Email or username:

Password:

Forgot your password?
Christine Lemmer-Webber

Three weeks ago I wrote "How decentralized is Bluesky really?" dustycloud.org/blog/how-decent

Shortly thereafter, @bnewbold wrote his response: whtwnd.com/bnewbold.net/3lbvbt

I have written my (final) response blogpost: dustycloud.org/blog/re-re-blue

And as last time, 🧵. Buckle up.

50 comments
Christine Lemmer-Webber

So first of all, thank you @bnewbold for not only encouraging me to write my first piece, but being so extremely lovely in your response. I genuinely mean that.

And I'd like to go meta for a moment before we get into everything I have to say, because the discourse has been interesting.

Christine Lemmer-Webber

I tried really hard to be fair and not mean in my writing last time. It is, genuinely, the way I prefer to behave. Nonetheless, I was expecting a lot of negative pushback, maybe from the Bluesky staff, but especially either from Bluesky's team or fans or actually even the fediverse.

Christine Lemmer-Webber

These days, one kinda gets the sense that "civil discourse" is dead, and maybe even undesirable. Everyone else is unpleasant, and if you even try to be thoughtful, you're going to be taken advantage of or actually dragged for it, so don't bother.

This exchange has made me feel hope that's not true.

Christine Lemmer-Webber

I was critical of both Bluesky and the existing fediverse in my writeup, though I was more critical in my analysis of Bluesky. Nonetheless, Bluesky's team by and large were encouraging and thoughtful and expressed appreciation that *I* was attempting to be thoughtful. Grateful for that.

Christine Lemmer-Webber

Communities on both Bluesky and the fediverse were also receptive, much more than one would expect. This was nice to see.

I did get a few criticisms that I was too harsh or even too nice, but not too many. In the aggregate, reception was positive.

Christine Lemmer-Webber

As with last time, I'm doing these threads on both Bluesky and the fediverse. This time I'm putting the links to each of you earlier in the thread. Feel free to see what the other side is saying:

bsky.app/profile/dustyweb.bsky
social.coop/@cwebber/113647109

Christine Lemmer-Webber

I'll also say that I've been doing this in an incredibly stressful time, *on top of* a mountain of work.

We're doing a fundraiser btw over at @spritely. If some of the things I talk about resonate with you, maybe look at our work and consider donating: spritely.institute/donate/

Christine Lemmer-Webber

It's difficult to know how much I should talk about @spritely. The previous analysis wasn't about it, but it came up a little because our work there aligns directly with where things should go. For the most part, neither piece is about it. But it will come up more later on this time.

Christine Lemmer-Webber

Anyway, for the most part the reason it'll come up more later is because @bnewbold specifically put an ask for me to talk about it in his piece. So it will, but later.

In the meanwhile, what I'm trying to say is that I'm a very tired and overworked woman and these are my words, not my work's.

Christine Lemmer-Webber

So the title of the piece I just wrote, yeah? It's "Re: Re: Bluesky and Decentralization" dustycloud.org/blog/re-re-blue

But the reason for that title is that I thought it was kind of funny that I was replying inline to a lot of @bnewbold's blogpost like it was an email thread

Christine Lemmer-Webber replied to Christine

Of course whenever I get some email subject like "Re: Fwd: Re: Re: Fwd: Re:" I'm like oh god I really don't want to read this thing and I thought, maybe I shouldn't name this blogpost that

But @mlinksva told me that he thought it was funny and perfect and to keep it and so I did

Christine Lemmer-Webber replied to Christine

And so here we go, oh god again, not another reply in the email thread right?

Well for my part this is the end, I am not *blogging* about this again, not on my blog

Well not unless something *extremely* saucy and impossible to ignore happens

Otherwise else this is it. THIS IS IT I PROMISE

Christine Lemmer-Webber replied to Christine

Also last time I hid three easter eggs in the thread, and I did that just because it was getting so *long* I thought it would be funny if I got replies that said "I found the egg" and at the end said "congrats you found all three eggs you collected the egg triforce, you can defeat gender ganon"

Christine Lemmer-Webber replied to Christine

This ended up turning out to be an interesting metric, one I was too lazy to *formally* measure, but I guess someone else could

Friends I gotta tell you people loved the eggs in both places but the fediverse did way better on collecting the eggs

Congratulations fediverse, you beat gender ganon

Christine Lemmer-Webber replied to Christine

But one way or another people did, after all, seem to actually read deep in the blogpost and especially deep in the thread, which leads to a question, why am I doing this to myself, why am I re-articulating what's already a long-ass blogpost into a long-ass thread?

