Google: Redefines Quality. And What about Ads?
August 23, 2022
When I was working on The Google Legacy (Infonortics, 2004), I gathered information about Google’s method for determining quality. Prior to 2006, Google defined “quality” in a way different from the approach taken at professional indexing and commercial database companies. Professional organizations relied on subject matter experts’ views. Some firms — for example, the Courier Journal & Louisville Times, Predicasts, Engineering Index, the American Petroleum Institute, among others — were old fashioned. Commercial database firms with positive cash flows would hire specialists to provide ideas and suggestions for improving content selection and indexing. At the Courier Journal, we relied on Betty Eddison and a number of other professionals. We also hired honest-to-goodness people with advanced degrees to work on the content we produced.
Google pops up with jibber jabber about voting, a concept floated by an IBM Almaden researcher, and the notion of links and their value. As Google evolved, I collected a list of what amount4ed to 140 or so factors which were used by Google to determine the quality of content. At one time, Dr. Liz Liddy used my compilation as illustrative material for her classes in information science.
By 2006, Google shifted quality from its mysterious and somewhat orthogonal factors to what I call “ad quality.” The concept gained steam when Google acquired Applied Semantics and worked hard to relax a user’s query, match the query to a stack of ads to which the query would relate, and display these as “personalized” and targeted messages. Quality, therefore, became an automated process for working through ad revenue.
Since 2006, Google has been focused on ad revenue. My personal view is that Google has one stream of revenue: Ad revenue. Its other ventures have not demonstrated to me that the company can match its first “me too” innovation. If you don’t remember what that was, think about the Yahoo settlement related to the “inspiration” Google obtained from the GoTo.com and Overture “pay to play” system. The idea was that those with Web pages would pay to get their message in front of a service’s users.
Where is Google quality now? Is it anchored in editorial policies, old fashioned ideas like precision and recall? Is the Google using controlled vocabulary lists designed to allow precise queries? Is Google adding classification codes to disambiguate terms like terminal as in “computer terminal” or “airport terminal”?
“Google’s Planned Search Changes Could Upend the Internet” reveals:
Google is trying to improve the quality of search results and reduce the number of misleading sites, misinformation, and clickbait users are subjected to.
I want to point out that the lack of precision and recall in Google’s approach is the firm’s notion that new Web sites are more important than older Web sites, traffic is more important than factual accuracy, and ad revenue goals are the strong force in the Google datasphere.
Thus, after a certain outfit headed by a search engine optimization crazed advanced the SEO “revolution”, the Google is, according the article:
As part of the change, the company will roll out its “helpful content update” to identify content that is primarily written to rank well in search engines and lower its rank. Sullivan says the update seems to especially benefit searches related to tech, online education, shopping, arts, and entertainment. The company is also working to improve access to high-quality reviews, ones that provide helpful, in-depth information.
Does this suggest that Google will focus on high-value content, explicit editorial policies, and professional indexing by subject matter experts?
Nope.
It means quicker depletion of the ad inventory and an effort to cope with the fact that those in middle school and high school use TikTok for information.
Google is officially a dinobaby just one not very good at anything other than selling ads and steering its coal fired steam boat away from the rapids in today’s data flows. For serious information research Google is too consumer oriented. Search based applications are what some researchers prefer. The content in these systems comes from specialized crawls and collections.
The quality list? Old fashioned and antiquated. How much of Google fits in that category? SAIL on, steam boat. Chug chug chug. PR PR PR. Toot toot.
But what about traffic to sites affected by Google’s content rigor?
Just buy ads, of course.
Stephen E Arnold, August 23, 2022
Google and the Thirst for Quantum Supremacy
August 17, 2022
The operative word is “supremacy.” For a very basic reason — money — Google wants to be the source of low cost training data, integrated models, and assorted software to own the fabled quantum computing sector. You (and probably the management and technical wizards) at the online advertising firm may disagree. That is okay with me. Since I wrote the Google Legacy in 2006 or so, Google has been a bifurcated outfit.
On one hand, the Google had oodles of money from its bar-room swinging door business model. In may view, the approach means that for some transactions, the Google gets paid with each swing. Ka-ching, ka-ching, ka-ching 24×7. Great for high school reunion talk and for the estimable company; maybe not so great for the advertisers. How does one add a gloss of intellectual excellence to this core business?
