I think we meant different things when we said "drop in replacement" and therefore were referencing different ideas of what makes something a "drop in replacement," which is why it sounds like there are feelings of goal posts shifting.

If the messages are still sent via Facebook servers, that is not a WhatsApp replacement, it's an alternative WhatsApp Client, it's still at it's core "performing" WhatsApp. It is not a WhatsApp replacement, but a WhatsApp client replacement. I moved to signal specifically to sever my relationship with Facebook because I don't trust Facebook. Not communicating with Facebook is the feature that made signal appealing and made me want to replace WhatsApp (not the client, but the service as a whole) with something else.

I think conflating the idea of replacing "WhatsApp the service" and replacing "the WhatsApp Client" is the crux of our talking past each other and why my focus is on how it's UI compatible and ignoring the idea of protocol compatibility.

> You're on HN, it's nerdville, expect more interest in the protocol than clients.

We are on HN, and it's nerdville, it's true. It is the place where the very same comment (rsync files around) I am using to criticize the hubris of nerds (myself included) was made. (https://news.ycombinator.com/item?id=8863). Plug and Play (referenced in the HN link) is a winning idea. The crux of my statements here is how amazingly plug and play signal is, specifically for WhatsApp users.

> This is as daft as Googling "email" and expecting a de facto client.

I am saying this in good faith and I hope you take it as a good faith comment and not an aggression, but have you googled "email"? I understand the point you were making and I think it applies to a lot of other protocols, but googling "email" returns gmail 1st and 2nd, then outlook 3rd. Wikipedia is the 4th result...

> Try searching for "Matrix clients". Element is the best thus far, IMO.

I hate to respond so directly to this too, but have you googled "matrix clients" because I did, and I am pretty confident the results don't prove the point you wish to be proven. The number 1 result is the matrix website matrix client page I had already found by searching "matrix app". The second and third results are top 10 lists. The 4th result which you have to scroll down to see is Element. How do top 10 lists outperform clients themselves? That paints a pretty bleak picture for the difference in quality between the best app and the worst app.

Engineers have a way of being arrogant when they think the thing they have is technically superior or have a grand vision. Betamax was better right? Being protocol first over customer first, to me, is a form of hubris. The customer experience is what wins, everything else is just an implementation detail for the vast majority of people, even engineers.

> Widespread adoption" includes the EU's military, healthcare and government, so I wouldn't be so certain you'll end up being right. It's hit 60m publicly addressable accounts, which doesn't include any private servers or any kind of healthcare, gov or military: https://news.itsfoss.com/matrix-sixty-million-users/

This is definitely interesting and something I find worth considering. I certainly have an American-centric view. Clearly it's in every countries best interest not to have all their communication going through foreign servers, so the idea of Europe migrating to their own chat, much like Korea chose Kakao, doesn't surprise me. I kind of suspect that the idea of federating chat before the balkanizing of it might be too forward thinking/pre-mature.

> EU is forcing interoperability standards.

It will be interesting to see if American fights this or embraces it.

> I have a gut feeling that you will look silly in five years time

I would not be surprised to see Element, for example, become the most popular client and obvious choice. All the comments, FAQ's, etc, already seem to acknowledge Element is the right choice, but the structures that make it easy to download (google ranking result, links from the main matrix page, people saying "use Element," not "use Matrix") are not yet in place. When Element eclipses Matrix or Matrix starts being talked about outside the context of chat apps I'll start to take matrix seriously. Certainly my criticisms are not based on immutable flaws, but what I think are strategic blunders.

As a final note this is directly from the matrix website:

  Empowering the end-user
    The user should be able to choose the server and clients they use
    The user should be able to control how private their communication is
    The user should know precisely where their data is stored

This means the user has to be informed about servers and clients, the user has to be informed about communication privacy levels, and the user needs to be informed about what data is and where it lives. While that might be nice, it's even more nice to trust someone to solve these problems for you so you can best think about how to spend quality time with the people you appreciate in your life, not the security level of your data.

`sum(dilemma) + sum(onboarding/required knowledge) <= resistance`. anything for which `reward < resistance` will probably not succeed. That's my calculus. Minimize choice, minimize required knowledge/onboarding, maximize reward. That is a winning combo. I don't think matrix is optimizing for that. I guess we'll see if I have to find the flaw in my reasoning or not in 5 years.

The flaw in your reasoning is that Matrix is like the web, and Element is like a browser. Just as mainstream folks don’t say “go look at my Chrome site”, but are smart enough to say “go look at my website with your web browser”, the same goes for Matrix too.

The only reason this doesn’t happen yet is that Matrix clients like Element are still too geeky, and joe public doesn’t care about the advantages of open decentralised e2ee comms. Our plan to fix that is to transform Element’s UX; hide the decentralisation complexities, and let it punch its own weight againdt the centralised alternatives - https://matrix.org/blog/2022/08/15/the-matrix-summer-special... has more details.

The user would still need to understand it’s talking on an open network though, because that’s what it is. But they don’t need to care that much about it.

FWIW, everybody has different results on Google. They vary between mobile and desktop, regionally, and can even be personalized.

Are you saying your results are different?

Yes. They can and will vary significantly. I'm in the UK FWIW.