WHY AM I DOING IT AGAIN

Christine Lemmer-Webber replied to Christine

Well the last blogpost was 24 pages printed and took me 8 hours to summarize and this one is definitely shorter, it's only..

oh shit it's 20 pages oh fuck

okay I've been awake since 4am and getting this post ready since 5am working 12 hour days lately haha it's no problem

I'm gonna need more tea.

Christine Lemmer-Webber replied to Christine

One more meta thing before we get into it. I want to say up front: working on and building systems is deeply personal, and deeply emotional.

We are talking about systems people pour their lives into. Trust me I know.

It's worth doing a critique, but it's also worth acknowledging the human aspect.

Christine Lemmer-Webber replied to Christine

So anyway. We are going to get into it. And the further we go, the more serious my critiques are going to get. Just because I'm being nice doesn't mean my analysis won't be harsh at points.

Nonetheless let's try to be decent to each other. We all deserve that. Thanks. 💜

Christine Lemmer-Webber replied to Christine

If you can't tell, my narration style buries the lede quite a bit. So lemme tell you what you can expect up front:

- Some fluffier bits I thought were good framing
- A dive-into-literature analysis of the terminology "decentralization" and "federation" and whether Bluesky is either

(cotd...)

Christine Lemmer-Webber replied to Christine

- A defense and analysis of my claim that Bluesky's approach explodes unsustainably ("quadratically") as you try to decentralize it
- "Actually what I'm worried about portable identity isn't what you may think"
- Public content vs community expectations on Bluesky

(cotd ...)

Christine Lemmer-Webber replied to Christine

- A comparison of Bluesky, ActivityPub, and @spritely's values and design goals (invited by @bnewbold specifically in his post! good idea)
- Stepping back and concluding: "Where to from here?"

So yeah. That's a lot to get through. Let's get going.

Christine Lemmer-Webber replied to Christine

Oh yeah and just like last time I WILL NOT BE READING NOTIFICATIONS until I get through this

Sorry, it's impossible. I'll never get through it otherwise. So see your comments on the other side!

Christine Lemmer-Webber replied to Christine

So yes, definitely burying the lede, we are now at "Interesting notes and helpful acknowledgements"

Bear with me, we'll get to the deeper analysis, but this does help frame things I promise

Christine Lemmer-Webber replied to Christine

So first of all, @bnewbold was very nice:

> I am so *happy* and grateful that Christine took the time to write up her thoughts and put them out in public. Her writing sheds light on substantive differences between protocols and projects, and raises the bar on analysis in this space.

Aw thx 💜

solastalgia kris replied to Christine

@cwebber

It is written, only fedilink can defeat gender ganon

julianproxy replied to Christine

@cwebber Defeating gender Ganon was the best!

Dan Sugalski replied to Christine

@cwebber @bnewbold It's cool to see this kind of reasoned back-and-forth on design, with the various intentions, requirements, non-requirements, and engineering compromises run through. I really appreciate reading it, and hopefully this helps at least some folks come to the realization that there are many ways to come at complex systems and how changes in requirements (or sometimes seemingly random small choices) can make large differences in the design you ultimately end up with.

The Duke of Fall :d6:

@cwebber A vital prerequisite to civil discourse is a level of comfort with possibly being wrong and admitting it.

Coincidentally, that level of emotional and intellectual maturity has been sanded off our collective psyche by a quarter century of adversarial social engineering. 😫

Andrew

@cwebber You don't get enough "engagement" if you don't generate enough "outrage bait", it seems. 🙁

Suspiciously kitsune shaped Amber

@cwebber@social.coop i'll be attending from this instance because my main one (transfem.social) has an issue where every time i post we get hundreds of requests per second and im still working on scaling ​:akko_woozy:​

Nemo_bis 🌈

@cwebber "But seriously though, could open source orgs have some of that fighter jet wing money?"

That's what MIT is for, no? :)

Mark Kraft

@cwebber @bnewbold Your critique seems fair, if not generous.

"Bluesky is a Public Benefit Corporation, which means..."
...arguably less than actual promises to the users, which are easily bent and broken, at little reputational cost.

One thing that LiveJournal did early on was have a list of about ten promises to its users on its website. Between about '00 - '03, these promises were repeatedly watered down, bent, and broken, until dispatched entirely with the VC-backed SixApart acquisition. 🧵

Mark Kraft

@cwebber @bnewbold From what I can see, public benefit companies going public are a new trend. Laureate Education did so, for instance.

But despite its lofty principles, they were involved in a big scandal, this one not long after their incorporation.
(Machine translated from Portuguese.)

Ultimately, money and profit make the ethical decisions in VC-backed companies. /2

apublica-org.translate.goog/20

@cwebber @bnewbold From what I can see, public benefit companies going public are a new trend. Laureate Education did so, for instance.

