The key difference between SMTP and ActivityPub is that SMTP arrived on the scene in an era when getting started on the internet (sorry, ARPANET) at all was a major technical feat. The median internet user in the two decades following the introduction of SMTP was plenty competent enough to handle the complexities of email, and once email became the standard way to communicate everyone had to figure it out. At first it was all that we had, and by the time centralized platforms started making things easier the network effects were so strong that they had to support SMTP.
Some people are putting Mastodon forward as a replacement for a centralized platform, but now the network effects are working in the same direction as the complexity friction—both are pushing you towards centralized platforms and away from Mastodon and company. For most people there's no incentive to overcome the barriers because they would be the first in their social circles to attempt it, and the centralized platforms so far have had little motivation to support ActivityPub because their users don't expect to interact with that ecosystem. If Threads really does implement ActivityPub I could see that changing, but that remains to be seen.
This doesn't mean that Mastodon doesn't have a place right now—if you're looking for a community that is specifically filtered to exclude those who aren't above average in technical competency, it'll do nicely. But it cannot serve the median internet user today.
The warning is necessary, so I'm not disagreeing with you per se... But I think it's a positive quality of the piece that it doesn't actually resolve to a thesis about the comparison (as might "email is fine, and therefore..."), but just points out the tension. It's up to the reader to use the comparison to take a moment to consider what they think the costs and benefits involved are. Plenty of folks (well, judging from comments sections) think that email fails at many of the things it claims to do, and would assent to an "Everyone sucks here" judgment about both email and the Fediverse.
Half the comments on long blog posts usually end up discussing the headline, or even just part of it. Combine that with the difficulty people on online forums have to recognise blatant satire and you'll end up with pages of discussion about SMTP before you know it.
I see at least four top level comments failing to see the satire or talking about email. The downvotes are pushing them down, but it shows the sarcasm warning was necessary…
I suppose I understand what the author's position is, but they wrote a long post explaining that, since we have a flawed system, we are allowed to create new systems that are as flawed as that one... and that doesn't sound too good
There's one big misconception here which really tears down the entire analogy from the beginning: you don't really choose your email server, and you were never intended to.
When email was developed, your email address would have been comprised of the computer you used (email is quite old) and your account name on said computer. As computers proliferated, the "computer" instead becomes "institution providing you with internet access", which would be your employer or your university (if a student), or your ISP if you manage to get home internet access. The email address is thus fundamentally tied to your source of internet [1], which also made it a useful unique identifier for the Web. But the desire for stability of this across ISP changes, as well as the need for multiple email accounts for a single household (kids don't want to use their parents' email address), propels the need for webhosted email accounts, email identities not specifically tied to your source of internet. And these email providers should be seen as exceptions to the general rule that email comes from your internet provider (although they probably provide the majority of email addresses now).
So the decentralization of email doesn't really create a "what server do I choose?" question, because the server is already chosen for you. And to the extent that the question does come up (when choosing a personal, permanent email provider), the evidence from the field is that the vast, vast, vast majority of people settle down for very few providers for this question (Gmail is close to a majority of personal email accounts, if not already there). From a user's perspective, email is not decentralized, and the confusion that Mastodon's decentralization brings isn't really an issue in email.
[1] And if you look up the version of email that was actually properly designed from the ground up as email (X.400), rather than accreting over time as SMTP et al did, it's even more apparent that email addresses are intended to come from your source of internet.
The email comparison often confuses people in my experience. They simply seem forget that email is federated when it comes to discussions like this.
I usually see these arguments from people who never had any interest in even trying Mastodon (or a related service).
In practice, there are only three large providers (Apple, Google, Microsoft, not necessarily in that order) that control whether or not your email gets delivered, so the comparison doesn't work entirely if you look at the modern web.
There also aren't that many Mastodon spam filters out there, but then again I haven't seen the same amounts of spam on there. I'm sure it'll get worse once Threads starts interacting with other servers and more people join the network, though.
