in the newspaper analytics? It seems like if your description is accurate, if I wanted more stories about a topic I could just have a botnet give those articles lots of views.

mlb_hn··on Facebook sues analytics firm Rankwave over data misuse
I think there are two issues with the FB data being conflated here: CA's targeted advertising and the population data itself.

For the advertising, concur there's a lot of disagreement over whether or not CA's models worked.

For the data though, having a dataset on population preferences and network connections is useful for things besides just advertising. For example, most people agree the Russian campaign was based on dividing the population on divisive issues; the data exfiltrated from Facebook should reveal any particular group's divisive issues.

TLDR; There's other uses for the data besides targeted ads in the broader topic of data privacy.

mlb_hn··on Facebook showed me my data is everywhere and I have absolute no control over it
> 1. In most responsible news organizations, there is a significant firewall between advertising and news. In two places I worked, the sales people weren't even allowed to go into the newsroom.

Do reporters not get any access to their analytics?

mlb_hn··on Better Language Models and Their Implications
A lot of the advancements that made the Manhattan project were published by the Germans. On hearing about Hiroshima, Otto Hahn was "'shattered,' and went on to say that he felt 'personally responsible for the deaths of hundreds of thousands of people,' believing that his discovery had made the bomb possible." (https://www.washingtonpost.com/archive/opinions/1992/03/01/t...)
mlb_hn··on Why can’t a bot tick the 'I'm not a robot' box?
The misconceptions about nonhuman/invalid traffic (NHT) seem like a problem brewing. There's an arms race between the NHT guys (e.g. ad fraud networks) and the guys trying to detect it (e.g. ad providers). Meanwhile, a lot of people are using analytics to inform decision making assuming most traffic is legitimate (e.g. news organizations, anyone doing A/B testing). The guys naively using analytics with weak feature detection may be totally unprepared to deal with nonhuman traffic from repurposed networks which have been optimized to defeat the more advanced countermeasures =/
mlb_hn··on Death and Valor on an American Warship Doomed by Its Own Navy
DOD issue, if not USG. Army had (has?) the same issue as well, 2015 whitepaper on sweeping everything under the rug until it breaks (https://ssi.armywarcollege.edu/pdffiles/pub1250.pdf)
mlb_hn··on As Facebook Raised a Privacy Wall, It Carved an Opening for Tech Giants
One reason would be that RBC was allowing people to send money from its app via Messenger - http://www.rbc.com/newsroom/news/2013/20131211-facebook.html
mlb_hn··on As Facebook Raised a Privacy Wall, It Carved an Opening for Tech Giants
That's what that article had - on Googling it looks like Spotify makes suggestions in messenger when talking about music as well.

https://www.theverge.com/2017/8/14/16143354/facebook-messeng...

mlb_hn··on As Facebook Raised a Privacy Wall, It Carved an Opening for Tech Giants
> Spotify, which could view messages of more than 70 million users a month, still offers the option to share music through Facebook Messenger.

Not sure why the others needed it

mlb_hn··on Disinformation Report: An Investigation into Russian IRA Activities in 2016
One substantial difference worth noting between the Russian astroturfing and astroturfing from other countries is that Russian crime have been refining their ability to automate human-like behavior for ad-fraud purposes. For those not familiar with ad fraud, there's been an arms race between criminal organizations to emulate human behavior clicking on ads and the advertising companies trying to detect it. The resulting algorithms have dual use properties that can be applied to astroturfing (e.g. https://www.trustwave.com/Resources/SpiderLabs-Blog/Bedep-tr...).
mlb_hn··on Inside the British Army's secret information warfare machine
One difference is the Russians have massive malware networks available from their criminal operations they can use for certain information operations, e.g. https://www.trustwave.com/Resources/SpiderLabs-Blog/Bedep-tr.... I have not seen anything on western governments doing anything close to that sort of thing.
mlb_hn··on YouTube comments are heavily manipulated, and I fear for our democracy
It's not just comments - the ability to drown out other views applies to actual news articles too. We saw back in 2009 with the Time Person of the Year campaign how easy it was for coordinated groups to drown out the public (https://techcrunch.com/2009/04/21/4chan-takes-over-the-time-...). As legitimate news organizations increasingly use analytics as part of their decision processes, that unsolved vulnerability presents an opportunity for click farms to drive traffic to articles and indirectly manipulate the production and placement of actual articles as well =/
mlb_hn··on The military’s obsession with energy drinks is contributing to PTSD
Yeah, the article is a bit misleading, looking at the linked study the authors point out those limitations,

> The findings also support the need for further research in this area, particularly longitudinal studies that would allow for stronger assessment of causal relationships.

> the measures for depression and PTSD each included two items related to sleep and the models were run with sleep problems as a predictor, which could lead to potential confounding in the analyses; however, this is consistent with research in this area.

