Talk:IPv4

The article suggest that in the 1980's address exhaustion was expected, based on laptops, PDAs, and smartphones. Laptops and PDAs do trace back to the 1980's, though as well as I know, not ones with IP addressing. From History of smartphones, smartphones are definitely later than 1980s. I do remember by late 1980's, the ability to connect IBM PC (ISA) machines to Ethernet, with PC-NFS and NCSA Telnet, and that might have included a small number of laptops, it seems surprising to me that it would be enough to predict exhaustion based on those. Gah4 (talk) 07:20, 3 November 2025 (UTC)

I lightly modified the text to conform more closely with reality. This conforms with the dedicated address exhaustion article. cheers. anastrophe, an editor he is. 18:29, 3 November 2025 (UTC)

Chronic1 insists on adding information about RFC 760 to the first sentence of the lead. RFC 760 is mentioned as sort of an aside in the History section. I reverted suggesting that if this is indeed worthy of mention in the first sentence, we develop the body further before adding it to the lead. Chronic1 suggests that we retain their changes and somehow develop the body from that. I'm looking for input from other editors on how best to proceed. ~Kvng (talk) 16:13, 16 November 2025 (UTC)

WP:LEDE strongly "argues" (it's a formal guideline) that the lead is a summary of the body. If the body does not have any coverage or inadequate coverage - or if the body gives appropriate coverage but that coverage simply shows that it's not a significantly material set of facts/claims, then it does not belong in the lede - not until it's been satisfactorily developed in the body or shown not to be significant.

The late president Ronald Reagan wore tan/brown suits during his tenure as the leader of the free world. This made national news - the New York Times reported on it. Does that therefore mean that his article should lead off with that fact in the first sentence? No. It has to do with relevance, significance, coverage, and a multitude of other factors.

What we can learn from my comment is that I need to ease off on the four-hundred horsepower coffee. cheers. anastrophe, an editor he is. 18:48, 16 November 2025 (UTC)

Regardless of whether this information belongs in the lead, it needs to be verifiable. I attempted to verify the claim that IPv4 was initially proposed under RFC 760 by the Department of Defense against cited sources, and failed. The RFC, if we can consider it to be a source on itself, says that it was written by ISI for the DoD, not by the DoD. There is one statement in the article body speaking about the DoD and RFC 760, and its source only states that In March 1982, the US Department of Defense declared TCP/IP as the standard for all military computer networking. Nothing about specific RFCs, or which party proposed it.

I would support a statement about IP's DoD origins in the lead if it were supported by a source used in the body of the article, but it currently is not. DefaultFree (talk) 00:16, 17 November 2025 (UTC)

I have reverted these changes again for now. anastrophe and I find it runs afoul of WP:LEDE and, more seriously, DefaultFree is concerned about a WP:V issue. We potentially have a consensus here since Chronic1 doesn't have a specific policy supporting their edit. We can continue discussion and un revert if things take an unexpected turn. ~Kvng (talk) 14:39, 19 November 2025 (UTC)

Yes, that makes a lot of sense. I was not aware of WP:LEDE and want to apologise for the inconvenience. At least I learned something. Chronic1 (talk) 15:14, 19 November 2025 (UTC)

The lead sentence currently relies on a 2013 citation to claim IPv4 routes "most" traffic. However, as of March 2026, we have reached a global tipping point.

• Global Adoption: Google's transparency reports now show global IPv6 adoption hovering around 47–50%, with many sources (Cisco/APNIC) indicating it is officially overtaking IPv4 as the majority protocol this quarter.
• Regional/Mobile: US adoption is well over 58%, and major mobile carriers (T-Mobile/Verizon) are consistently above 90%.

I've flagged the sentence with an {{Update inline}} tag. The lead should be updated to reflect a "near-even split" or "ongoing transition to majority status" rather than implying IPv4 is still the undisputed primary protocol.

Sources:

• Google IPv6 Statistics (March 2026)
• APNIC Labs IPv6 Measurement

Awhit003 (talk) 20:08, 14 March 2026 (UTC)

I don't doubt that's all true, but there probably should be a finer distinction made in terms of the 'audiences' that have adopted it. Carriers obviously use it routinely for backbone traffic, CGNAT and whatnot, but 'end user' adoption is still in the tiny fractions. As well, the vast majority of destination sites on the internet are all still on IPv4; we're still decades (if ever) from the entirety of the internet being IPv6 with IPv4 no longer in use. cheers. anastrophe, an editor he is. 21:03, 14 March 2026 (UTC)

I retract my point. The article is not misleading in a sense of scale or trends, simply woefully outdated like vast sections of Wikipedia. The lead needs a re-write that I am not qualified to write. Someone else will have to delete this talk section, also; perhaps in the year 2039. Awhit003 (talk) 21:48, 19 March 2026 (UTC)