Twitch Aces YouTube Again
June 21, 2019
I did a quick check of YouTube Live, the finder for live streams available on YouTube. You can locate this dashboard at this link. I scrolled through the results on YouTube at 0630 US Eastern time. I located this video link on the YouTube Live page:
YouTube Live has zero Hong Kong protest streams which are actually “live.” Queries run at 0630 am US Eastern time.
The “Police HQ Blocked” points to a “recent live stream.” That’s okay but the link appears on the YouTube Live page, and there is no live stream of the Hong Kong protest streaming live.
What? A live index pointing to an archived file.
Now contrast that with Twitch.tv, an Amazon property. I entered the query “Hong Kong protest” in the Twitch search box at this link: https://bit.ly/2sRPekp and got hits to actual live streams. Here’s a screen shot taken shortly after my visit to the YouTube Live page.
A Twitch live stream captured about 630 am US Eastern time.
The quality of the video is excellent. None of that low res stuff.
A couple of observations:
- YouTube Live is supposed to provide links to live content. Obviously YouTube does not have live video of the historic Hong Kong protests on June 21, 2019, US time zone, or YouTube chooses not to make these streams available
- Twitch.tv provides live streams of high quality from different Twitch content providers and the Twitch.tv search engine makes the content easy to find. This is a feat that mainstream US media sites cannot achieve.
- The cognitive disconnect of YouTube Live’s listing archived footage as “live” is baffling to me.
Net net: Amazon Twitch continues to provide interesting and often significant content of news value. YouTube looks increasingly arthritic when compared to the more agile Twitch service. Plus Twitch delivers high quality streams. To be fair, Amazon does display some annoying and repetitive advertisements. That’s a small price to pay for feet on the street information about activities in Hong Kong.
Twitch is focused and apparently on the steraming ball. Google is not in the game when it comes to Hong Kong’s protests.
If you were Hong Kong government authorities, which service would you use to track protest activities? Sure, the government’s camera network is a first choice, but right behind might be the Twitch.tv service. YouTube? Probably not.
Stephen E Arnold, June 21, 2019
Crashing Teslas: No Dark Web Inputs Needed
June 21, 2019
Do you want to interfere with a Tesla’s electronics? You don’t need to prowl the Regular Web for hacking forums. You don’t need to fire up Tor and visit Dark Web sites. A quick search of Bing and Google delivers a link to “Tesla Model 3 Spoofed Off the Highway.” The information in the write up is a summary of an experiment. DarkCyber understands that.
The products and methods disclosed in the write up will make it easier for some vendors of intelligent driver assistant systems to improve their products. The publicity generated by the listing of the off-the-shelf products used in this experiment will benefit the manufacturers. DarkCyber anticipates a spate of high school science fair projects which present the results of potential engineers’ experiments.
We learned in the write up:
Yonatan Zur, Regulus Cyber CEO and Co-Founder, emphasized this goes way beyond Regulus Cyber and Tesla: “We designed a product to protect vehicles from GNSS spoofing because we believe it is a real threat. We have ongoing research regarding this threat as we believe it’s an issue that needs solving. These new semi-autonomous features offered on new cars place drivers at risk and provides us with a dangerous glimpse of our future as passengers in driverless cars. By reporting and sharing incidents such as this we can ensure the autonomous technology will be safe and trustworthy. “
The purpose of the research is to inform people about vulnerabilities and to boost sales of Regulus anti-spoofing technology. The company will provide a curious person with a free research summary — in exchange for an email address.
DarkCyber is not sure whether to thank Regulus for making the vulnerabilities of autonomous technology to hacks and spoofs or whether to suggest the firm a less sensational way to publicize the results of their experiments.
If you own a Tesla or other vehicle with certain types of assist mechanisms, those systems may be vulnerable. Bad actors and frisky high school science club students are likely to do some experimenting.
Will Regulus’ research results have the sales impact the company wants? Will other impacts be observable? Have US government oversight agencies pursued similar research into security and safety issues for smart vehicles?
