MMA Announces Formation of Mobile Location Data Accuracy Group To Drive Trust and Scale in Location Marketing | MMA

MMA Announces Formation of Mobile Location Data Accuracy Group To Drive Trust and Scale in Location Marketing; press release; Mobile Marketing Association (MMA); 2014-07-30.
Teaser: Specialist Group to Develop Industry Guidelines and Standards

Mentions

  • Location Leadership Council
    comprises

    • Location Committee
    • Privacy Committee
  • Location Terminology Guide; Mobile Marketing Association (MMA); 2013; 24 pages; landing; regwalled (pay with PII); separatedly noted.
  • Mobile Location Data Accuracy Group
  • Media Ratings Council (MRC)

Accuracies

order desc

  • Indoor positioning system (?)
  • Bluetooth
  • GPS
  • Wi-Fi hotspots
  • Wi-Fi triangulation
  • Cell tower triangulation
  • IP Address
  • Consumer self-attestation

Participants

Company Who Title Role
xAd Monica Ho SVP Marketing co-chair
Joule Michael Lieberman CEO co-chair
Factual Vikas Gupta Director of Marketing
ThinkNear Brett Kohn Director of Marketing
Mobiquity James Meckley CMO
Nielsen Tom Eaton VP, Client Services
MEC Bav Panchal Mobile
OpenX Rob Kramer GM, Mobile
Ubimo Ran Ben-Yair CEO
Verve James Smith CRO

Linda Programming Model named after Linda Lovelace, contra Ada after Ada Augusta Lovelace

Linda is named for Linda Lovelace

Referenced

1. David Gelernter’s Romance With Linda; John Markoff; In The New York Times (NYT); 1992-01-19.

<quote>

As generations of college professors can attest, sometimes there is no accounting for graduate student humor. And that explains how Linda, the parallel processing language, got its name.

When David Gelernter began his doctoral studies at the State University of New York in Stony Brook in 1978, the major parallel processing programming language was named Ada. It had been developed by the Defense Department, and Mr. Gelernter considered it inelegant and bulky.

He developed his new language around a set of six simple additions that could be made to existing computer languages — Fortran, for example — so that software authors would not have to start from scratch learning a new language if they wanted to use parallel processing.

But Mr. Gelernter’s new language might have been little more than a graduate student thesis if he had not met Nick Carriero, who arrived at Stony Brook to begin his graduate studies just as Mr. Gelernter was finishing up. Both men became captivated with the potential of parallel computing and Mr. Carriero followed his mentor to Yale University and became Mr. Gelernter’s first graduate student.

When it came time to name the language, Mr. Gelernter said he noted that Ada was named after Ada Augusta Lovelace, the daughter of Lord Byron, the English poet. Miss Lovelace is regarded as the first computer programmer because she worked for the computer pioneer Charles Babbage.

Another woman named Lovelace was in the news when Mr. Gelernter was casting about for a name — Linda Lovelace, a star of pornographic films. So he named the language Linda and it stuck. Asked about it now, Mr. Gelernter grins and shrugs. “I was a graduate student at the time,” he said.

</quote>

2. David Gelertner, In Jimi Wales Wiki.

<quote>In the 1980s, he made seminal contributions to the field of parallel computation, specifically the tuple space coordination model, as embodied by the Linda programming system (named for Linda Lovelace, an actress in the porn movie Deep Throat, mocking Ada‘s tribute to Ada Lovelace).[3] Bill Joy cites Linda as the inspiration for many elements of JavaSpaces and Jini.[4] </quote>

3. Linda (coordination language), In Jimi Wales Wiki.

<quote>Linda is named after Linda Lovelace, an actress in the porn movie Deep Throat, a pun on Ada‘s tribute to Ada Lovelace.[4]</quote>

One Trillion Edges: Graph Processing at Facebook Scale | Ching, Edunov, Kabiljo, Logothetis, Muthukrishnan

Avery Ching, Sergey Edunov, Maja Kabiljo, Dionysios Logothetis, Sambavi Muthukrishnan (Facebook); One Trillion Edges: Graph Processing at Facebook Scale; In Proceedings of the Conference on Very Large Data Bases (VLDB); 2015-09-04; 12 pages.

