Post AIdDeP8c2lwt4s5VbM by humanetech@mastodon.social
 (DIR) More posts by humanetech@mastodon.social
 (DIR) Post #AIdAjHQ1s3D8irrqdM by humanetech@mastodon.social
       2022-04-17T08:58:21Z
       
       0 likes, 0 repeats
       
       Interesting Hacker News discussion on a "Slow #Social" application."Slow Social, a #SocialNetwork built for #friends, not influencers"https://news.ycombinator.com/item?id=31054029Noteworthy the comment of the creator of the platform:"It seems like #Mastodon and other federated networks definitely have user wellbeing and autonomy as a goal. However, for whatever reason, most of the more popular federated networks seem to aspire to be "X, but federated" instead of having drastically different user experiences."
       
 (DIR) Post #AIdAjHxhqq9KPJck7s by eylul@social.eyluldogruel.com
       2022-04-19T11:47:39Z
       
       0 likes, 0 repeats
       
       @humanetechdoes an iteration need to be innovation and a radically different thing? twitter and facebook didn't just appear in a vacuum. they looked at myspace, livejournal, that in return self hosted forums usenet etc...I use mastodon because it is better than twitter (that I can follow a blog, a pixelfed and a peertube account is a huge bonus, so is bigger text limits. (if anything twitter became more mastodon like in last year)starting as something people recognize isn't a bad thing.
       
 (DIR) Post #AIdAjIeFIgB2XFWiQq by humanetech@mastodon.social
       2022-04-19T12:38:23Z
       
       0 likes, 0 repeats
       
       @eylul no certainly doesn't have to be, and perfectly fine if it is not.I agree with you and enjoy the fedi a lot. But besides that there are opportunities now that may give traditional social media the finger. Unique dynamics where an open interoperable ecosystem grows into a broader and tight-knit "social fabric". If you come to think of it most online applications, not just what we think of when hearing 'social media', have a whole bunch of social aspects to them that may lend to federation
       
 (DIR) Post #AIdAjJE39YoiKIHJEu by eylul@social.eyluldogruel.com
       2022-04-19T13:27:59Z
       
       0 likes, 0 repeats
       
       @humanetech I definitely see what you are saying here. :) but those opportunities are also being developed. What I was trying to say is that innovations can happen in smaller bits (as it does) in mastodon and rest of fediverse. It doesn't need to be a big shiny thing that shows up at once. Also another question, is it that most popular alternatives do not choose to innovate, or that people actually prefer to use what they understand?(1/3)
       
 (DIR) Post #AIdAjJtAgfi6NpW9Kq by TerryHancock@mastodon.art
       2022-04-19T14:25:41Z
       
       0 likes, 0 repeats
       
       @eylul @humanetech Another fair point is that startup businesses with commercial models have done a pretty good exploration of the possibilities. If you try to think "what socially-connected software do I need?" what do you come up with that isn't already somehow filled with a centralized commercial tool?So, it may be hard NOT to be "X but federated" -- it's just a way of describing it to new users.I mean, Mastodon is SIMILAR to Twitter, but certainly not identical.
       
 (DIR) Post #AIdAjKV6Pe3GHTGRSS by eylul@social.eyluldogruel.com
       2022-04-19T14:29:13Z
       
       0 likes, 0 repeats
       
       @TerryHancock @humanetech this is essentially what I was pointing out! yes! :)
       
 (DIR) Post #AIdAjL38N7H1z1BcVE by humanetech@mastodon.social
       2022-04-20T05:27:34Z
       
       0 likes, 0 repeats
       
       @TerryHancock @eylul It is VERY hard to think further, agreed. In my advocacy I usually give suggestions of what's possible, in hopes the domain experts be inspired.For Fediverse it is not just the federation, but the Linked Data that can bring most innovation. Might involve letting go of the 'app silo' thinking.Highly inspiring is MercuryOS https://uxdesign.cc/introducing-mercury-os-f4de45a04289"Why do we still have the cluttered desktop paradigm?"Such paradigm will never come about the way fedi evolves now.
       