Easy. Solve “real world” technical problems; for example, make death go away, allow autos to drive themselves (sort of), and master the future world of quantum computing or quantum stuff from pharma drug “discovery” to inventing new materials with a short ride on the bioengineering bypass.
However, doing these innovative things is hard. Writing papers, giving talks, and writing blog posts is a little easier. “DeepMind Feuds with Russian Scientists over Quantum AI Research” puts Google in a defensive posture. That’s uncomfortable for today’s bastion of management excellence.
To further irritate the Google, academic researchers from a couple of countries like
Here’s the passage I circled because it is important and pretty easy to understand by a dinobaby like me:
According to the commenting researchers, the training process that DeepMind used to build its neural network taught it how to memorize the answers to the specific problems it was going to face during benchmarking — the process by which scientists determine if one approach is better than another.
Since the Google cannot fire the academics beavering away in Russia and South Korea, Google denies the allegation that its snorkeling-like approach to creating training data suffers from a flaw. Google wants, as I understand the goal, to become the Big Dog of all things quantum. Supremacy is the goal, and some academics criticizing the intellectual skyscrapers erected by Google is not permitted. In fact, if the Russian and South Korean academics are a little bit right, the claim to quantum supremacy and all-thingns-quantum is tilting like the Millennium Tower in San Francisco.
Net net: My view is that quantum research is chugging along at the turtle-racing pace of other next-big things. Google’s need to be the leader in something substantive and not yet directly associated with the worshipful business of online ad sales is growing. Thus, one thing is clear: None of the researchers has demonstrated much in the way of Googley behavior. If their analysis is correct, what will be the value of a digital Millennium Tower?
Stephen E Arnold, August 17, 2022
High School Science Club Management Goes Man in the Gray Flannel Suit
August 17, 2022
I read the stories about Facebook and Google trying to manage their paid humanoids. Both companies, not surprisingly, are pulling tips from the “Universal Guide to Running a High School Science Club” and its Annex 1: Never Do These Things. The two estimable companies skipped the Annex. Why read something at the back of a user manual. That’s for those who are smart, just not brilliant.
Among the tips in my copy of the Universal Guide was this one: “Never tell a fellow science club member to work harder.”
Another precept was: “Never tell a fellow science club member to quit if the alleged humanoid did not like what the president told them to do.
Both Facebook and Google appear to have pushed to the “work harder” and “go away” approach. Brilliant, right?
Even the Silicon Valley type of “real” news outfit Protocol published an article focusing on this management approach. “Don’t Be Meta or Google: How to Tell Workers They Need to Be More Productive” has some management advice for the fellow travelers; to wit:
the idea that underperforming individuals are solely responsible for their companies’ large-scale financial troubles is probably inaccurate, and you don’t want your productivity pep talk to give that impression. Launching a companywide campaign to improve productivity is absolutely reasonable, as long as you’re not alienating employees in the process.
Yes, Harvard Business School, here we come!
I am not sure what’s crazier: The management methods of the high school science club or the faux-Drucker inputs from a “real” news Silicon Valley type online publication.
The write up adds:
Sharing a specific game plan to improve productivity is key to avoiding chaos.
Yes, is the corollary “sharing is caring”?
That method was not part of the Woodruff High School Science Club in Central Illinois. My fellow members believed themselves to be budding wizards. One of the best and brightest had his first date and ran the train signal. The train won. Not a best nor brightest moment as I recall.
“Management” was, in my opinion, a no show at some of the zippy Silicon Valley outfits for which I labored until I threw in the dead fish in 2013. The idea that the methods of a high school science club would contribute to management science would have been laughable about a decade ago. Now that Facebook and Google type outfits have to manage, the adolescent guidelines of the unread Annex seem oddly appropriate.
Had Google solved death, Mr. Drucker would be available to provide some management guidance to the “real news” and the Facebooks and Googles of the world. I am not sure “don’ts” work… at all.