Personally, one of my worst moments deployed was when our Ripit/RedBull/Tobacco supply got blown up (well, technically the DFAC and PX actually collapsed after being blown up due to shoddy construction) and we had to have people fly them in from BAF. I'm sure the lack of energy drinks contributed to PTSD for those who didn't have alternative logistics.

mlb_hn··on Apple Further Refutes Bloomberg Report Suggesting Spies Tampered with Servers
You may be misunderstanding how anonymous sources work. The NYT knew who the sources were, it's the public who did not (see https://www.nytimes.com/2004/05/30/weekinreview/the-public-e... for NYT's writeup of some of their mistakes)
mlb_hn··on The Diplomatic Couriers Who Deliver America's Secret Mail
For the USG, it's currently authorized under Executive Order 12333 (https://www.archives.gov/federal-register/codification/execu...).

Paragraph 2.1 states "2.1 Need. Accurate and timely information about the capabilities, intentions and activities of foreign powers, organizations, or persons and their agents is essential to informed decisionmaking in the areas of national defense and foreign relations. Collection of such information is a priority objective and will be pursued in a vigorous, innovative and responsible manner that is consistent with the Constitution and applicable law and respectful of the principles upon which the United States was founded."

mlb_hn··on The U.S. Air Force learned to code and saved the Pentagon millions
Sounds like more to do with not taking battlefield conditions/training into consideration.

When push comes to shove, if you give people cool tech, unless it's easy to mindlessly pick up and use they won't use it. When I was in, I had boxes and boxes of crap that was on the books that we never used (and didn't know how to use but had to inventory), then the stuff we actually used.

mlb_hn··on Complicating the Narratives
Yep. There have been lots of studies starting in the 1950s showing that journalism wasn't as objective as they wanted to be starting with White's study on "The Gate Keeper" and implicit bias in stories (http://www.aejmc.org/home/wp-content/uploads/2012/09/Journal...).

But then everyone immediately forgets.

mlb_hn··on Justice Dept. Seizes Times Reporter’s Email/Phone Records in Leak Investigation
That is accurate: https://www.pbs.org/newshour/nation/obama-administration-set...

Background on FOIA: Basically, in 1966 the US Government created the Freedom of Information Act that allowed people to ask for records from federal agencies. In theory, it was supposed to create transparency. Unfortunately, it's a bit outdated and in many areas it's turned into many in the Government thinking requestors are just out to get them (and a lot of it is also commercial requesters). Under Obama, the USG denied more requests than ever. In their defense though, it's got really easy to do a FOIA request online and flood the government with requests (it only applies at the federal level, states have their own version of public information laws).

mlb_hn··on Facebook Back on the Defensive, Now Over Data Deals with Device Makers
This is a followup to the first article where they confused the ability to use a device to access Facebook functionality (e.g. upload photos) without using the app or browser with giving a third party the ability to access Friend data.

The common understanding seems to be that Facebook is allowing device manufacturers to download user data the same way that Cambridge Analytica and other third party services did.

So far, they haven't shown that device manufacturers are siphoning that data to their own databases short of the statement "Facebook acknowledged that some partners did store users’ data — including friends’ data — on their own servers" (in the original article, no explanation of context). What they have shown is that Facebook gave device manufacturers the ability to let users use the device to interact with their Facebook accounts without using the web browser.

mlb_hn··on Why We Disagree with The New York Times
I'm not sure it's meaningless. It seems to me that there is a distinction between [device manufacturer X is building their own copy of Facebook's graph by acquiring friend data] vs [device manufacturer X allows a user to login to Facebook and access friend data on their device without having to use the web browser].
mlb_hn··on Why We Disagree with The New York Times
Facebook's making the distinction between allowing a device to offer Facebook-like services (which require Facebook functionality) and third-party apps that suck up all your friend data.

On the one hand, Facebook's got a point, that if you want to be able to use Facebook on a device without going through the Facebook app or the website the device needs to be able to authenticate onto some sort of API.

On the other hand, the NYT article makes the claim that the makers of the devices got access to the Facebook data, writing "Facebook acknowledged that some partners did store users’ data — including friends’ data — on their own servers". However, Facebook never followed up on that in their article, just pointing out that if you are logged into Facebook on a BlackBerry that the BlackBerry can make the same requests you could if you were logged into Facebook through the web browser.

The question that matters which neither side addresses well is how much of that data makes it to device maker servers (for a while, the NYT homepage was claiming 'dozens' but they removed that and it doesn't appear to be substantiated in the article).

There are related worries with the device having access to the Facebook data itself, but at that point you need to start worrying about malicious activity by device makers in general. E.g. will my phone start sending my web history back to its maker as well? my bank account numbers?

← PreviousPage 3 of 4Next →