 (DIR) Post #AIdAjLE7iG42X6KOdE by eylul@social.eyluldogruel.com
       2022-04-19T13:33:12Z
       
       0 likes, 0 repeats
       
       @humanetech one side of the problem is the difficulty of adoption without huge marketing (something to convince, the audience that they need this new shiny thing) - which is why I always talk about the importance of "volunteering influencers" so to speak, or ambassadors or advocates, whatever you wish to call them. this issue is something those of us in FLOSS have been defining and trying to fix and verbalize for a while now. (2/3)
       
 (DIR) Post #AIdAjLkjl09UAFaRSy by RyunoKi@layer8.space
       2022-04-20T05:34:29Z
       
       0 likes, 0 repeats
       
       @humanetechHold on a sec!Linked data?As in SoLiD (by Sir Tim Berners-Lee)?@TerryHancock @eylul
       
 (DIR) Post #AIdAjN6Ojx4aLijFrs by eylul@social.eyluldogruel.com
       2022-04-19T13:37:34Z
       
       0 likes, 0 repeats
       
       @humanetech Also, on commercial domain there is a need for venture capital to "start" so you launch a product etc. when on FLOSS there is no such pressure. There is also the question that while new ideas are good, finding new ideas focused around actual use cases (and designing around them) is hard. So as the products that responds and can justify the use case to users win they tend to be things that people need which often look like what already exists with small changes. (3/3). :)
       
 (DIR) Post #AIdAjNRfSs5RPggF1M by humanetech@mastodon.social
       2022-04-20T05:33:14Z
       
       0 likes, 1 repeats
       
       @TerryHancock @eylul Inspiring along the same lines is Bret Victor's satyrical talk in 2013 where he presents as if it were 1973 and what "The future of programming" might look like in 2013. Hint: We're far off from that vision.https://vimeo.com/71278954There's some real cool research going on and relating (hopefully) to fediverse as well, such as https://spritelyproject.org and https://dream.public.catBut forces beyond fedi (metaverse, bluesky, etc.) may come to thwart our dreams.
       
 (DIR) Post #AIdAjP9J86aYhK6bgG by eylul@social.eyluldogruel.com
       2022-04-19T13:38:43Z
       
       0 likes, 0 repeats
       
       @humanetech sorry for the wall of text and rambling, this is something I talk and think about a lot and I am glad you brought this as a discussion (and on some level I do agree with you that we need to aim for "better" than X product, than just a copy, I am just beginning to ask myself what that "better" looks like) *shuts up*
       
 (DIR) Post #AIdBAWJ0d8AaiCZagS by humanetech@mastodon.social
       2022-04-20T05:39:25Z
       
       0 likes, 0 repeats
       
       @RyunoKi @TerryHancock @eylul Yes, as in https://solidproject.org but differently. ActivityPub is also a Linked Data standard. But Solid is going in yet another direction, where a formal standards body (their core team and Inrupt, a commercial entity) are driving forces. Plus they move into a direction of, what I call, universal interoperability. The opposite of ad-hoc interop (where interop deteriorates). In both directions the complexity grows non-linearly.
       
 (DIR) Post #AIdCjCUyMdGlN4GmVU by humanetech@mastodon.social
       2022-04-20T05:44:35Z
       
       0 likes, 1 repeats
       
       @RyunoKi @TerryHancock @eylul This image demonstrates the 2 opposite directions..
       
 (DIR) Post #AIdD5hLIMrPfP1TFb6 by humanetech@mastodon.social
       2022-04-19T14:10:08Z
       
       0 likes, 0 repeats
       
       @eylul I like to think of this too :)"X but federated" may lead to X limiting the imagination. But I'm not so worried about app evolution and adoption.More pressing is that Fediverse is an ecosystem based on open standards that no one cares for, except on individual app level. Broad collab in our grassroots culture is very hard to sustain. Innovation complexity grows over time as app protocol flavours abound.The potential is in the protocol, community and collaborative evolution.
       