Stephen E Arnold, August 17, 2022
YouTube: Some Proof about Unfindable Content
August 17, 2022
I read “5 Sites to Discover the Best YouTube Channels and Creators Recommended for You.” The write up presents five services which make YouTube content “findable.” What I learned from the article is that YouTube videos are, for the most part, unfindable. A YouTuber can stumble upon a particular video and rely on Google’s unusual recommendation system. In my experience, that system is hobbled by its assorted filters and ad-magnetic methods. If I want to locate a video by eSysman (a fellow who reports about big money yachts loved by some money launderers and oligarchs), Google refers me to NautiStyles, YachtsForSale (quite a sales person is visible on that channel), or the flavor of the day like Bering Yachts. eSysman is the inspiration for one former CIA professional, and her edging into the value of open source intelligence. Does Google’s algorithm “sense” this? Nah, not a clue. What if I want some downhome cookin’ with Cowboy Kent, the chuck wagon totin’, trail hand feedin’ Oklahoma chef. Sorry, promoting Italian chefs are not what I was looking for. Cowboy cookin’ is not Italian restaurateurs showing that their skills are sharper than fry cooks in French restaurants. But what about YouTube search? Yes, isn’t it fantastic? Enough said.
What about the services identified in the article? Each offers different ways to find a video or channel on a specific or semi-specific topic. You can navigate to the source document and work your way through the list of curated “finder” sites.
The write up points out:
YouTube has over 50 million channels, but as you might have guessed, most of them aren’t worth subscribing to.
That’s the type of “oh, well, don’t worry statement” that drives me bonkers. Just let someone tell you what’s good. Go with it. Hey, no problemo. Who wants to consider the implications of hours of video uploaded every minute or the fact that there are 50 million channels from the Googlers’ service.
Several observations:
- No one knows what is on YouTube. I have some doubts that filters designed to eliminate certain types of content work particularly well. The idea that the Google screens each and every uploaded video with tools constantly updated to keep track of possibly improper videos is interesting to contemplate. Since no one knows what videos contain, how can one know what’s filtered, allowed in mistakenly, blocked inadvertently, or processed using methods not revealed to the public. (Lists of user “handles” can be quite useful for some purposes.)
- Are the channels no one can find actually worthless? I am not too sure. There are channels which present information about how to game the Google algorithm posted by alleged Google “partners.” I engaged in a dialogue with this “professional” and found the exchange quite disturbing. I located the huckster by accident, and I can guarantee that keeping track of this individual is not an easy task. Is that a task a Googler will undertake? Yeah, sure.
- YouTube search is one of the many “flavors” of information location the company offers. In my experience, none of the Google search services works very well or delivers on point information without frustration. Does this comment apply to Google Patent search? Yep. What about Google News search? Yep yep. What about regular Google search for company using a common word for its name? Yep yep yep. (Google doesn’t have a clue about a company field code, but it sure pushes ads unrelated to anything I search. I love mindless ads for the non-US content surveillance products that help me express myself clearly. Hey, no I won’t buy.)
Net net: YouTube’s utility is designed for Google ads. The murky methods used to filter content and the poor search and recommender systems illustrate why professional libraries and specific indexing guidelines were developed. Google, of course, thinks that type of dinobaby thinking is not hip.
Yes, it is. Unless Google tames the YouTube, the edifice could fall down. TikTok (which has zero effective search) may just knock a wall or trellis in the YouTube garden over. Google wants to be an avant guard non text giant. Even giants have vulnerable points. The article makes clear that third parties cannot do much to make information findable in YouTube. But in a TikTok world, who cares? Advertisers? Google stakeholders? Those who believe Google’s smart software is alive? I go for the software is alive crowd.
Stephen E Arnold, August 17, 2022
Google: If True, This Is a Management Moment
August 15, 2022
I read tabloids when I ride the subway in New York, Paris, or London. Nope, to Madrid’s and Moscow’s undergrounds. No thanks.
I spotted a New York Post article called “Google Execs Threaten Workers with Layoffs: There Will Be Blood on the Streets.” Wow, blood on the streets of Mountain View. Pretty exciting. I also liked the “threaten” idea. What an outstanding and sensitive management move… if accurate.
Let’s assume, just for chuckles, that the article is accurate.
I learned:
Google executives are telling their employees to shape up or ship out, warning that layoffs are coming if results don’t meet expectations. Employees who work in the Google Cloud sales department said that senior leadership told them that there will be an “overall examination of sales productivity and productivity in general.”
