The last remaining Sears in Chicago, the department store titan's hometown, is down the street from my house. Above the entrance, looking over Six Corners, is a massive ad for their auto services. It's so sun-faded that it's only half-legible.

I've purchased things from it before, including a sprightly red Casio watch I quite like, but my experience shopping there matches up with the reporting of Business Insider's Hayley Peterson, who's done some of the best work on the long-struggling institution (especially this lengthy piece). It's like a Potemkin village in reverse: It is a real place of business that puts forth the minimal effort in appearing to function like one.

According to the reporting of Peterson and others, this is due to a shift in strategy, as Peterson reported: "Lampert's plan is for Sears to one day be a tech company, more like Apple or Facebook than a traditional retailer, according to three former executives." But Sears appears to be going the opposite direction from Apple, a tech company that has not only expanded its retail operations, it has invested considerable money in the experience of its stores. When Crain's Brigid Sweeney wrote a thorough piece on the company in 2012, she found that the company "spends less than a quarter of the $8 per square foot typically invested into stores."

The results are increasingly worse. Sears lost more than two billion dollars last year, its sixth straight year of losses, and five of those have involved losses of over a billion dollars. Comparable store sales fell 16 to 17 percent from last holiday season, one in which the competition did OK considering the retail environment.

What's the company to do? Continue on the path of becoming a tech company?

They've tried that, actually—in some ways, quite successfully, though they lost a pile of money on it. Sears was one of the creators of Prodigy, a one-time giant of the internet back when "portals" were the easiest way of accessing it. In 1984, Sears, along with IBM and CBS, created the dial-up service; after CBS dropped out, it launched in 1988, two years before the first web browser and website were created, and five years before the Mosaic web browser was created at the National Center for Supercomputing Applications at the University of Illinois.

It's not as famous as AOL, which conquered the portal space with its infamous mass-mailed CDs and managed to survive the demise of the portals, but it was critical training wheels for the internet, not least because its flat-rate pricing made it an affordable option for families like mine: $9.95 in 1988 is the equivalent of about $21 a month today.

I also had access to a dial-up service through my mother's job, but prior to Mosaic, the chief ways of using the internet were text-based programs like FTP and Gopher. Even with access to the sophisticates' internet, I used it for the same reason that Jim Carpenter, a programmer who has resurrected some of the portal, did:

"Honestly, I wasn't a huge fan of Prodigy," says Carpenter, a 38 year-old freelance programmer based in Massachusetts, recalling his time on the service around the turn of the 1990s. "I had already been using the Internet for a couple of years and Prodigy seemed so closed in. But I still used Prodigy every single day. It was the graphics."

That Atlantic piece, by Benj Edwards, gets into the details of how it worked: a  data-efficient protocol that grew out of Teletext TV graphics from the 1970s. (In the 1980s, newspapers like the Los Angeles Times and the Sun-Times experimented with the protocol to present breaking news.) What was efficient enough for primitive graphics via Teletext allowed for what was at the time considerable richness via Prodigy.

That's why Sears bought into the idea. Prodigy launched in Chicago in 1989, when it had reached around 85,000 households. Computerworld explained its appeal to Sears. "Unlike text-based on-line information services, Prodigy employs the North American Presentation Level Protocol Syntax (NAPLPS), for its graphical user interface," Ellis Booker wrote. "While NAPLPS adds a free face to on-line service and can portray a graphical representation of the camera that the consumer is thinking about buying through the Sears' on-line catalog, the speed of the service is appreciably slower than text-based services."

The Los Angeles Times's Stuart Silverstein profiled Prodigy that same year, calling it "an unprecedented bid to hook a broad cross-section of America on computerized shopping," and what an online product marketer presciently called "a marketplace that's probably 10 years away from becoming a reality" (Amazon's IPO was in 1997). Silverstein also wrote, very presciently for 1989, that "telephones with display screens and keyboards, for instance, might prove to be more inviting for computer-wary consumers than a PC-based system."

But Prodigy did a lot of things right, or at least did a lot of things that would become most of what people do on the internet now:

Unlike older competitors that cater mainly to computer buffs and devote most of their capacity to other electronic information services, the Prodigy service was set up mainly as "a merchandising medium" from the start, said Gary Arlen, a consultant in Bethesda, Md. It is aimed at harried but affluent two-income families who don't want to squander their precious time waiting in lines or parking lots.