 (DIR) Post #AIdD5hruPbV72AjIQq by TerryHancock@mastodon.art
       2022-04-19T14:27:30Z
       
       0 likes, 0 repeats
       
       @humanetech @eylul I think if you want something genuinely new, you're going to have to look at things that CAN'T be done without federation -- where there is no sensible central silo application.But I'm hard pressed to think of examples.
       
 (DIR) Post #AIdD5iPwN4isjieTTc by eylul@social.eyluldogruel.com
       2022-04-19T14:36:01Z
       
       0 likes, 0 repeats
       
       @TerryHancock @humanetech community-centric communications (same reason why we have slack and company emails in this century), user based moderation that is sustainable yet more accurate (and error correcting capability) than a central AI or anything. what @humanetech also pointed out which is inter-operation of different tools. You can install/join the tool you need yet listen to others and that is powerful. :)
       
 (DIR) Post #AIdD5j0oA0DIa3tuwS by humanetech@mastodon.social
       2022-04-20T05:53:26Z
       
       0 likes, 1 repeats
       
       @eylul @TerryHancock I see P2P + federation (aka hybrid decentralization) more as a quality of the ecosystem and in large part something that people do not care about.It's the emergent heterogenous "social fabric" that becomes very interesting once anyone can contribute and deeply integrate new parts to this fabric. And here the interoperability mechanism and the Linked Data aspects of the protocols are the biggest enablers of that.Community I feel is key, and modeled more as in real life.
       
 (DIR) Post #AIdD5jdnp1PCX093iq by Seirdy@pleroma.envs.net
       2022-04-20T06:00:58.415435Z
       
       0 likes, 0 repeats
       
       @humanetech @eylul @TerryHancock I'd probably place the IndieWeb somewhere similar to Solid on that chart.(Psst...alt-text sometimes gets cut off after a certain character count. You can put a longer transcript in a reply and use shorter alt text that ends with "see reply for full transcript". It's so great to see people caring about the issue though!).
       
 (DIR) Post #AIdDRetk8WeOBz3TkG by humanetech@mastodon.social
       2022-04-20T06:00:39Z
       
       0 likes, 1 repeats
       
       @TerryHancock @eylul In that talk Bret Victor states it well when it comes to adopting new ideas:"Technology changes quickly, people's minds change slowly. It is easy to adopt new technology, but it can be hard to adopt new ways of thinking"
       
 (DIR) Post #AIdDeOVyMR2Z920eNE by humanetech@mastodon.social
       2022-04-20T06:04:39Z
       
       1 likes, 0 repeats
       
       @RyunoKi @TerryHancock @eylul Image #captions (1/2)(for when they are truncated. Thanks @Seirdy )"Diagram titled "Interoperability in practice" shows two horizontal opposite arrows. One on the left in the direction of increasing Ad-hoc interoperability where everyone creates their own extensions to the protocols on-the-fly. And on the right towards Universal interoperability where semantics are fully machine readable."
       
 (DIR) Post #AIdDeP8c2lwt4s5VbM by humanetech@mastodon.social
       2022-04-20T06:05:27Z
       
       1 likes, 0 repeats
       
       @RyunoKi @TerryHancock @eylul @Seirdy Image #captions (2/2)"Center-left is the ActivityPub standard, but the Fediverse itself is moving more towards Ad-hoc interop.Center-right is the Solid project by Tim Berners-Lee moving towards Universal interoperability and more and more formal standards that define everything.In both directions the complexity of establishing interoperability grows non-linearly over time.There should be a sweet spot somewhere between ad-hoc and universal interop"
       
 (DIR) Post #AIdMPsKM8RK0qnjARc by humanetech@mastodon.social
       2022-04-20T06:37:37Z
       
       0 likes, 0 repeats
       
       @bob @RyunoKi @TerryHancock @eylul @Seirdy Sure, and not disputing any of this, nor validity of the approach.But we are in a situation where the popular X is a 'Microblogging' domain, and becomes so dominant in our thinking, that we create X* and X** and when we have an app in a different domain Y and want to federate we start to think "should I add X*** features?"This often makes no sense, so we may say "Nah, not interested in federation".But A, B, C, X, Y, Z are all valid on their own.
       
