The experience of using the internet to teach or learn about computers has not improved since 1992, but that is because there was a robust population of people using the internet of 1992 to teach and learn about computers.
If you wanted to teach or learn about a particular chronic health condition on the other hand you would struggle to find others on the internet of 1992 who shared your interest.
In 1992 the web existed, but it played a minor role. What was then called Netnews, but would be renamed a few years later (by Time, Newsweek and other news organization who took offense at the name "Netnews") to Usenet was definitely the front page of the internet, and if a particular document was available on the internet, it was more likely to be available on an ftp server (that had been configured to allow anyone to sign in) than on a web server.
The web was the first significant internet service to offer a GUI. before the web you had to use terminal-based interfaces to access any internet service.
The web (and to a lesser extent the discovery of the internet by mainstream journalists) caused a massive democratization of the internet, which had good and bad effects.
The good effects are obvious, so let me give an example of a bad effect.
I recently had the need to learn more about a mental disorder called dissociative identity disorder, which is trickier for humans to think rationally about than most mental disorders are. i.e., it is quite tricky to think rationally about, so tricky that most humans and even many trained psychotherapists fail. I used Google Search and was successful in learning what I wanted to learn. Even though the vast majority of web pages (on this particular subject and in general) have been written since 2000, the 2 most useful documents for my learning were written in the 1990s. One appears to have started life as a web page in 1996 or 1997. The other was a FAQ regularly posted in the 1990s to the newgroups alt.support.dissociation, alt.abuse.recovery, etc.
The point is that most of the person-hours spent on the internet have been spent since 2000, and I would have expected and would have hoped to see more signs that all those person-hours haven't just gone to waste than I have actually seen during my searching and browsing of the web.
I've kind of switched the focus from 1992 to 2000 here. 2000 represents in my mind a kind of high water mark for the internet. I know we cannot go back again to the internet without commercial activity, so I cannot get too excited about the internet of 1992.
There are classes of Google Searches I performed often in 2000 that I have learned to refrain from now because they no longer constitute a productive use of my time. For example in 2000 I spent hours in Google Search trying to learn whether AI research might cause the end of the world -- and I was able to drastically improve my understanding of the issue. These days I still do a lot of Google Searches, but they are mostly about "settled science", e.g., what is a thiol?, e.g., whether any iphone model has an OLED display, e.g., what is a bounding box in computer graphics?
I also continue to search often for information that is easy for me to verify. e.g., if I search for recommendations for a browser extension to block ads, it is relatively easy for me to verify the recommendation by installing the extension, then watching to see how well it blocks ads.