Mastodon does solve the authenticity problem that SPF, DKIM, and DMARC have yet to solve in email. When you receive a message from steve@example.com, you don't need to dig through the headers to make sure it wasn't a Gmail signed envelope from some entirely different domain that happened to use the same server infrastructure like you need to with email.
I imagine if you tried “releasing” IMAP and SMTP email (with whatever cleanups and encryption you wanted to choose) now as a new open standard, in a world where email never existed, there is no chance it would take off.
My theory is that email took off because of how well it used already existing concepts from paper mail. People picked up the concepts behind paper mail as children. New concepts are hard and I think we tend to overlook that fact.
> Oh and even better, apparently Gmail is starting to lock down access to your emails with only a username and password in the interest of "security", so you have to go through convoluted hoops in order to check your email in something that isn't the gmail web UI. It is literally impossible for me to check my work mail in something like Aerc.
Application specific passwords. If you have 2FA enabled on your Google account, you can create an app password for your account to use a client like Aerc or whatever.
You also need to have 2fa set up. And they don't have totp as an option unless you also set up either a security key or any number of options (sms, mobile push, others?) that involve getting entangled with google again.
Google really prefers you'd use sms 2fa over totp.
It's probably meant ironically (I see more reasonable date formats elsewhere on the site), but Americans will truly do anything to avoid using international standards.
Comparing Mastodon (or ActivityPub-related endeavors) to email is silly.
Mastodon is an open platform where people can post things publicly.
Email is private communication between individuals.
Petty disputes exist on Mastodon because admins can see what's being posted and talked about in other instances and so can choose to ban them just for personal spite.
How about your instance admin peeping into everyone's DMs and seeing which servers hold the most people talking against the Democratic party and then blocking all those instances?
Would you like that?
Well, that's what would need to happen with email to achieve a same 1:1 comparison
Are you saying email administrators are somehow above this all? They absolutely make weird personal choices when it comes to blocking servers and users. Microsoft will purge your account if it detects what appears to be a names child.
Take, for example, email delivery: if you're simply running a Tor relay (not an exit node) on the same IP address. You'll end up on a weird lists of blacklists not because there's any chance of abuse, but simply because they don't like Tor (or because they're too incompetent to distinguish between the two or too lazy to look up what those IP address lists mean).
The biggest difference between Mastodon and email is that you're less likely to interact with the petty administrators because people send fewer emails than they send toots. I promise you that free, community run email servers will be just as petty, shady, or unreliable as any hobbyist Mastodon instance.
How is you private life have way less data then you public life?
Email accounts have waaaaayy more information about you than a public social media account. I can find out your private thoughts, the things you buy, who you owe money too, your social graph, tons and tons of data. I can find out way more about than from some comments on a nerd forum.
If an email admin looks into your private correspondence your first thought is to get the hell out of his servers, whereas if a mastodon admin constantly monitors public posts from your and other instances then it doesn't seem like a big deal
I thought this was actually about Mastodon given that people already have experience with email, and it was trying to draw parallels with the success of email. But it just ends by giving up..?
I seems to be missing the mark and satirizing the users for not understanding or liking a poor user experience, when it would make more sense to satirize the platform itself for failing to appropriately consider that when given a choice, users aren’t going to choose the difficult and annoying thing, but rather the simple and understandable one.
Criticisms of mastodon/lemmy/fediverse are 100% warranted. People try it and give up because it’s a confusing mess compared to any alternative. So either the product is so complex it can’t be simplified, or the developers of the project are so ignorant to user feedback that they aren’t willing to prioritize simplifying it. Both of these do not bode well for long term success or wide adoption.
There is also this sense I get when interacting with some fediverse enthusiasts that perhaps the complexity is good because it keeps the communities small, or if you can’t figure it out you shouldn’t use it. All I have to say about that perspective is that it’s pretty damn pretentious, as well as at odds with the messaging that these are meant to be replacement services for widely adopted social networks.
> So either the product is so complex it can’t be simplified, or the developers of the project are so ignorant to user feedback that they aren’t willing to prioritize simplifying it. Both of these do not bode well for long term success or wide adoption.
