Re: Naming Clash: The W3C Web 3.0 Stack and "Web3"

Dear all,

a few thoughts on this proposal.

First, none of Web 2.0, Web 3.0, web3, Web 1.0, Web 4.0, and more are 
well-defined concepts with an officially-sanctioned definition. All of 
them have in common that someone looked at the web as it existed at the 
time and thought "we need to change something relatively big about how 
the web works." And ever since that dark and stormy night in 1999 when 
Darcy DiNucci minted the terms "Web 1.0" and "Web 2.0", people have been 
reusing the versioning metaphor as a signal that they're proposing a big 
change.

Now that is actually a problem? I don't think so. The web is unwell and 
we should be *encouraging* people to think of big things to change. Come 
up with Web 17 for all I care if it gets us unstuck from captured 
infrastructure and native envy. No doubt there'll be disagreement about 
some of the changes — great! We should be debating what's broken in web 
architecture. Sure enough, every now and then, instead of actually 
proposing a real change, someone will just cobble together a fad salad 
and slap a version number on it to make it seem credible. (I'm looking 
at you Web 4.0[0].) But this is tech, of course people will fake 
expertise to get money.

At the end of the day, Web vX isn't a reserved namespace and no one has 
naming authority over it. The TAG has no standing in deciding who gets 
to use these terms, neither does the W3C. You could propose an IANA (or 
W3C) registry for this, but frankly I'm not sure what we'd do with it, 
what the value would be, or how we'd enforce it.

The TAG *could* review the architectural principles attached to one of 
these labels, but that's a very different request from the one you're 
making.


Second, I find it gently ironic that every single argument you list to 
support your position that "web3 isn't the web" are arguments from 
authority. I think it's funny precisely because a key goal of web3 is to 
increase user agency by shifting authority to people (as discussed in 
[1] as well as in [2]).

If the TAG wants to pick up a discussion on architectural issues with 
self-certifying protocols (or with the other major component of web3, 
cooperative computing) then I think that could be quite interesting. 
There are specific and fascinating problems to consider with IPFS[3], 
CRDTs[4], petnaming[5], UCANs[6], new consensus methods[7], automated 
governance systems[8], Merklised all-the-things, Zooko's triangle[9], 
CIDs and their addressing trade-offs[2], mismatches between Fetch and 
content addressing, local-first storage[10], or client-centric 
primitives like Peergos[11] or Web Tiles[12]. There certainly is *a lot* 
of discussion about changing the web for the better that relies on 
concrete proposals that the TAG isn't touching.

Picking up specific issues for specific reasons would be great. Making a 
decision because someone respected, maybe even someone we like, 
pontificated about it? I don't think so.


Finally, it's not even clear to me that there is an architectural 
incompatibility between 3 and 3.0. Admittedly, I only know so much about 
3.0. It seems to be mostly about Solid-like stuff with control over 
one's data and authorisation to it. That's a smaller scope than 3's 
goals of user authority + cooperative computing but it would seem to 
align well with WNFS+UCAN[10][6] (and some RDF sprinkled on top)? Or 
maybe you would think about it more like an ATProto PDS, with 
repositories of Merklised IPLD?[13]

Either way, again, I think that the TAG could potentially answer 
specific questions like "given the intent to achieve X on the web, one 
architecture does Y whereas another does Z — are there significant 
differences and principled reasons to prefer one over the other?"


Anyway. I'm not on the TAG and can certainly not speak for it. As a 
member of the web community, I do see that there are many architectural 
topics that the TAG isn't looking at. I'm not sure what I would 
recommend to do about it, though, because there is so much going on that 
there simply isn't time to have in-depth architectural discussions about 
all of it in this group. But precisely because there is so much to look 
at, I don't think that the TAG should be spending time making statements 
about labels.