Tough questions to answer.
Stephen E Arnold, June 21, 2019
Have Fun Searching Nonprofit Tax Records
June 21, 2019
If you work at a nonprofit organization, the word free is magical! Databases are also a magical source of information and the life blood for anyone writing grants. A free, authoritative database is like a magic wand. ProPublica is a news source focused on nonprofits and it recently published the story about a free way to search IRS records: “You Can Now Search The Full Text Of 3 Million Nonprofit Tax Records For Free.”
Along with being a newsroom ProPublica also launched a brand new tool: the Nonprofit Explorer database that searches the full text of three million digitally filed IRS nonprofit tax filings. Nonprofit Explorer contains tax records from more than 1.8 million nonprofits as well as names for key employees and organization directors. Users can search for terms anywhere in the tax records. The only catch is the that the nonprofits needed to file their taxes digitally, but nearly two-thirds do so.
How can you use the Nonprofit Explorer:
“For one, this feature lets you find organizations that gave grants to other nonprofits. Any nonprofit that gives grants to another must list those grants on its tax forms — meaning that you can research a nonprofit’s funding by using our search. A search for “ProPublica,” for example, will bring up dozens of foundations that have given us grants to fund our reporting (as well as a few filings that reference Nonprofit Explorer itself).
Just another example: When private foundations have investments or ownership interest in for-profit companies, they have to list those on their tax filings as well. If you want to research which foundations have investments in a company like ExxonMobil, for example, you can simply search for the company name and check which organizations list it as an investment.”
Usually a database like this requires a yearly subscription. Most nonprofits cannot afford subscription fees, so ProPublica is providing a public service that will assist millions. ProPublica probably uses their own database to apply for grants to fund it.
Keep in mind that some bad actors set up non profit organizations for some interesting purposes. Access to these records may provide useful to some investigators.
Whitney Grace, June 21, 2019
Owlin Pivots Attracts Funding
June 21, 2019
Financial-tech startup Owlin is bound to be celebrating—TechCrunch announces, “Owlin, the Text and News Analytics Platform for Financial Institutions, Raises $3.5M Series A.” This is especially good news, considering the company lost ground when its original backer went bankrupt; that twist cost the company two founders, we’re told. Now, though, Velocity Capital is leading this round of funding. Writer Steve O’Hear reports:
“The fundraise follows the fintech company’s pivot from a real-time news alert service to a more comprehensive ‘AI-based’ text and news analytics platform to help financial institutions assess risk. … This is seeing Owlin enable 15,000 counter-party risk managers worldwide to track risk events that are not captured by traditional credit risk metrics. ‘We are adding news and unstructured data to their risk monitoring. In the end, our clients don’t just gain insights, they also gain time,’ adds the Owlin CEO.”
Apparently, the platform is unusually successful at augmenting certain types of data, making for more accurate risk models. Regulators love that, we’re reminded. Founded in 2012, Owlin is based in Amsterdam Some of the companies global clients are Deutsche Bank, ING, Fitch Ratings, Adyen, and KPMG.
Cynthia Murrell, June 21, 2019
Building Trust: Current Instances of Dubious Credibility
June 20, 2019
I buzzed through the overnight email and scanned the headlines dumped in my “Pay Attention” folder. Not much of interest to me. Sure, Congress is going to ask questions about the new sovereign currency from the People’s Republic of Facebook. That’s going to be a rerun of the managerial version of “So You Think You Can Dance.”
I did spot three items which make clear the ethical swamp in which some companies find themselves lost. Let’s look at these and ask, “Yeah, about that bridge to Brooklyn you sold my mother”?
ITEM ONE: Vice reports that Twitter is working on a bug fix which tells a user, “You know that person you unfollowed. Well, good news, that person is now following you.” The write up “A Nightmare Twitter Bug Is Sending Users Notifications When They’re Unfollowed” states:
For several days, untold numbers of Twitter users have been getting push notifications whenever someone unfollows them. To add insult to injury, the notifications say the user has “followed them back” when in fact the opposite is true.