The long-term market for Prodigy consists of people "who don't really want to use a computer," said Scott Corzine, Prodigy's director of merchandise marketing. "They want a utility that lets them run their lives more easily."

Besides computerized shopping with 70 retailers, Prodigy provides such features as news briefs, Consumer Reports summaries, educational games, banking, stock trading, flight reservations and electronic mail, which allows subscribers across the country to send messages to each other.

It also screwed up in ways that internet titans screw up now:

(Electronic mail is one of the most popular services, but it can be a headache for Prodigy. The company recently scrapped an electronic mail "bulletin board" on health that became a forum for a heated debate between homosexuals and religious fundamentalists. Prodigy said it eliminated the board because it was lightly used and denied charges that the company was trying to avoid controversy.)

And, in fact, a lawsuit against Prodigy by the subject of the film The Wolf of Wall Street, regarding bulletin-board postings about him, "led to much of the basic legal framework that governs Internet content." Prodigy, which wielded relatively strict control over what was written by users on its site in ways that grew into absurd fights that foreshadow how difficult content moderation would become, lost the lawsuit. The backlash against that decision is a big reason content providers have little liability for user content.

But Prodigy's list of offerings was odd, fascinating, and forward-thinking. Within its tightly controlled, family-friendly walled garden, it did allow for a primitive form of social media, which has since disappeared into the ether—hence Carpenter's project to rebuild bits of it—but reporting from the time details why it stood out. In 1989, the LA Times wrote that Prodigy had an "impressive list of 'experts' who write daily columns and who also answer questions sent by users over the service's electronic email," which included Gene Siskel, Robert Novack, Jack Germond, Jane Fonda, Dick Schaap, and Heloise.

In 1994, the Baltimore Sun described it as "a great haven for political junkies, with an unusually thorough database of information on Congress, including authoritative political biographies, ratings by major interest groups and campaign contribution summaries." The Atlanta-Journal Constitution signed up 20,000 subscribers through its $4.95-a-month Prodigy-based Access Atlanta service. In 1994 the Times-Mirror company, then the publisher of the Los Angeles Times and Newsday, signed up not only to publish their newspapers on Prodigy but to "use the near-infinite space of the digital format to create not only an electronic version of what is already available in the daily paper, but also something akin to an interactive community database":

For example, Southern California subscribers to TimesLink will be able to enter their ZIP codes and call up recent news stories about their neighborhoods, a list of local libraries, real estate information and restaurants. Clicking on another icon will conjure up a portfolio on the O.J. Simpson trial, complete with photographs, profiles of the main characters and the juror questionnaire.

"No one's been able to figure out how to get ads to work in the on-line context," an analyst told the Los Angeles Times when asked about the plan. "So the concept of content driving circulation which in turn drives advertising revenues is broken."

So, they discovered that, too.

All along the plan was to maintain Prodigy's comparatively low subscription costs through a hybrid model that should sound familiar: an affordable monthly subscription plus advertising plus a cut from the retailers who used the system as a marketplace. (Early on, San Franciscans could use it to get groceries delivered, because of course.) Compare that to what the New York Times does now: It acquired The Wirecutter, a publication that does laboriously thorough reviews of products and makes money off affiliate links, a $30 million acquisition that seems to have brought in $5 million in the first quarter of 2017 alone.

Prodigy ultimately didn't work. Neither the e-commerce nor the advertising model brought in enough money to subsidize its flat-rate fee. A 1993 Wired piece explained why: real-time or near-real-time communication, like chat, bulletin boards, and email, required frequent long-distance phone calls between their servers and users. That was expensive, and Prodigy underestimated how much their users wanted such services and overestimated how much they would shop and how much money advertising would make. The company flailed around making up the difference, raising its flat rate, implementing tiered pricing, and charging for emails beyond 30 a month.

Prodigy's bungling of its pricing changes likely caused it to give up its lead over AOL, which survived the transition from portals to internet service providers to the internet of today; despite a chaotic run during that period, AOL sold for $4.4 billion to Verizon in 2015. Sears and IBM, which each held a 50 percent stake in Prodigy, sold it for $200 million in 1996, after spending more than a billion on it over 12 years.

They misread the market of its time, but understood what it would become. "A very expensive technological effort," Edwards writes, "would end up inadvertently crafting  a consumer online world for the everyman that eerily presaged the Internet we know today—if in a Bizarro Superman type way."

What Eddie Lampert will do with Sears's continuing tech shift is anyone's guess; if it continues to bleed money, it can only hope to be as prescient as Prodigy was even in failure.