The fediverse is not a product. The fediverse is an organization of humans willing to interact together. There are dozens of softwares, documentation on how it works and how to migrate, tens of thousands of instances each with different rules. The fediverse is a universe, not a plastic-wrapped package you buy and consume. The goal is not to capture the population as fast as possible, as seems to be the only valid metric on here, because that would mean coming back to where we started and not having advanced a single step
> if you can’t figure it out you shouldn’t use it
If you don't want to take the time to understand what it is, then you maybe you shouldn't use it. It's like coming to the house of a stranger: you don't know them, you try to understand the rules, ask around, maybe remove your shoes at the door or at least make sure it's ok if you keep them. But if you expect to enter without saying a word, picking beers in the fridge and shouting at people that's just rude. The fediverse is built to prevent exactly that.
> these are meant to be replacement services for widely adopted social networks.
No, there shouldn't be a replacement for social networks, the same way the web is not a replacement for something else: the web is one application that equalizes access and publication. The fediverse is a universe for everyone to have a voice and discuss and organize. Make conversing easy by giving back control to people.
> The fediverse is a universe for everyone to have a voice and discuss and organize. Make conversing easy by giving back control to people.
I think my main issue is that while this is great idealism, it’s fundamentally hypocritical in practice. The fediverse is not a place for everyone when the vast majority of people do not understand how or why to use it. It is not accessible across cultural barriers, and it’s not accessible to those with lower levels of technological literacy.
Bitcoin proponents play the same game about decentralization: centralized industries no longer in control, power to the people, etc. Except instead of just centralized power to a new group, whoever ended up being early adopters.
The fediverse is a small group of technologically literate folks doing the same thing, but with social networks. It’s an (intentional or otherwise) homage to the old web, before Eternal September, which I totally understand the desire for. But it claims to be a new voice of the people, when really it’s just a new platform for the voice of the few. Those fed up with the actual voice of the masses (it turns out most people don’t have much interesting to say, and probably don’t agree with you). It instead replaces the mass of voices with a smaller, pre-filtered set of voices. It is its own echo chamber.
We certainly need viable competition to the massively centralized, corporate controlled internet that we live in today. The fediverse is a good experiment, and quite likely in the right direction. But it’s activists and participants need to a reckon it’s actual nature rather than the idealist model they claim it to be. Right now there’s a whole lot of drinking of the kool aid, which will only be to its detriment long term.
If the fediverse is a universe for everyone, then it should be accessibility über alles
I understand what you mean and I fully agree. We want everyone to get rid of capitalist networks, take matter in their hands and decide by themselves. It's not far from democracy actually. But it's totally true that doing so takes time and energy that only a minority possesses, and learning how to use the fediverse is an activity itself.
But contrary to what you say, there is an ever growing amount of resources and orgs that take the active role of welcoming and guiding people into the network. This is where change happens: by teaching others, by taking time with them, by showing them, opening an account for them, having a space where they can come. It takes a lot of time and it's why the growth of open networks isn't as fast, but it's ok because that's not the point. The point is to make it together. It's not a technical problem, it's a social problem, and we're going to solve it with social solutions, not more javascript.
So, yes, it is only self-learnable by an elite. But no, it doesn't mean it is only for the elite. It just means tech isn't the only way.
I don't think it's satirizing the users; it successfully highlights an existing problem of people leaving before trying, and I agree with this highlight as an avid user of the Fediverse.
> Every mailing list has their own unspoken rules on how you're supposed to use your client, reply to emails, and more. If you break these rules, you're threatened with banning.
It's an interesting and mostly correct article overall, but I must come to the defence of email here.
I would say this is a rather unfair characterisation of Internet mailing lists. I've been a member of dozens, if not hundreds, of mailing lists; I have never seen a member threatened with a ban, let alone being ejected. This may have been the predominant culture twenty years ago (and the archives witness that), but times have changed.
