Weeknote 5, 2023

§1 Google’s Featured Snippets

Yesterday I searched Google for the phrase personal property and noted that the first result was a definition from Thomson Reuters’ Practical Law.

This definition is a Google Featured Snippet.

How Google’s featured snippets work

Google’s search results sometimes show listings where the snippet describing a page comes before a link to a page, not after as with our standard format. Results displayed this way are called “featured snippets.” You might find featured snippets on their own within overall search results, within the “People also ask” section, or along with Knowledge Graph information.

We display featured snippets when our systems determine this format will help people more easily discover what they’re seeking, both from the description about the page and when they click on the link to read the page itself. They’re especially helpful for those on mobile or searching by voice.

Featured snippets commonly contain one listing, but more than one may appear.

How featured snippets are chosen

Featured snippets come from web search listings. Google’s automated systems determine whether a page would make a good featured snippet to highlight for a specific search request. Your feedback helps us improve our search algorithms and the quality of your search results.

Tip: If you’re a website owner, learn how to manage featured snippets in Google Search.

My follow up question was answered on this page:

You can’t. Google systems determine whether a page would make a good featured snippet for a user’s search request, and if so, elevates it.

This answer surprised me. I thought Featured Snippets were connected to Google’s Knowledge Graph and/or Knowledge Panels.

§2 Postmedia shoots the Windsor Star

Earlier this week it was announced that Postmedia was going to close the printing press and make additional job cuts to our local newspaper, The Windsor Star.

From what I can tell, no local coverage of this announcement framed these cuts by the company as a direct message to the Federal government like Canadian Dimension has done in its piece, Postmedia shoots more hostages to keep debt payments flowing to New Jersey hedge fund. It begins:

New Jersey’s largest newspaper chain shot another dozen Canadian hostages last week as the end game of its northern extortion scheme grows ever closer. If the federal government doesn’t allow Postmedia Network and the rest of the country’s press to start taking money from the pockets of Google and Facebook soon, more of the country’s captive newspapers could get whacked. That’s because the Jersey boys have just about bled dry the chain that owns most of Canada’s largest dailies. Its latest financial report shows that Postmedia is sinking fast, and it could take most of Canada’s press with it.

The former Southam newspaper chain was acquired out of bankruptcy in 2010 by New York hedge fund Goldentree Asset Management, which had bought up most of its debt on the bond market at pennies on the dollar. It kept the debt on the renamed company’s books so it would be paid first every month, then took over the country’s second-largest newspaper chain, Sun Media, in 2014. Goldentree sold out to New Jersey-based Chatham Asset Management two years later in a deal that “happened so quietly,” noted the New York Times, “that Postmedia’s own financial news site described it as a debt restructuring in a report that included a single mention of Chatham as ‘one of the investors.’” That fast move flew right over the head of comatose regulators in Ottawa. “Few people noticed,” added the Times, “including some of the chain’s employees.” The deal wiped about $300 million of debt off Postmedia’s books but raised its foreign ownership to a dizzying 98 percent, with Chatham alone holding 66 percent.

§3 Consider the Time

I enjoyed this column from Sarah A. Sutherland called, Considering the Time Element in Law:

Law is a unique and important dataset: to a large degree it is a record of governance. It also tends to be conservative, so people can know what is likely to happen in the future based on what has happened in the past. Structurally, it has elements in common with other large text-based collections, such as aggregations of literary works. However, socially it has more in common with other high stakes bodies of information like medical research, with concerns like privacy and direct impact on people’s lives being necessary considerations. These attributes combine to make law as data a strange and wonderful problem to work on: this complexity is compounded by the way the law interacts with time.

Thinking about the time element reminds me of this article from L.M. Sacasass: We Are Not Living in a Simulation, We Are Living In the Past:

1. On the internet, we are always living in the past.

The internet, as a mediator of human interactions, is not a place, it is a time. It is the past. I mean this in a literal sense. The layers of artifice that mediate our online interactions mean that everything that comes to us online comes to us from the past—sometimes the very recent past, but the past nonetheless. This is so because …

2. On the internet, all actions are inscriptions.

These inscriptions include verbal expressions as well as images, audio, videos, memes, likes, shares, buttons hovered over, forms filled but never submitted, location data, etc. The inscriptions are legible either to humans or to machines, or to both. The fundamental fact of our online experience is that we are at all times expanding the massive reservoirs of the documented past. By contrast, consider the evanescent spoken word, part of which has already vanished before it is fully articulated by the speaker…

I do not have the capacity nor the bandwidth to consider what the internet can learn from Law in terms of understanding the context of time (or vice versa) but I’d happily read such an article … when I have the time.