Abstract

Analyzing large graphs provides valuable insights for social networking and web companies in content ranking and recommendations. While numerous graph processing systems have been developed and evaluated on available benchmark graphs of up to 6.6B edges, they often face significant difficulties in scaling to much larger graphs. Industry graphs can be two orders of magnitude larger – hundreds of billions or up to one trillion edges. In addition to scalability challenges, real world applications often require much more complex graph processing workflows than previously evaluated. In this paper, we describe the usability, performance, and scalability improvements we made to Apache Giraph, an open-source graph processing system, in order to use it on Facebook-scale graphs of up to one trillion edges. We also describe several key extensions to the original Pregel model that make it possible to develop a broader range of production graph applications and workflows as well as improve code reuse. Finally, we report on real-world operations as well as performance characteristics of several large-scale production applications.

Previously

References

46 references

The messy battle to protect your data from your own Internet provider | The Washington Post

The messy battle to protect your data from your own Internet provider; Brian Fung; In The Washington Post; 2015-08-20.

Mentions

  • Consumer surveillance practices
    • AT&T
    • Comcast
    • Verizon
  • AT&T charges $29 more for gigabit fiber that doesn’t watch your Web browsing; ; In Ars Technica; 2015-02; previously filed.
    Teaser: AT&T goes head to head against Google in KC on fiber and targeted ads
  • Tom Wheeler, Chairman, Federal Trade Commission (FTC)
  • Mignon Clyburn (D), Commissioner, Federal Trade Commission (FTC)
  • Michael O’Rielly (R), Commissioner, Federal Trade Commission (FTC)
  • Customer Proprietary Network Information (CPNI)
  • Section 222 of the Communications Act defines Customer Proprietary Network Information (CPNI)
  • Jurisdiction
    • Federal Trade Commission (FTC) regulates trade
    • Federal Communications Commission (FCC) regulates communication
  • Something about <quote>FCC’s first-ever case against a telecom company for failing to prevent a data breach, leading to a $10 million fine for YourTel and TerraCom, </quote>
  • Event
    • A conference hosted by the Technology Policy Institute in Aspen, Colo.
  • Quoted
    for color, background & verisimilitude

    • Tom Wheeler, Chairman, Federal Trade Commission (FTC)
    • Michael O’Rielly (R), Commissioner, Federal Trade Commission (FTC)
    • David Redl, a Republican policy staff, House Energy and Commerce Committee.

Will Content Blocking push Apple into advertising’s wheat business? | Doc Searls

Doc Searls; Will Content Blocking push Apple into advertising’s wheat business?; In His Blog; 2015-08-29.

tl;dr → ads are bad; Apple iAds exists; Apple launches Ad Block in iOS9; they are hypocrites.

Mentions

  • Links to his own previous opinements
  • Links to trade press on iAds, etc., ordered by date, in chronological order.

Previously

Why Daniel Gelertner, Dittach, Is Not Looking to Hire Computer-Science Majors

Why I’m Not Looking to Hire Computer-Science Majors; Daniel Gelertner; In The Wall Street Journal (WSJ); 2015-08-28.

Lost me at “hegemon.” That’s a $10 word from academe where the departments =~ m/\bstudies$/i; and they use “privilege” as a verb in an argot. Without irony. Fancy. As well, even credulous Millennials will see the author’s endeavor, as stated, isn’t a business so much as a checkbox on a subpanel on a stock-standard expected feature of the MUA. Those Millies are wily, they can smell this; they want to work on World Changing Ideas, like taxi hailing, laundry scheduling or pizza delivery.

Background

  • Daniel Gelertner
    • Dittach, CEO, 2014-08→now
    • Lifestreams, bizdev, 2012→now (concurrent with Dittach), licensee of Mirror Worlds
    • Managed Interface Tech, engineer?, 2009-11→2011-01.
    • University of the Republic, designer, 2007-04→2008-06.
    • Yale, Class ’10, B.A. Music
  • Dittach, CrunchBase
    <quote spelling=”theirs”>Dittach is a moblie app that syncs with you emaill account making finiding attachments much easier and more efficiant.</quote>
  • Daniel Gelertner (son) vs David Gelertner (father)
  • el’s previous & current endeavor was LifeStreams of Mirror Worlds
  • Computer Scientist David Gelernter; Alexandra Wolfe; In The Wall Street Journal (WSJ); 2013-11-29.
    Teaser: Can the computer scientist help make computers more intuitive?
  • Mirror Worlds, In Jimi Wales Wiki