Yep, a bug, not another programming error, not a failure of code QA prior to pushing the ones and zeros to a production system, not an example of a senior management team looking for fire extinguishers. Just a bug. Forget the cause, and, of course, the Twitteroids are going to fix it.
ITEM TWO: The somewhat frantic and chaotic methods of YouTube are going to more attention. “YouTube Under Federal Investigation over Allegations It Violates Children’s Privacy” reports:
A spokeswoman for YouTube, Andrea Faville, declined to comment on the FTC probe. In a statement, she emphasized that not all discussions about product changes come to fruition. “We consider lots of ideas for improving YouTube and some remain just that — ideas,” she said. “Others, we develop and launch, like our restrictions to minors live-streaming or updated hate speech policy.”
Okay, let’s clam up and face facts: The methods used to generate engagement, sell ads, and stave off the probes from Amazon Twitch are just algorithms. Once again, no human responsibility, no management oversight, and no candid statement about what the three ring video extravaganza is willing to do with regard to this long standing issue.
ITEM THREE: Facebook’s crypto currency play aside, I noted this admission that Facebook users have zero expectation of privacy, and, if I understand Facebook’s argument, you will get zero privacy from our platform. Navigate to “Facebook Under Oath: You Have No Expectation of Privacy” and note this statement:
In a San Francisco courtroom a few weeks ago, Facebook’s lawyers said the quiet part out loud: Users have no reasonable expectation of privacy. The admission came from Orin Snyder, a lawyer representing Facebook in a litigation stemming from the Cambridge Analytica scandal.
Now I am not sure this is an admission. It strikes me as a statement of a Facebook bedrock foundational principle.
What do these three current items trigger in my mind? Let me answer that question, gentle reader:
- Large, powerful high technology firms say what’s necessary to get past a problem.
- Situational decision making creates what are unmanageable business processes.
- The senior managers and spokes humans are happy to perform just like the talent on “So You think You Can Dance.”
For me, that show show is becoming tiresome, repetitive, and the intellectual equivalent of chowing down on KryspyKreme chocolate- iced, glazed-with-sprinkles donuts. The music is getting louder, and the tune is “Deflect, aplogize, keep goin’.” Boring.
Stephen E Arnold, June 20, 2019
Google Has Changed Search Results Again
June 20, 2019
Google cannot let anything rest. Whether that is a good thing or a bad thing is a matter of opinion, to others it turns into annoyance. Has Google’s latest changes to search results confounded its users once again? Inc. looks into Google’s newest change in the article, “Google Just Announced A Major Change To Your Search Results.”
The new change appears to be a simple one. Google will no longer show multiple results from the same Web site, except occasionally. What is even more uncanny is that Google deployed it right under our noses. The search engine did not change anything in the search algorithm, however, this could be bad for content creators and businesses centered on content curation.
More unique Web sites will be introduced into the top search results, most people click on the first few results displayed. What this means for content people is that:
“A lot of you spend a lot of resources creating content for the very purpose of showing up at the top of organic search results. This change means that you’ll have to consider how to adjust your content and search engine optimization strategy knowing that less of your pages will show up for relevant searches…Google says there is an exception. When the company’s search algorithm thinks that more than one result from the same site is especially relevant to a search, it will continue to display them in the top results. It doesn’t give specific examples as to what this means or when it will make this exception, but Google does say they will continue to make adjustments as it rolls out across search results.”
Google fiddles the ad giant engine has made assurances that multiple results from the same page will display they are the most pertinent. As a librarian, perhaps some of these changes will restore relevance and add a pinch of precision and recall?