[0] https://ec.europa.eu/commission/presscorner/detail/en/QANDA_23_3719
[1] https://jaygraber.medium.com/web3-is-self-certifying-9dad77fd8d81
[2] https://specs.ipfs.tech/architecture/principles/
[3] https://ipfs.tech/
[4] https://automerge.org/blog/automerge-2/
[5] https://spritely.institute/static/papers/petnames.html
[6] https://ucan.xyz/
[7] 
https://raw.githubusercontent.com/consensus-shipyard/IPC-design-reference-spec/main/main.pdf
[8] https://metagov.org/
[9] https://en.wikipedia.org/wiki/Zooko%27s_triangle
[10] https://github.com/wnfs-wg/spec
[11] https://peergos.org/
[12] https://hackmd.io/@robin-berjon/tiles
[13] https://atproto.com/guides/data-repos

On 17/07/2023 05:30, Melvin Carvalho wrote:
> Recently the W3C linked data and the semantic web stack, commonly known 
> as "Web 3.0" has come into a naming clash with the term "web3"
> 
> I was wondering if the W3C TAG could take a more definitive stance on 
> this matter, recognizing the nuances, lack of education, and conflicts 
> of interest that may exist among W3C members.
> 
> In support of this request, I would like to highlight comments I would 
> like to highlight recent comments made by Tim Berners-Lee and Roy 
> Fielding regarding the term "Web3." Tim (according to CNBC), expressed 
> his view that we should "ignore the Web3 stuff" and stated that "Web3 is 
> not the web at all." [1]
> 
> Similarly, Roy Fielding characterized Web3 as a scam, emphasizing the 
> importance of not giving it any form of respectability. [2].
> 
> Highly respected former TAG member Alex Russell:
> 
> "A few months ago I tried to get current and past members of the @w3ctag 
> to sign onto a statement that, in effect, read "'web3' isn't the web; 
> please kindly stop web-washing". [3]
> 
> Having in-depth knowledge with both W3C linked data effort, and 
> distributed payments technology, for over a decade, I couldn't agree 
> more with these sentiments.
> 
> Furthermore, the term is increasingly creeping into standards 
> discussions in multiple W3C groups, so it might be a good opportunity 
> for the TAG to weigh in with some opinion a bit more.
> 
> Considering the strong viewpoints expressed by these influential figures 
> in the web community, it may be an opportune time for the W3C TAG to go 
> beyond previous statements and take a more decisive position in 
> rejecting the term "Web3." I acknowledge that there may be varying 
> interpretations and perspectives surrounding Web3, but it is essential 
> to address any potential confusion or misrepresentation to maintain the 
> integrity and clarity of the W3C's work.
> 
> I suggest the W3C TAG consider this naming clash, and/or engage in an 
> open discussion, to provide guidance and education to the web community 
> to address any misunderstandings associated with Web3. By doing so, the 
> W3C can play a role in ensuring that the web's evolution remains aligned 
> with its core principles and values.
> 
> tl;dr Web3 isn't the web, could we just come out and say that?
> 
> [1]  CNBC - "Web inventor Tim Berners-Lee wants us to ignore 'Web3'" - 
> [Link to 
> Article](https://www.cnbc.com/2022/11/04/web-inventor-tim-berners-lee-wants-us-to-ignore-web3.html <https://www.cnbc.com/2022/11/04/web-inventor-tim-berners-lee-wants-us-to-ignore-web3.html>)
> [2]  IETF URI Review Mail Archive - [Link to Roy Fielding's 
> Comment](https://mailarchive.ietf.org/arch/msg/uri-review/4Sj8k6rLZzqZsGgMEe6fLM-wE4U/ <https://mailarchive.ietf.org/arch/msg/uri-review/4Sj8k6rLZzqZsGgMEe6fLM-wE4U/>)
> [3] Twitter - 
> https://twitter.com/slightlylate/status/1478099457304449025 
> <https://twitter.com/slightlylate/status/1478099457304449025>

-- 
Robin Berjon (he/him)
Governance & Standards at Protocol Labs
https://berjon.com/ - https://mastodon.social/@robin

Received on Monday, 17 July 2023 16:32:22 UTC