The Mobile Video Ad Lie | Rob Leathern

Rob Leathern; The Mobile Video Ad Lie; In Medium; 2015-08-27.
Teaser: NYPost.com site loads 10Mb on iPhone, with no videos or video ads to be seen
Rob Leathern is ex-CEO, Optimal

tl;dr → adtradesman observes that mobile web pages loads slowly, because adtech. QA travel log follows.  Ad blocking on iPhone iOS9 is mentioned; ad fraud is insinuated.

Mentions

  • Charles Proxy
    payware HTTP proxy for Windows, MacOS, Linux & iOS (iPhone)
  • Cookie Syncing between N2 partners is documented LOG
  • Statistics
    • ~900 HTTP calls
    • 10MB of componentry
  • Ghostery
  • Brian DeFrancessco
    • CEO, LKQD
    • ex-Vindico
    • ex-Specific Media
  • Viewability
  • Ad Fraud
  • autoplay

Partners

  • 495communications
  • AdapTV
  • Branovate
  • LiveRail
  • lkqd
  • Marimedia
  • outfox
  • TinyLove
  • fusevid.com
  • LKQD

Domains

In order of appearance within LOG.

  1. olark.com
  2. rfihub.com
  3. crwdcntrl.net
  4. smartclip.net
  5. turn.com
  6. nypost.com
  7. adadvisor.net
  8. invitemedia.com
  9. nypost.com
  10. advertising.com
  11. nypost.com
  12. rubiconproject.com
  13. stickyadstv.com
  14. bidtheatre.com
  15. innovid.com
  16. googleapis.com
  17. tinypass.com
  18. virool.com
  19. krxd.net
  20. amgdgt.com
  21. mxptint.net
  22. yume.com
  23. scorecardresearch.com
  24. crwdcntrl.net
  25. rubiconproject.com
  26. krxd.net
  27. doubleclick.net
  28. adform.net
  29. chango.com
  30. liverail.com
  31. fusevid.com
  32. krxd.net
  33. rpxnow.com
  34. adgrx.com
  35. dpclk.com
  36. eyereturn.com
  37. doubleclick.net
  38. comodoca.com
  39. meltdsp.com
  40. gstatic.com
  41. krxd.net
  42. chango.com
  43. runadtag.com
  44. turn.com
  45. cloudfront.net
  46. de17a.com
  47. advg.jp
  48. springserve.com
  49. demdex.net
  50. bing.com
  51. adition.com
  52. scanscout.com
  53. videohub.tv
  54. innovid.com
  55. nexac.com
  56. btrll.com
  57. criteo.com
  58. nypost.com
  59. a3cloud.net
  60. adnxs.com
  61. media6degrees.com
  62. rlcdn.com
  63. fusevid.com
  64. imrk.net
  65. moatads.com
  66. visiblemeasures.com
  67. yume.com
  68. liverail.com
  69. netmng.com
  70. extend.tv
  71. exelator.com
  72. amazonaws.com
  73. exelator.com
  74. nexac.com
  75. ru4.com
  76. adsrvr.org
  77. rtbidder.net
  78. rundsp.com
  79. fusevid.com
  80. dyntrk.com
  81. navdmp.com
  82. rubiconproject.com
  83. adsymptotic.com
  84. rfihub.com
  85. googlesyndication.com
  86. polarmobile.com
  87. phluidmedia.net
  88. chartbeat.net
  89. btrll.com
  90. revsci.net
  91. quantserve.com
  92. adap.tv
  93. quantserve.com
  94. sitescout.com
  95. tapad.com
  96. w55c.net
  97. yahoo.com
  98. rundsp.com
  99. eyeota.net
  100. doubleclick.net
  101. owneriq.net
  102. net.ua
  103. adap.tv
  104. rlcdn.com
  105. gwallet.com
  106. virool.com
  107. metrigo.com
  108. tubemogul.com
  109. innovid.com
  110. ixiaa.com
  111. opendsp.com
  112. a3cloud.net
  113. adap.tv
  114. advertising.com
  115. vindicosuite.com
  116. innovid.com
  117. advertising.com
  118. parsely.com
  119. nypost.com
  120. virool.com
  121. chartbeat.com
  122. parsely.com
  123. adotube.com
  124. unrulymedia.com
  125. addthis.com
  126. angsrvr.com
  127. active-agent.com
  128. adap.tv
  129. advertising.com
  130. mathtag.com
  131. spotxchange.com
  132. tidaltv.com
  133. lkqd.net
  134. liverail.com
  135. atwola.com
  136. yieldoptimizer.com
  137. bluekai.com
  138. rubiconproject.com
  139. unrulymedia.com
  140. eyeviewads.com
  141. nypost.com
  142. gradientx.net
  143. simpli.fi
  144. videoamp.com
  145. yashi.com
  146. moatads.com
  147. nypost.com
  148. vindicosuite.com
  149. unrulymedia.com
  150. liverail.com
  151. google-analytics.com
  152. bidswitch.net
  153. vindicosuite.com
  154. crwdcntrl.net
  155. chartbeat.com
  156. googleapis.com
  157. cdn77.org
  158. walmart.com
  159. tinypass.com
  160. doubleclick.net
  161. audienceiq.com
  162. doubleclick.net
  163. rfihub.com
  164. icloud.com
  165. chartbeat.net
  166. quantserve.com
  167. casalemedia.com
  168. adap.tv
  169. advertising.com
  170. unrulymedia.com
  171. googlesyndication.com
  172. virool.com
  173. posumeads.com
  174. amazonaws.com
  175. gstatic.com
  176. bidswitch.net