Whitney Grace, June 20, 2019
Grammar Rules Help Algorithms Grasp Language
June 20, 2019
Researchers at several universities have teamed up with IBM to teach algorithms some subtleties of language. VentureBeat reports, “IBM, MIT, and Harvard’s AI Uses Grammar Rules to Catch Linguistic Nuances of U.S. English.” Writer Kyle Wiggers links to the two resulting research papers, noting the research was to be presented at the recent North American Chapter of the Association for Computational Linguistics conference. We learn:
“The IBM team, along with scientists from MIT, Harvard, the University of California, Carnegie Mellon University, and Kyoto University, devised a tool set to suss out grammar-aware AI models’ linguistic prowess. As the coauthors explain, one model in question was trained on a sentence structure called recurrent neural network grammars, or RNNGs, that imbued it with basic grammar knowledge. The RNNG model and similar models with little-to-no grammar training were fed sentences with good, bad, or ambiguous syntax. The AI systems assigned probabilities to each word, such that in grammatically ‘off’ sentences, low-probability words appeared in the place of high-probability words. These were used to measure surprisal[sic]. The coauthors found that the RNNG system consistently performed better than systems trained on little-to-no grammar using a fraction of the data, and that it could comprehend ‘fairly sophisticated’ rules.”
See the write-up for a few details about those rules, or check out the research papers for more information (links above). This is but a start for their model, the team cautions, for the work must be validated on larger data sets. Still, they believe, this project represents a noteworthy milestone.
Cynthia Murrell, June 20, 2019
Facebook: Amazon Is Shifting Gears and May Be Grinding Toward Social Media Land
June 19, 2019
The zippy 2019 news cycle is writing about Facebook: Its digital currency, a thriller for those working to enforce assorted rules and regulations about money. Facebook’s trust issues percolate in the stories as well; for example, “Facebook’s Crypto currency Has a Trust Problem,” which explains — well, obviously — Facebook’s method of saying one thing and just chugging along mostly doing what it wants to do. One must not overlook the legal tussles like the Cambridge Analytica matter; for example, “The Cambridge Analytica Debacle: A Legal Primer.” Exciting stuff.
There was an announcement which the DarkCyber team noted; specifically, “Game Streaming Site Twitch Buys Social Network Bebo.” Yawn. Bebo, a social media service founded in 2005. That’s so yesterday. Bebo pivoted to social apps, but that did not work out as planned. Then Bebo tried a hashtag messaging app. The idea was that a message had content and it had user assigned index terms just like Twitter. More recently, Bebo has dabbled in young people playing games. Think intramural sports with games like Fortnite. Bowling leagues for people who prefer digital games to those which can result in two a days, bruises, and rides on a team bus.
Wikipedia provides more details of the Bebo trajectory. The reports about this deal like “Amazon’s Twitch Acquired Social Networking Platform Bebo for up to $25 million to Bolster Its Esports Efforts” hit the basics:
- Twitch will be buying pizzas for the Bebo team
- Amazon paid an alleged $25 million for the Bebo property (fungible and intellectual property like Monkey Inferno)
- Hope for the future.
DarkCyber’s view of the deal is mostly in line with the publicly available news reports. However, Amazon has access to data about Twitch, including outputs from users who exchange messages, inputs from “creators” or “live streamers” who want special features without having to arm wrestle with third party software, and Amazon’s own big thinkers who understand that “games” are not part of the fabric of outfits like Amazon, Facebook, and — are you ready for it? — Netflix.
There are also implications for intelligence and law enforcement and, of course, for Facebook, a digital country with its own fledgling sovereign currency. The Bezos bulldozer might be making tracks for Palo Alto to redevelop a certain billionaire’s compound.
Stephen E Arnold, June 19, 2019
Factualities for June 19, 2019
June 19, 2019
I must congratulate the purveyors of “real” news. Even though my screen time was limited for various, unexciting reasons, I did spot some fantastic numbers. Herewith I present some of the more fascinating data generated by the great minds fiddling with Excel and soon Salesforce with Tableau analytics. I am looking forward to those numerical confections. Let’s go, Salesforce.
