Replacing one fallacy with another
Replacing one fallacy with another
Posted Mar 4, 2019 23:38 UTC (Mon) by pjhacnau (subscriber, #4223)Parent article: Rosenzweig: The federation fallacy
a) I'd hardly call it "lengthy" - took me 5 minutes to read. (For reference I would consider "lengthy" to start at around 20 minutes for a first pass) - and that's a criticism of the expectations raised by the lead-in rather than the article.
b) It degenerates rapidly (IMO) into under-defined jargon - e.g. "information dictatorship" and "information anarchy".
c) (the point I'm going to expand on) It talks about federation as a "fallacy" then presents a false dichotomy.
Let's look at what I consider the key point in the article:
"
In reality, guess how many instances encompass half of the user base. Maybe 1,000? Alright, there are some big instances in there, so perhaps 100? Well, there are a lot of really tiny instances mixed in, so possibly only 20?
The answer?
Three.
Just three instances encompass 50.8% of users.
"
So there are only three main nodes. Well, that's still 300% of the externally visible nodes of Facebook, Twitter, Instagram, ....
Also, that covers 50.8% of users. That means that 49.2% of users are on nodes other than those three.
Likewise email. Yep Google Mail has a huge hunk of email users. But there are other large servers, and plenty of smaller servers.
The false dichotomy is the implicit message I got from the first half: "A large number of roughly equal weighted is federated, any other distribution is centralization". As far as I'm concerned Mastadon _is_ still federated. People choosing not to manage their own personal server, and instead use someone else's server (either for free or for money) does not affect whether something is federated or not. The protocol is open and any individual or organization could set up new nodes and create a significant redistribution of users. A point that's actually made further down the article:
"True, Mastodon is de facto centralised, but despite the size of the largest instances, it retains the ability to federate with other Mastodon instances. Further, Mastodon is able to federate with other free software friendly networks via a pair of common protocols, creating the familial fabric of the “fediverse”. Centralization and federation can certainly co-exist in harmony to improve efficiency while retaining user choice."
So which is it? Is federation "dead" or is it still there?
There's major issues for federated systems; last time I checked Mastadon had not displaced Twitter for the general public, and to me _that_ is a bigger problem than how many federated Mastadon nodes are popular. The Web is also a federated system. I would argue that the inherent federated nature is a reason things are not 100% centralized now. Matrix offers another interesting federated option; it will be interesting to see how the adoption by the "French Ministry of Digital" plays out.
I would summarize the last section generally as "the issue is less the technology; the culture we build on top of whatever platform is at least as important". A gross simplification I acknowledge, but I was struggling for a phrase longer than "information democracy" and short enough to fit here. A federated system doesn't necessarily ensure freedom, but it makes it much easier to construct freedom, centralization of servers or not. In that I _think_ we are in agreement.
So I actually agree with where the article finishes, I just think the whole "federation is dead", "federation is a fallacy" was over-played and a distraction from the good bits.
Posted Mar 5, 2019 2:42 UTC (Tue)
by ejr (subscriber, #51652)
[Link] (9 responses)
Posted Mar 5, 2019 3:07 UTC (Tue)
by areilly (subscriber, #87829)
[Link] (2 responses)
Posted Mar 5, 2019 4:32 UTC (Tue)
by alyssa (guest, #130775)
[Link] (1 responses)
It's not a perfect measure of activity across the fediverse, but seeing as status updates are the primary function of a microblogging platform, it should give a pretty close estimate.
Posted Mar 6, 2019 2:40 UTC (Wed)
by areilly (subscriber, #87829)
[Link]
The alternative to taking a cut-down feed (which is fractional de-federation, I suppose) is to have an account on an enormous server, where the feed volume is aggregated between the active users, and the message database is physically shared.
WhatsApp might just be XMPP, but it works because it's single node, an enormous SMP system has been competently engineered and refined to the limits of current technology. Cleaving that into federated pieces would necessarily complicate and slow down message delivery and introduce all sorts of synchronization and protocol issues. Seems like a reasonable design trade-off, if you can make it work. More secure than SMS and it's inter-exchange signalling issues.
The multi-dimensional discrepancies between network terms-of-service, national speech laws and population-group sensibilities is one of the more interesting struggles of our age, IMO.
Posted Mar 5, 2019 5:01 UTC (Tue)
by pjhacnau (subscriber, #4223)
[Link] (5 responses)
Posted Mar 5, 2019 5:06 UTC (Tue)
by pjhacnau (subscriber, #4223)
[Link] (4 responses)
3 servers have 51.8% of the users. From a followup comment I gather 5 servers have 51.4% of the traffic. So I can accept that there are 3-5 servers that matter and the rest are of rapidly diminishing significance. I'd still call 3-5 servers a federation.
Posted Mar 5, 2019 15:03 UTC (Tue)
by ejr (subscriber, #51652)
[Link] (3 responses)
Posted Mar 5, 2019 16:51 UTC (Tue)
by dan_a (guest, #5325)
[Link]
Today, it's not that difficult to start running an email server but there is an increasing difficulty getting your emails accepted by other servers due to spam protections.
In the world with 5 big email services, if you can start running your own email server and start sending emails to people on all 5 of those services and receive replies then it's still open and federated and that's all good. On the other hand, if the mail from your little email server gets rejected by the big 5 (because anyone who is not one of them must be a spammer) then it's no longer an open and federated service, but a closed federation which you can't join in with, and that's not good.
I don't know how the Mastodon world works, but to my mind as long as anyone has the ability to join the federation then it doesn't really matter if there's a big concentration of users on a small number of servers.
