[ih] Why did TCP win? [Re: Internet-history Digest, Vol 63, Issue 3

[Brian E Carpenter]

On 04-Feb-25 06:51, touch--- via Internet-history wrote:
> 
>> On Feb 2, 2025, at 10:20 PM, Dave Crocker via Internet-history <internet-history at elists.isoc.org> wrote:
>>
>> On 2/2/2025 3:16 PM, Brian E Carpenter via Internet-history wrote:
>>> I've always answered that with: Because it worked, had a full suite
>>> of applications, and it was free on Unix, and modestly priced on
>>> most other operating systems.
>>
>> /Specs/ and reference software were both free.  Big advantage over OSI. (I think the Unix version was nice, but Unix had such a tiny market share at that point, I suspect it didn't sway the market choice as we might think.)
>>
>> 20 years of operational history.  Big advantage over OSI.
>>
>> Telnet, FTP, email. In use over that entire time.
>>
>> Quick updates and quick development of specs.  OSI, well, umm...
>>
>>>
>>> After 1995 you could add "because of the Web".
>>
>> TCP had clearly and definitively won by the latter 1980s, before the Web was created.

I didn't intend to imply otherwise. But in terms of persuading management that TCP/IP was the way to go, long after the technical community had decided, the Web was very definitely a factor.

To underline the point: the *reason* that Tim Berners-Lee designed the Web as he did was because the community he was in had already concluded that TCP/IP was the future. It is not coincidence that the original proposal for the Web [1] and my paper "Is OSI too late?" [2] were written in offices about 30 metres apart in the same year.

[1] https://info.cern.ch/Proposal.html

[2] https://dl.acm.org/doi/10.1016/0169-7552%2889%2990040-8

    Brian

> 
> This is my recollection as well. I think too much credit has been given to the Web per se when GUIs on commodity home PCs were the real “game changer”, i.e., Windows 3.0 (which basically caught up to MacOS from 1984) running on machines that were half the price of a Mac.
> 
> I remember the explosion of email and other graphical link-based systems (Fetch from 1989, Gopher 1991, etc.).
> 
> I.e., the time was ripe and one of these systems was going to “win”. But giving credit to the web is like saying Facebook invented social media (vs Myspace? Friendser?). In many markets, “one wins”, but not necessarily the first or the best and nearly never by design.
> 
> IMO, it’s also very hard to notice when an exponential adoption curve starts - in the beginning, it was just universities, then universities and some workplaces, etc. Much like protocols, many user interfaces were walled gardens that didn’t interoperate (Prodigy, the original AOL, BBS systems) - until they started to, and when they did, they all used TCP.
> 
> It’s also widely known that historical “recollection" favors the individual over the group, which mirrors “who wins”.
> 
> IMO, the “history” here involves everything.
> 
> Joe
>