Should the Eth 1.x name be dropped?


This to me was the most interesting conversation in the first community call. It seems that a lot of the confusion around Eth 1.x could be that people don’t realize it’s just the short term roadmap for Ethereum, not a separate initiative. Also since 2.0 sounds more attractive than 1.x, perhaps it’s not getting enough attention despite a lot of the work being eventually used in 2.0. Should we drop the name and just instead call this effort Ethereum?

There seems to be too much overlap these days to have them be so separate. Just off the top of my head the overlap includes:

  • Finalizing eth 1.0 chain with 2.0 beacon chain
  • Fee market overhaul via eip-1559
  • State rent/stateless clients
  • ewasm

I’d probably prefer the roadmap be merged and instead of splitting by version we use themes which stories fall into such as Scaling, UX and more which then get merged into forks as they are ready. AKA a normal dev process :slight_smile:


I agree that implementing too many designations can cause/add confusion. To me, the dotX implies an unknown number of sub-releases with unknown specific criteria to each. It raises the question of what, for example, would ETH 1.1 consist of? Now, I’ve never seen anyone use specific 1.1, 1.2, etc. naming in discussion, but it’s certainly alluded to by using 1.X. Perhaps worse so, it implies those sub-release specifics haven’t been decided on or perhaps even considered.

I’ve thought more about it since the end of the call and think I’m leaning toward abandoning the term personally. It makes sense to have a separate designation for ETH 2.0 if for no other reason than it’s actually a new chain.

+1 for ETH 1.X --> ETH

1 Like

Like I’ve mentioned in the call, I agree on dropping the Eth 1.x name. The reason being, for me at least it seems like maintenance work. Not something that will benefit the Eth 2.0 chain. I believe many others view it this way also, and because of this I get the impression of why waste time on hearing about any changes on Eth 1.x if it’s not going to matter in Eth 2.0. But if the changes made with Eth 1.x will be included with Eth 2.x, then I see no reason to distinguish the two.


Agreed. It was clearer when there was just Ethereum, and names and goals for upgrades. The 1.x name causes unnecessary confusion.


agree, honestly the biggest takeaway from the call was how terms may be limiting perception. 1x is Ethereum, it is ETH 2.0! it just sounds like separate initiatives to those that aren’t neck deep in this stuff.

Improve Ethereum Now to prepare for Ethereum Future.

1 Like

Agree with all of the above. I think the best way to get this started is to put together a blog post that we can publish via the EthHub Medium publication as a community initiative and then spread it around on Twitter, Reddit etc. We’ll also get the relevant EthHub pages updated to reflect the changes.

I can start putting together a draft this weekend unless someone else wants to take it up?

I also think it would be a good idea for someone with graphic design skills to create some easily digestible graphics to clearly show a “roadmap” of sorts that outlines that all of the development on Ethereum is happening in parallel.


Let me ask the question in a different way: what if anything would we lose by dropping it? And what would be required to drop it?

Seems like we’d lose nothing, but only if we ensured that 1) key elements of the 1.x “roadmap” (such that it exists) get pulled under the banner of Eth 2.0, or 2) we create some sort of branding around it as “Critical Ethereum Sustainment.” Frankly, I don’t love the Option 2 message of parallel sustainment to developing Eth 2.0, either. The best outcome to me is finding a way to discuss these efforts as a part of the broader Eth 2.0 roadmap under Option 1.

If we don’t do one of those two things and eliminate the 1.x branding, then we risk fragmenting the 1.x efforts with no overarching campaign to support them (although arguably, this is effectively the status quo).


This can’t get folded into ETH2 because it’s unclear when or if ETH2 will ship.

Ethereum is Ethereum until ETH2 completes all phases, with a question mark about what happens to current Ethereum state (become a shard, migrate for those that want to, or other options).

The current Ethereum roadmap is more cohesive and more planned than ETH2 at this stage.

So, what would be required is an acknowledgement that this is the chain we have today, it is usable and valuable today, it is getting new features and improvements, and it is benefiting from ETH2 research (E.g. Finality). And just call it Ethereum.

This is the Osborne effect issue.


Really cool to talk about this issue and I am glad it came up!

There needs to be clear communications about the platform and its evolution; it ultimately drives how decisions are made about where we invest our time and resources. We watch the software industry move their names and versions around, how strategies shift, and it can be jarring for developers and users.

Several points about 1.x

It is important to note that the proposed Ethereum 1.x upgrades, particularly the state fees component, are not short-term. State fees is a very big upgrade to the current EVM-based platform, and makes it sustainable beyond a few years from now. Ethereum 1 sustainability is a key requirement for supporting the early Ethereum 2.

Another thing is that bundling all of these upgrades under the moniker “1.x” helped bring attention to a serious and complex problem in the community. A lot of investment was going toward innovating the future platform and to maintaining the current platform (some amazing optimization work there). But what was being missed was near-future competitiveness and sustainability.

The key question for Ethereum 1 was and still is: with so much invested in the EVM ecosystem, what can Ethereum do to compete while Ethereum 2 gets built out? Ethereum 1.x has emerged as the answer.

What is the numbering trying to solve?

Numbering does hem things in and leads to weird expectations, especially the Osborne effect as @boris pointed out. It would actually be beneficial to lose the rigid numbering. @VitalikButerin and the sharding team did try to break free of it with “Serenity”. But I see Serenity as something bigger than just a release.

Perhaps we need epochs or ages or something like that, to gather sets of releases into general themes.

1 Like

Alexey has created a post where he discusses how the process of 1.x is unique and there is value in that and it’s not just roadmap items.

The post is very well laid out and I think actually identifies some flaws in the base EIP process as is. He seems to argue (and I agree) that more complex ideas that can’t be written straight into an EIP and need some pre-discussions. Stuff that requires that seems to just be falling into the 1.x bucket because of the process differences.

I have a background in product management and how I’ve typically handed an idea that is bigger than an extremely simplistic user story is:

Idea > Discussion > Spec > User story breakdown (usually multiple) > Discussion of stories > Estimation of effort > prioritization > coded > tested

It seems the current process skips the first 3 steps:

User story > Discussion > Estimation effort > prioritization > coded > tested

Because many of the current 1.x ideas require up front discussion and more formal specs, the idea of working groups has been created.

Is this more a flaw in the original process and 1.x is leading us to realize it should be transformed so that we don’t create a split in the roadmap between easy and complex issues? My fear here is the complex issues will just never happen because they’ve been “cast off”.