Posted Mar 5, 2019 22:14 UTC (Tue)
by pjhacnau (subscriber, #4223)
[Link] (1 responses)
a-ok? No. Still Federated? Yes. Scope for improvement? Definitely. Likelyhood of improvement? ............
Longer answer:
I'd consider only having gmail , o365 and yahoo still a better situation than the other "messaging" options. I think email is in an "interesting" situation as it is, AFAIK, the longest still-used federated protocol. Firstly the whole barrier-to-entry issue has, and continues to, whittle away at the federated nature of e-mail. But it's hanging on by its fingernails. Part of the problem is that the original email protocols have . . . er . . . "issues" and most attempts to fix them hit vested interests and/or undermine the original federated nature. But right now I can (and do) run my own email sever. After having to switch to the NBN I lost my static IP - DDNS to the rescue. Sub-optima but useable. I then found that my IP address was being allocated from a pool which is blacklisted (all of it - thanks Telstra). OK, so outgoing email gets routed via a freind's hosted email server and I've set up SPF records so that everyone[1] accepts the incoming mail. Would a sane person bother with all this? No. But . . . for the sane, and for me if/when I give up on all this, I can immediately think of an additional 3-5 big Australian commercial providers I could use for my email. Which for me (living in Australia) means my mail, for the most part, remains subject to the same set of laws the rest of my life is subject to.
Three points I'd pull out of that:
1) Email the protocols is a mess. Email the system is compromised on multiple levels, (usually) insecure, the "improvements" often create more pain than gain. But the "cures" are (IMHO) universally worse.
2) A Federated system makes it possible for commercial entities to jump in, make money, and have a big presence without automatically being in full control. Google doesn't really control mail in the same way that, say, Facebook controls Facebook, Facebook Messenger, and (now) WhatsApp. Federated systems have a built-in resistance to becoming a monopoly.
3) A Federated system doesn't guarantee freedom for users, "Digital Democracy" or anything else down that path. Just because a given federated system resists monopoly doesn't mean it can't become one. But it remains a (very) useful tool.
Separate to email I want to mention Mastadon again. In the context of the original article I feel like I'm really missing something. The author (and a couple of comments) seem to be trying to tear down something that, to me, isn't that big in the first place. As I said earlier, for me the biggest point of Mastadon (or Diaspora before it, or . . . ) isn't how "federated" it is (or isn't) but that it has really had no impact on the existing social media players. It's interesting, but doesn't really warrant huge praise or criticism. What gives?
[1] Ironically the _only_ problem I've had with that hasn't come from gmail, o365, or yahoo - it's actually a small Australian ISP who whinges about a reverse-DNS lookup issue.
Posted Mar 5, 2019 23:56 UTC (Tue)
by Garak (guest, #99377)
[Link]
Posted Mar 5, 2019 4:58 UTC (Tue)
by flussence (guest, #85566)
[Link] (2 responses)
Posted Mar 6, 2019 14:57 UTC (Wed)
by jkingweb (subscriber, #113039)
[Link]
Posted Mar 6, 2019 17:46 UTC (Wed)
by zaitcev (guest, #761)
[Link]
Also, allow me to interject for a moment, but
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
castles made of sand and fallacies made of straw
Separate to email I want to mention Mastadon again. In the context of the original article I feel like I'm really missing something. The author (and a couple of comments) seem to be trying to tear down something that, to me, isn't that big in the first place. As I said earlier, for me the biggest point of Mastadon (or Diaspora before it, or . . . ) isn't how "federated" it is (or isn't) but that it has really had no impact on the existing social media players. It's interesting, but doesn't really warrant huge praise or criticism. What gives?
To me, the biggest point of FOSS decentralized/federated communication platforms ('social media players') isn't about federatedness, or even impact on existing high-usercount platforms. To me the biggest point is Free Speech. From that perspective, the derived value of 'federatedness' comes into play. As each large-outlier member of the federation becomes a too-attractive chokepoint target for those who would wish to censor or 'shape the human terrain' of the free speech within that realm. While I still haven't read the whole article, I will point to the second paragraph-
Thus, permeating the community are calls for decentralisation. To attack the
information silos, corporate conglomerates, and governmental surveillance,
decentralisation calls for individuals to host servers for their own computing,
rather than defaulting on the servers of those rich in data.
First I think there is a fallacy with the phrasing of 'decentralisation' as an entity. It's perhaps more important to understand that the 'permeating calls for decentralization' are on a political spectrum as diverse as the spectrum of (in the US) democrats and republicans and libertarians and christians and muslims and jews. I.e. I think the false assumption is that the important aspect of those 'permeating calls' is maximization of decentralization, versus maximization of free speech capability and/or privacy concerns.
In general the article seems a bit agenda/narrative pushing to me. 'unbridled aura' in the first sentence triggers my suspicion that someone has a 'bridling' narrative/agenda they are pushing.
Immediately in the third graf we get
In the decentralised dream, every user hosts their own server. Every toddler
and grandmother is required to become their own system administrator.
This seems to be a strawman narrative attacking people with positions such as mine. Strawman because it mischaracterizes the 'call for decentralization' into an extreme 'every toddler and grandmonther is required to...'.
That reads to me like someone who is trying to smear positions such as mine that advocate (see other comments) every *adult* *have the option*(not the requirement) to become their own system administrator. And that such *options*(NOT REQUIREMENTS) are a critically necessary aspect of achieving real hard-line Free Speech on the global information superhighway.
My apologies to the author if the toddler/grandmother issue is something they are responding to, versus creating. But from the tone of the first few paragraphs I get the impression Rosenzweig is either pushing, or completely falling for the toddler/grandmother/sysadmin anti-home-server narrative.
Replacing one fallacy with another
Replacing one fallacy with another
Replacing one fallacy with another
https://zaitcev.livejournal.com/251546.html