-20. The percent decrease in cash flow if a British company does not implement artificial intelligence. (Note: There was no definition of artificial intelligence in the write which makes the number more special.) Source: Telegraph
10. The percentage of graphic liners in a UK nuclear plant reactor which have cracked; that is, failed. Source: BBC
13. Percentage of UK adults who trust big tech firms handling health data. Source: Techerati
18. The number of minutes it takes a New York Times’ reporter to read Facebook’s privacy policy. Source: New York Times
46. Age in years of the Department of Education’s computer system. Source: GAO
60. Percentage of “meat” which will come in 2040 from laboratories, not formerly alive animals. Source: Big Think
67. The percentage of UK doctors who find patient feedback negative. Source: Health Care IT News
94.5. Percentage of software vulnerabilities not exploited by bad actors. Source: ZDNet
2,300 to 12,594. Number of nitrite attributable cancers, possibly due to US drinking water. Note: Maybe a year, maybe more time. Source: Science Direct
$10,000. Cost of repairing a MacBook Pro screen which was not broken. (Software setting issue). Source: The Register
187,000. The number of users from whom Facebook collected data via an app Apple banned. Source: Techcrunch
7.7 million. Number of Lab Corp customers who experienced data loss due to the collection firm’s failed security system. Source: Krebs
$39 million. The cost of a gallon of scorpion venom healing compound. Source: Stanford
2.4 billion. Number of online game players worldwide. Source: Vox
$47 billion. The amount of money Google earned from news. Source: Free Press Journal
$52 million. Cost of a trip to the International Space Station. Source: Wired
Stephen E Arnold, June 19, 2019
Alphabet Google: Reality Versus Research in Actual Management Activities
June 19, 2019
For a few months, I have been using my Woodruff High School Science Club as a source of ideas for understanding Silicon Valley management decisions. I termed my method HSSCMM or “high school science club management method.” A number of people have told me that my approach was humorous. I suppose it is. One former colleague from a big name consulting name observed that I was making official, MBA-endorsed techniques look like a shanked drive at a fraternity reunion golf scramble. (MBA students seem to be figuring out that their business degrees may open doors at Lyft or Uber, not McKinsey & Company.
Where the HSSCMM differs from “situational thinking without context” (this is DarkCyber jargon), Google research has identified best practices for management. However, HSSCMM is intuitive and easy to explain. My touchstone for management appears in the article “Google Tried to Prove Managers Don’t Matter. Instead, It Discovered 10 Traits of the Very Best Ones.” Google’s original goal involved figuring out if a sports team manager was important or not. Google’s brilliant analysts crunched numbers and found that coaches matter. The best ones shared some data-backed characteristics. Let’s compare what Google found with the HSSCMM.
I made an a MBA-influenced table to keep thoughts clear.
| # | Google Research Says | HSSCMM Approach | Observations |
| 1 | Be a good coach | Be arrogant because you understand differential equations | Google is working on discrimination |
| 2 | Empower | Others are stupid | The smartest person is in charge |
| 3 | Inclusive team | Exclusivity all the way. | Google hires best talent, and Google defines “best” |
| 4 | Be results oriented | Do what you want. Outsiders don’t get it. | Boost ad sales |
| 5 | Communicate | Don’t get it? You’re fired. | Explain YouTube is too big to be fixed |
| 6 | Have a strategy | React. Ignore the uninformed | Make quick decisions like buying Motorola |
| 7 | Support career development | Learn it yourself | Find a team or leave |
| 8 | Advise the team | Figure it out, or you don’t belong | First day at work confusing? Try flipping burgers |
| 9 | Collaborate | Work alone | Fix the problem or quit |
| 10 | Be a strong decision maker | Do what I say, dummy | Obvious, right? |
Answer this question: How many of the characteristics from each column match actions from Silicon Valley-type companies like Amazon, Apple, Facebook, Google, etc.?
 |
| Which type of management method is exemplified in this allegedly true incident? The Google management research findings or the high school science club management methods? Answer: HSSCMM. |
As you formulate this answer, consider the decision making evidenced in this allegedly accurate article from 2017 about a Silicon Valley executive.
Stephen E Arnold, June 19, 2019
-
- Subscribe to Beyond Search
Feature archive
News archive
-
