The "web" generally refers to http/html which were extremely new in 1995, much newer than generic tcp/ip networking. The first netscape navigator was released in 1994!

Yup. "The web" = www.*

Or rather, http://

It was an interesting time with Windows 95 being shipped without IE or a TCP/IP stack by default.

This lead to needing to install stuff like Trumpet Winsock.

This mirrored the local networking capabilities between Windows 3.x and 3.11 Windows for workgroups.

Eventually, Microsoft would realize the need to include all of this by default in Windows, which led to the antitrust suit.

It's worth noting that this was a contrarian view in 1995, and that most of us saw things going the other way.

I agree. I remember hearing Clifford Stoll interviewed on the radio around 1995 and reading reviews of his book "Silicon Snake Oil," which was published the same year, and thinking that he was obviously wrong. It was just such a game-changer for me to be able to sit in my home in Tokyo, and, through a dial-up modem, get information I needed from throughout the world—information that otherwise would have required a trip to a major library or bookstore, if it was even available there. And to be able to find people interested in chatting about topics that no one else in Japan would have cared about. The Internet was clearly the future.

Cliff's viewpoint was definitely contrarian, but on the other hand many people thought that the internet would be a kind of utopia - either full of rational speech leading to better decision-making in the real world, or even a separate independent "nation" that governments could never control. Neither of those turned out to be fully true either.

When the internet was handles you could say anything and rarely worry it would follow you back in life. Now people use real names on social media.. which makes it about defining what imagine I want to be.

Real names attach constraints.

I dunno, weren’t university logins tied to real identities for many decades?

For some reason the myth that in the old days of the internet people used "handles" or nicknames and not real identities crops up here all the time. I can only assume people making the claim weren't on the internet in the 20th century and/or were on AOL.

On the Facebook of the 1990s, Usenet, people routinely posted with their real name/identity and real email address, and message headers showed each server the post passed through from its origin, the IP address of the poster (which could and did result in doxxing) and often an "abuse@ISP" email address to complain about the post/poster.

True, but that was for convenience, not because we thought we needed to have a "real identity". (IP addresses were later, ISPs were much later, but yes, UUCP hops were recorded. FWIW, my UUCP paths ended in something close to my legal name.)

Remember some of early Usenet, I wouldn't be surprised if some people had taken advantage of Californian naming laws (at the time, one could use almost any name one wanted, as long as it was consistent and not intended for fraud) to make their legal name more consistent with their online handle.

Mark V. Shaney is an example of a user who was not using their real name/identity (nor even had one).

Maybe https://news.ycombinator.com/threads?id=thequux can tell us how pseudonymous usernames were in his time?

    Quaxity quuxity,
    Teitelman's InterLISP
    Has a DWIM feature that's
    Really a screw;
    MacLISP has evident
    Superiority,
    Letting its customer
    Mean what he do.

As a teenager at the time it was impressed on us how important it was to not use our real name on the internet. I didn't use my real name for anything on the internet for at least 10 years after I started using it.

My university login was a pseudonym. Sure, you could have tried to get my real name from the admins, but it would have taken at least some effort.

Moreover, online forums (remember phpBB?) were overwhelmingly pseudonymous on practice.

Pretty sure you could have done:

    finger [email protected]

and found out who I was all the way through to the mid 90s.

University logins were handles (usually close to "real", but not tied) in my time. I firmly believe the whole real identity thing has always been driven by advertising considerations.

(Our undergrad computing lab had donated CAD workstations, so someone came up with a graphical display of the room, showing who was logged in where. The TA was displeased with me after figuring out that I had set my avatar to the "unoccupied" graphic. IIRC, he added a square outline for logged-in workstations to the display, around the same time.)