Placed opines that Location Accuracy on Ad Exchanges is Bad | Ad Exchanger

ocation Accuracy: Ad Exchange Data Falls Short; Marcus D. Collins (Placed); In Ad Exchanger; 2015-08-27..
Marcus D. Collins, staff, Placed.

Mentions

  • Placed runs an opt-in panel of confederates:
    • 1,000 points/day
    • <vague>1 in 450 adults empaneled</vague → 309M-persons-US÷450 → 680,000 persons-US empaneled
    • has participants with high-frequency or persistent location traill contributions.
  • Placed surveyed RTB data
    • 30M high-accuracy location points (post-funnel)
  • Definition: high-accuracy bid request locations, via the Procedure, below.
    • sub-minute frequency
    • multi-minute repetition bursts
    • and meter-level accuracy.
  • Claims
    1. low-frequency observations of targets makes inference of intent impossible
    2. almost all? businesses do not generate quality locations on exchanges.
  • Insights
    <quote>

    • <vague>The overwhelming majority</vague> of real-world business visits don’t generate [useable] locations on the exchanges.
    • ~73% of bid requests occur in transit between two locations.
    • The targets can only be observed moving between venues, rarely appearing to sojourn at a venue.
    • 91% of business visits <snip/> generated no ad requests at all.
    • ≤1% of the location data from ad exchanges is accurate enough<snip/></quote>
  • Stories
    • the target walks past the coffee shop; the target does not present itself into the coffee shop, instead visits a market which is around the corner.
    • the target does not use the phone while in the shop, it is busy conducting e.g. commerce, entertainment, eating.

Procedure

  • 4B RTB imp/day, fractionally with (Lat,Lon)
  • <vague>only a fraction</vague> corresponds to real-world human locations and behavior.

Unclear how to disentangle his prose into a waterfall. This may be what he performed.  This does not match the funnel depicted in the nearby image.

  1. filtering out imprecise locations
  2. Given Line 1, with precise locations
    drop 90%, 10% remaining

    • 90% of remaining locations were “overrepresented” and couldn’t be trusted.
    • Example: 10% of Bay Area bid requests had a single location in San Jose, CA.
  3. Given Line 2, precise with reasonable representation
    drop 10%, 90% remaining

    • are inhuman (but not bots)
    • too random
    • not random enough
  4. Given Line 3, precise, with reasonable location, not inhuman
    drop 67%, keep 33%

    • Bots accounted for roughly 67% of remaining locations
    • Observe: <vague>up to several thousand</quote> impressions per hour for individual users.
  5. Final, after Line 4, reliable locations from reliable devices
    • 30 million location points
    • 5 million devices
    • 30 locations  day × 6 days/month
    • Define: such points as high-accuracy bid request locations.

Actualities