I'd say some friction can be a positive thing to a sytem. It SHOULD be hard to do bad things, or to spread potentially spam or toxic content, etc. Intentional friction or "circuit breakers" at those points may solve problems or prevent problems for users.
What I don't think helps anyone is things that become needless UX friction to helping good things happen that delight and bring benifits to users.
@tchambers@mastodonmigration@laurenshof@fediversenews@cheeaun@rolle@scottjenson I joined Mastodon in late 2022, I think it was when Elon bought Twitter. I found a list of servers, some of which were closed to new users because of the mass migration. I ended up applying to social.linux.pizza because I was running Linux Mint and it was open. I was accepted, then I started figuring out how to work it. After a while, I started using Phanpy.
Some of the things you say are defects or omissions might actually be features. I have been using computers for a long time, though I am not professionally involved with them. My first online experiences were on the GEnie BBS. I also don't use any of the typical social media. I had a Twitter account (no longer), but I only used it to follow the Ukraine war. I didn't mind having to figure out Mastodon. It meant that the users had to have a certain level of commitment and curiosity. I think it cuts down on the trolling and blathering. The necessary information is out there, if you look for it.
I look forward to part 2. In the meantime, I have two comments that someone might find useful.
First is that I was part of the November 2022 fuck Elon migration. That was before ExTwitter locked down their API, so someone had written a little bot that would search your follows there for anyone who had a fediverse handle in their profile (the odd double-@ format, I'm guessing). You could download that as a .CSV and upload it to Mastodon and have an instant large group of familiar people to follow.
Very different than starting from scratch or trying to guess what someone's handle might be and finding them here, even if you know they are in fact here.
I think that's huge for many people because unless you join a very active location-based or interest-based server, it'll feel really lonely for quite a while.
@Mikal Love the very descriptive phrase of the "fuck elon migration." And agree there were two different (at least two) tools that did what you say, and yes, those deeply helped....Am about 3/4th done with my second article and will add that bit of tech history to it. Thanks!
Looking forward to your ideas on how to improve things. Have you considered taking a page from role playing computer games. When you start a game you arrive on "newbie island" when you walk through a few simple tasks that familiarize you with the UI and your various options. But pretty soon you have to (get to!) move off into the real game equipt with some knowledge, the choices you've made, and a modicum of skill.
@tchambers very clear list and agreee on all of them. I hope most of those will be fixed on Mastodon by the end of the year (at least that’s the plan).
#3 is still one that I don’t have many doable ideas on how to fix properly 😞
Smithereen fixes the DM UX thing by actually treating direct messages as something entirely separate from posts. Different UIs, dedicated inbox and outbox pages, a button in each profile to send a message, addressing in a separate field like in an email client instead of mentions. I don't even allow setting per-post visibility on my side (but I do store and enforce it for incoming posts/comments)
We would like to fix it otherwise, by having the authoring software mark a link as being a link to a specific AP object so when displaying the status you can know it right away and do what is correct.
@renchap@tchambers I don’t think there’s one answer, but a big step would be to stop sending web users to unfamiliar servers in the first place. When I’m using my own server’s web interface, links to content on other servers need to be handled in the web interface I’m already using instead of linking directly to the other server.
@tchambers This is spot on. I'd like to trust the team that got all of this wrong to fix it. But, that seems unlikely. So, it seems like the solution you used, to manually help new people is what we're left with.
I see what you mean: I couldn't access the full thread of comments from the reply that you boosted.
About Sin No. 1, picking an instance in 2022 was quite puzzling, as I didn't want to intrude anywhere or step on anyone's toes. At that time, there was a strong attitude of "we don't want intruders." Maybe that general attitude has changed now?
Hmmmm. I think the Sins I list here are not as easily fixed as you suggest. And are fediverse wide (at least across many different fediverse offerings) -- will write up some fixes I do think will work in my next article.