I get that this is a satire post on the usual criticisms of Mastodon, but honestly the points on Email is valid, and the reasons are a big point on why non-business Email usage is declining…
Especially the points 'there's no real way to migrate between email servers either, so whatever I choose is going to be my permanent home' and 'I can't run my own server' parts are very valid criticisms of the Email system…
I do like the way BlueSky lets you use your own domain much like hosted email providers do - presumably your account should be migrateable to other AT protocol systems as a result (eventually, hopefully, caveat, caveat).
The whole thing is talking about the fediverse. Presumably if you wanted to set up a server of your own on there and run a spam filter it would need to be so large.
The satire may land better if the hyperbole was more exaggerated. The specs are crazy for personal use, but probably reasonable for heavy/enterprise tools. I left wondering if they were considering the latter; misguided people make mistakes like that.
What got so complicated? Things like reverse PTR and SPF are trivial. DKIM is a bother to set up but you do it once during an annoyed afternoon and never need to look at it again. DMARC is as easy to set up as DKIM.
> why isn't our mail just routed to a three-word address?
If only there was something that could take words like "grdotcloud gmail com" that we could use to deliver mail!
That's not my experience. MS Outlook seems to have it's off days, but email usually arrives just fine. That's even on Contabo's network, which is listed on that one "your ISP doesn't do enough to stop spam" blacklist Microsoft annoyingly refers to.
I don't see any spam in my inbox either, the little spam I do receive ends up in the spam box where it belongs.
Whatever mail setup Mailcow defaults to has worked great for me. Mailcow and mail-in-a-box have made the classic email setup very painless.
EDIT: didn't realize this was satire when I wrote this
Setting it up from scratch is extremely annoying but there are foss solutions which manages everything mentioned and more, you only need to install & add the specified DNS records. Modoboa does a great job at this: https://modoboa.org/en/
> Also apparently there's no real way to migrate between email servers either
This seems false (unless I'm misunderstanding), you can just setup a second mailserver, change DNS records and run IMAPsync to get everything transfered (backend-indepently), the only thing that would need additional consideration is passwords.
Haha. Email was a strange beast to understand back in the day. Of course, you needed to have access to the internet, your ISP, so you would default to their server. Whichever platform you were on had leading software, so you use the default. Not really that hard to figure out—it was all just new.
Since then, I’ve obtained my own domain name to have a ‘permanent’ (while I pay for the domain) email address.
Now I just forward these emails to my gmail account. I don’t have any hang-ups about google using my emails to train my AI overlord. Maybe the AI will recognize me like a country cousin?
This is the model I used: https://civitai.com/models/4468?modelVersionId=57618. I also used a few extra LoRA filters (remove bad features, foxgirl ears, etc) to make the image a bit better. The model is literally named Counterfeit.
Yeah what's with that? If you walk around in vrchat with an avatar that's non human you get this already. And it's actually very hard to find decent human avatars on there.
It's been a while, but as I recall your ISP will provide you with an email address. So if you have internet, you have an email address. Sure, if you change your ISP then you have to update your email address everywhere. But the same is true for phone numbers if you change phone companies.
As for email clients, just use whatever Netscape comes with.
Not sure where you live but in Canada and the USA the phone companies legally have to allow you to port your existing number over from one company to another.
I remember porting numbers becoming a thing and I was born in 1989 so it must have been late 90s earliest. In Finland though, so don't know about north america
Some people are putting Mastodon forward as a replacement for a centralized platform, but now the network effects are working in the same direction as the complexity friction—both are pushing you towards centralized platforms and away from Mastodon and company. For most people there's no incentive to overcome the barriers because they would be the first in their social circles to attempt it, and the centralized platforms so far have had little motivation to support ActivityPub because their users don't expect to interact with that ecosystem. If Threads really does implement ActivityPub I could see that changing, but that remains to be seen.
This doesn't mean that Mastodon doesn't have a place right now—if you're looking for a community that is specifically filtered to exclude those who aren't above average in technical competency, it'll do nicely. But it cannot serve the median internet user today.