But despite its lofty principles, they were involved in a big scandal, this one not long after their incorporation.
(Machine translated from Portuguese.)

Ultimately, money and profit make the ethical decisions in VC-backed companies. /2

apublica-org.translate.goog/20

Mark Kraft

@cwebber @bnewbold Can we trust Bluesky to adhere to their lack of clear promises?

Dunno. Can we trust OpenAI?!
/end

Mark Kraft

@cwebber @bnewbold On a more structural level, bsky's ATProto sounds more like a competition killer, in that adoption of it as the requirements for entry exponentially expand pretty much require corporate / competitive ambitions... which tends to get ugly, fast.

This brings up some ugly aspects of LJ server's history. The software was initially fully open source. Technically, it was always possible to tie LJ server sites together in a meaningful way.

But... /1

Mark Kraft

@cwebber @bnewbold ...the founder repeatedly made decisions undermining them, rejecting relevant code contributions, etc.

One of the things I argued for is that LJ server should focus on being easier and more distributed, to the point that a lot of the early blogging systems which didn't have robust commenting features at that time could use it to host commenting.

I viewed a very large opportunity there for open source adoption, and a reliable % of people contributing, if asked.

No dice. /end

Gos de l'hortolà :estelada:

@cwebber

Thanks for that!!

I just read the blog version.. so while it's fresh in my head I have 3 quick comments: one "typo", one "really?!" one "Ah!"

1) "group off people"

2) self-hosting email has become a hard thing? Grrrr

3) square-law means it is against interests of current nodes to encourage others

@bnewbold

EVHaste

@cwebber @bnewbold Still working my way through this one but it’s been a fun read so far. I particularly enjoyed your thoughts on Bluesky’s assumption of being adversarial to itself in the future, and I’m also wondering where the whole ad thing is going to go.

Strypey

"I asked if the piece seemed mean to her, because she was one of the people I kept in mind as 'someone I wouldn't want to be upset with me when reading the article', and she said something like 'I think the only way in which you could be considered mean was that you were unrelentingly accurate in your technical analysis'."

@cwebber

dustycloud.org/blog/re-re-blue

What a great compliment on your writing! I suspect I often fail, but this is always what I'm aiming for in my analysis of tech.

"I asked if the piece seemed mean to her, because she was one of the people I kept in mind as 'someone I wouldn't want to be upset with me when reading the article', and she said something like 'I think the only way in which you could be considered mean was that you were unrelentingly accurate in your technical analysis'."

Matěj Cepl 🇪🇺 🇨🇿 🇺🇦

@cwebber @bnewbold

I still hold that the best explanation of the opposite of “message passing” systems is to call them “Usenet-like” ones.

bryan newbold

@cwebber thanks for the reply!

really happy for use to have these back-and-forth blog posts as an artifact. it sounds like it took a lot of time on your end, and I know it did on mine as well, but doing it "in the thick of things" made it real and memorable.

I read the whole blog reply, but likely won't read the full summary thread (different media!).

I'm not planning to do a full-length blog reply, but will post a couple quick comments here. and do hope to "be in conversation" going forward.

Erik

@cwebber I'm not a techy. Last time I tried to program anything was a text based adventure on a C64 in the earliest 90s. Gave up after that. Yet I've been able to read this entire text because it was so well-written and clear, even to a language teacher with only very mild interest in networking. Thanks! I couldn't get halfway through Newbold's text.

As a linguist, though, I'm curious as to why Newbold *chooses* to use incomplete definitions: has that use been informed by crypto culture?

Reply Guy

@cwebber @bnewbold Thank you for sharing this conversation, it prompted me to map your terminology into economy. Centralization = monopoly. Credible exit = free market or at least oligopoly. Fully federated = own your means of communication (ownership, not socialism!). Since communication enables public discourse, it maps further into politics: centralization = absolute worldwide monarchy, credible exit = nation states with absolute monarchies, federation = heterogeneous orgs participating.

Ascendor

@cwebber @bnewbold

I have been online for some 25 years or so. This is one of the best, most respectful and well-prepared debates I have ever seen: social.coop/@cwebber/113647109

All three blog entries have been worth every minute of time: I know of no example where people criticized on such a high level without getting personal.
Thanks so much, it gave me hope (while learning a lot about #Fediverse's and #Bluesky's fundamental architecture.

Ascendor

@cwebber @bnewbold Deep respect and a big "thank you" to both participants.

Eliot Lash

@cwebber Thanks for another great write-up. I chose to read this in blog format again. So I don't think I have any easter eggs to prove I finished it, you'll have to take my word for it. :)

⊢≪ Noctilua ≫⊣

@cwebber @bnewbold

And this I see just in time when I started to wonder about Bluesky’s architecture and it’s capabilities. The article explains a lot, thanks.