@tchambers@mastodonmigration@laurenshof@cheeaun@rolle@scottjenson@zeldman Mastodon was always sold to the end user as the community when its actually just the software that can run a community. The term join Mastodon is fundamentally flawed, it's like saying join the world wide web instead of saying go to my web site. The emphasis should be building great communities that just happen to interooerate with each other. That shift in perception addressees a few of the great sins (def not all)
I only WISH I were too young to remember, but love your analogy.
Yep Web 1.0 had a ton of such sins like this. And that the open web worked those out gives me great hope that the open social web can too, and faster than before to boot!
@Mikal The #Fediverse "been built by and for users and not fashy billionaires, it requires more direct participation, more tolerance for all the rough edges" True, true.
Would also like to mention, regarding the article by @tchambers, of all the critiques of the Fediverse, Mastodon, etc, it's nice to see one coming from love instead of a tech journalist that is not willing to understand, at this point of development, the Fediverse, esp Mastodon, has "been built by and for users and not fashy billionaires, it requires more direct participation, more tolerance for all the rough edges"
I think that is why the community is taking this critique to heart in the comments.
The way I explain the fediverse is to compare it to virtually any non-prepackaged social relationship.
Want to try a worker-owned business? A communal household? Co-housing community? Nonmonogamy or other non-traditional romance or relationship forms? CSA or other direct farm-to-consumer food production? Any organization or group that runs by consensus?
Virtually all of my lefty friends have some experience with one or more of these things, so they understand that any time you step outside of the frictionless, pre-made capitalist box to create the life you want, it will be more difficult and often awkward, time consuming and confusing. And probably require a bunch of meetings. But maybe worth it in the end.
That's how I explain the fediverse. Because it's been built by and for users and not fashy billionaires, it requires more direct participation, more tolerance for all the rough edges, longer to make changes and improvements. They usually get that, even if they ultimately decide, "nah."
Great writeup of the Original Sina, Tim! And behind these ones on the list is a long list of the Thousand Papercuts, from misplacing mentions at start of the reply body text (when a perfectly good mention field exists in the protocol) to inability to remember the readers' position in their (reverse chron and random) timeline. The papercuts could be solved by the client, but increase friction all the same. @tchambers
I am a SysAdmin for small to medium businesses in Edmonton who need a website, cheaper email hosting etc. I have no clue how much of a need there is for a regional social media platform. But I have a computer powerful enough to host 1,000 or so people, so that they can move away from Facebook. That is good enough to start and see how much of a need there is. I am not trying to snitch users from other instances of Fediverse, lets federate.
I see how badly my previous post was worded. I shouldn't have rushed to press reply before going into the tunnel. What I meant was "Do you think running a city-wide regional instance would exaggerate the issue of centralization? It would basically act like a subreddit for Edmonton, and can be interpreted as an echo chamber by some people. What would I need to do to avoid the mistake of creating echo chamber or other mistakes the mainstream social media platforms done? Is federating with other instances such as Qlub.social , NWT.social , toronto.place , mstdn.social , thecanadian.social and more enough? I am asking because I was born into the era of mainstream social media platforms and my world view might be more different than most of you, and I didn't get to experience pre-socialmedia times. I want to be part of the solution without repeating the same mistakes".
Note: "You don't need to download an app" is surprisingly effective way to market it. So many people looks surprised that this thing works and functions amazingly well on a regular boring internet browser, as if it was on a purpose built app. That is Edmontonian Social's audience.
Asking as someone who was a cry baby at 90s, born into the social media era, and having somewhat of difficulty visualizing a better world, and still has the audacity to wish for one.
@mike@tchambers@mastodonmigration@laurenshof@cheeaun@rolle@scottjenson@zeldman Hey! I’m an old too! GEN X REPRESENT! From a Vic20 and my friend’s TRS-80 to my first internet connection back in 91 managing a computer lab at USF in Tampa. Tis twas a different time. Decentralized is simply “rejoining” the internet the way it was (and probably should’ve stayed). Biggest regret was centralizing information to social media companies.