And, pray tell, are those expectations. More chat applications? How about some big time acquisitions that go exactly where? Motorola, hello, hello. What about a high lift project like solving death? Oh, sorry. Been there and failed at that. Maybe mending fences in Australia? Yeah, that is a good idea once the most recent fine has been paid.
The write up quote the Google top dog as saying:
… he wanted to solicit ideas from his employees on how to get “better results faster.”
Okay, that certainly opens to do to some creating thinking. Based on my attending law enforcement and intelligence conferences, the quick money comes from fraud, human trafficking, selling contraband, and possible a few weapons deals. I would toss in getting in the fulz business as another “let’s noodle that” idea.
My personal perspective is different. I think the actual term is my viewshed, but I could be wrong, a practice I have explored in my 77 years of muddling along. Here we go:
- Google embodies the management methods of what I call a “high school science club.”
- The company had been purposeful when refining its “clever” search system and obtaining inspiration from GoTo, Overture, Yahoo ad sales technology. Once achieved, the once sharp lens lost its ability to focus
- The culture of the Google in interesting. Foosball, bean bags, volleyball, and car washes complement the weird caste approach to food. Change may not come easily into this good organization.
But maybe the write up is incorrect. Google is nothing more than the best managed online advertising outfit in the world. See. Money works to explain the reality. Blood? Hmmm. Possibly an overstatement?
Stephen E Arnold, August 15, 2022
Google: Oh, Oh, Another Example of a Government Not Being Googley
August 12, 2022
I read “Google LLC to Pay $60 Million for Misleading Representations.” The write up reports, if the information is spot on:
The Federal Court [of Australia] has ordered Google LLC to pay $60 million in penalties for making misleading representations to consumers about the collection and use of their personal location data on Android phones between January 2017 and December 2018, following court action by the ACCC. The Court previously found that Google LLC and Google Australia Pty Ltd (together, Google) had breached the Australian Consumer Law by representing to some Android users that the setting titled “Location History” was the only Google account setting that affected whether Google collected, kept and used personally identifiable data about their location. In fact, another Google account setting titled “Web & App Activity” also enabled Google to collect, store and use personally identifiable location data when it was turned on, and that setting was turned on by default.
What did this decision tell me? The Australian Court is not Googley. Ah, lawyers.
Stephen E Arnold, August , 2022
The New Yorker Magazine Gets Close to a Key Precept of Google Senior Management
August 11, 2022
I suggest that anyone interested in the bizarre personnel decisions which have become as notable as Google’s amazing announcements about its technology read “Google’s Caste-Bias Problem: A Talk about Bigotry Was Cancelled Amid Accusations of Reverse Discrimination. Whom Was the Company Trying to Protect?” The article did not appear to be behind a paywall, but you may be asked to spit out some cash to read the interview.
I am not going to discuss the ins and outs of the interview, its factoids, or the motivation for the comments.
I have several observations:
- The New Yorker has identified and made visible behaviors which have been ignored by other “real” news outfits; for example, Yahoo News. (Yahoooooo!) The question is, “Why?”
- The consequences of certain decisions have been fascinating. Dr. Timnit Gebru departed and set out to do the Don Quixote thing? Blake Lemoine, the fellow who thought software was alive, is now free to share his insights on podcasts. By doing this, he highlights some of the thought process of Google professionals. Then there was the cult. I don’t want to think about that.
- The high school science management methods of the Google have certain deep roots. I am not sure if these are cultural, bro-behaviors, or some other protein firing in the carpetland crew. What’s clear is that only Meta’s management methods are in what I would call the Sergey-Larry league. Maybe it’s the water in Silicon Valley.
Net net: The New Yorker’s Delphic soothsayers are definitely on to something that business school gurus have been skirting for years.
Stephen E Arnold, August 11, 2022
Google Kicked Out of School in Denmark
August 11, 2022
Like its colleagues in Netherlands and Germany, the Denmark data protection authority has taken a stand against Google’s GDPR non-compliance. European secure-email firm Tutanota reports on its blog, “Denmark Bans Gmail and Co from Schools Due to Privacy Concerns.” Schools in the Helsingør Municipality have until August 3 to shift to a different cloud solution. We learn:
“In a statement published mid July, the Danish data protection agency expresses ‘serious criticism and bans … the use of Google Workspace’. Based on a risk assessment for the Helsingør Municipality, the data protection authority concluded that the processing of personal data of pupils does not meet the requirements of the GDPR and must, therefor, stop. The ban is effective immediately. Helsingør has until August 3 to delete pupil’s data and start using an alternative cloud solution. … This decision follows similar decisions by Dutch and German authorities. The issues that governmental institutions see themselves faced with has started with the invalidation of Privacy Shield back in 2020. Privacy Shield has been a data transferring agreement between the USA and the European Union and was supposed to make data transfers between the two legally possible. However, the agreement has been declared invalid by the European Court of Justice (ECJ) in 2020 due to privacy concerns. One major problem that the EU court pointed out is that data of foreigners is not protected in the USA. The protections that are there – even if limited – only apply to US citizens.”
So the NSA can gain unfettered access to the personal data of Europeans but not US citizens. We can see how authorities in the EU might have a problem with that. As the Danish agency notes, such a loophole violates rights considered fundamental in Europe. Not surprisingly, this Tutanota write-up emphasizes the advantages of a Europe-based email service like Tutanota. It is not wrong. It seems Denmark has woken up to the Google reality. Now what about Web-search tracking?
Cynthia Murrell, August 11, 2022
Xoogler on AI Ethics at the Google: Ethics? Explain, Please
August 8, 2022
I read a write up which seems to be information I have seen cycled and recycled. Nevertheless” “An Engineer Who Was Fired by Google Says Its AI Chatbot Is Pretty Racist’ and That AI Ethics at Google Are a Fig Leaf” contains an interesting observation; to wit:
“These are just engineers, building bigger and better systems for increasing the revenue into Google with no mindset towards ethics,” Lemoine told Insider. “AI ethics is just used as a fig leaf so that Google can say, ‘Oh, we tried to make sure it’s ethical, but we had to get our quarterly earnings,'” he added.
The statement is interesting from several different vantage points:
First, the Xoogler is directing public criticism at the GOOG by linking Google’s approach to ethics to a fig leaf. The metaphor to a fig leaf. Wikipedia points out:
Some paintings and statues have the genitals of their subjects covered by a representation of an actual fig leaf or similar object, either as part of the work or added afterward for perceived modesty.
Yep, perceived modesty and the evocative idea of a cover up.
Second, the recycling of negative comments could create a target for some of Google’s legal eagles.
Third, the allegations about ethics and its possible subordinate role to achieving goals supports assertions offered by Timnit Gebru et al.
Will a burning pile of fig leaves attract attention? Unlikely. Hey, it’s the Google.
Stephen E Arnold, August 8, 2022
Google Innovates Again: Quick or Is That Semi-KWIC?
August 5, 2022
Innovation at the Googleplex never stops. Never. I read the online story “Google Updates Search Result Snippets for Queries with Quotes.” The write up reports that after more than two decades of defining Googling as searching:
Now Google will now show the quoted text in the snippet where that exact phrase appears on the page.
The idea upon which the quantumly supreme Googlers hit is that some context, not much, but some is helpful. No one has ever had this scintillating insight before. Amazing. Think about it. A person’s search for a quote returns some semi-context. I learned:
Google said they made this change based on searcher feedback, Google wrote “We’ve heard feedback that people doing quoted searches value seeing where the quoted material occurs on a page, rather than an overall description of the page. Our improvement is designed to help address this.”
A few observations? Sure, why not?
- Google ignores bound phrases and user defined phrases in quotes. Don’t you love the strike out for the key words in a query that do NOT appear in the results list? I do. Will this helpful feature be decremented or ignored?
- Key words in context has been a function for a long time. I am not motivated to dig through my 50 year archive of “search” ideas to locate the very first KWIC option. I think some of the long-forgotten online search systems offered this feature? Maybe Dialog circa 1980? Somewhere around there. I recall Carlos Cuadra talking about the function at an Information Industry Association 45 years ago. Yo, SDC experts, any thoughts?
- Is this the magic learnings of a former Verity wizard transporting inspirations to the GOOG?
- Will the Google allow the user to specify the size of the KWIC window? Sure, when Google discovers the function. What’s next? Boolean logic?
Wow. KWIC.
Stephen E Arnold, August 5, 2022
-
- Subscribe to Beyond Search
Feature archive
News archive
-