 (DIR) Post #AIdMPsqyBBPSTwzDHM by humanetech@mastodon.social
       2022-04-20T06:46:56Z
       
       0 likes, 1 repeats
       
       @bob @RyunoKi @TerryHancock @eylul @Seirdy I've had this discussion with @forgefriends Very interesting. A new domain for fedi: Code Forges. And #ForgeFed paved the road already.But what does it mean to add federation? To add social networking? Is the Code Forge important? No, it isn't. Forges are just tools that help programmers. The core domain is Programming or Software Development.Now we see that programming is uniquely social in nature. We translate human interaction to code.
       
 (DIR) Post #AIdMPxN1NxQOUW0rnE by humanetech@mastodon.social
       2022-04-20T06:50:09Z
       
       0 likes, 0 repeats
       
       @bob @RyunoKi @TerryHancock @eylul @Seirdy @forgefriends This seems obvious, but the complexity of our technologies means we often get mired in it, and overlook the social aspects, develop something the client doesn't want.When thinking #SocialNetworkingReimagined in the domain of free software development our interop extensions may open up an ecosystem that is as broad as any tool provider that is now building on top of Github. But ours is an open and democratic ecosystem.
       
 (DIR) Post #AIdMPze6vOHbY61Xhw by humanetech@mastodon.social
       2022-04-20T06:51:24Z
       
       0 likes, 0 repeats
       
       @bob @RyunoKi @TerryHancock @eylul @Seirdy @forgefriends But we might easily get stuck in a box-thinking of Code Forges alone and slice things differently, basically modeling in one domain what we find in terms of features in Github, Gitlab, Gitea, etc.
       
 (DIR) Post #AIdMQ1inCxDTzCEJHc by humanetech@mastodon.social
       2022-04-20T06:53:21Z
       
       0 likes, 0 repeats
       
       @bob @RyunoKi @TerryHancock @eylul @Seirdy @forgefriends For each of those specific domains (and there are numerous ones) it helps if a diverse community of domain experts and technologists bundle force and collaboratively further the ecosystem. But that happens to be a very hard nut to crack in our grassroots cultural environment.
       
 (DIR) Post #AIdMWYF1DH2ma8PkbQ by humanetech@mastodon.social
       2022-04-20T06:56:43Z
       
       0 likes, 1 repeats
       
       @bob @RyunoKi @TerryHancock @eylul @Seirdy Yes, again 100% with you. When saying #SocialNetworkingReimagined I am talking about the potential of the #Fediverse and the protocols / standards that are used are the implementation details to that.I see ActivityStreams as some core library providing some generic social primitives. It is the extensibility mechanism with linked data vocabularies where you can model different domains, and also use many vocabs that are well standardized already.
       
 (DIR) Post #AIdWxbYG0qFiD0x4j2 by eylul@social.eyluldogruel.com
       2022-04-20T07:55:00Z
       
       0 likes, 0 repeats
       
       @humanetechor the developer can focus on getting the audience first by replicating similar tools because that is what the users need (again those tools ended up on those previously used software for a reason)with sometimes radical changes. but interoperatability without a reason isn't a radical change unless there is a benefit to the user. what benefit to the user we lost so far because players actions? which actions? :)@bob @RyunoKi @TerryHancock @Seirdy @forgefriends
       
 (DIR) Post #AIdWxc7ht2bnyxXNyq by humanetech@mastodon.social
       2022-04-20T08:03:22Z
       
       0 likes, 0 repeats
       
       @eylul @bob @RyunoKi @TerryHancock @Seirdy @forgefriends Rather than trying to formulate an answer I'd like to say that you are asking the right kind of questions.And imho the fedi as a whole will benefit if - besides continuing on the valid path we already are on - more people asked these questions more often and collectively worked on finding answers + exploring solutions.In a way this does happen, but so so fragmented, with lack of 'substrate formation', and a risk that things'll stall.
       
 (DIR) Post #AIdWxcpfFblqBI6UUq by eylul@social.eyluldogruel.com
       2022-04-20T08:51:40Z
       
       0 likes, 0 repeats
       
       @humanetechhere is a question though. is fragmented a bad thing. various actors or group of actors more as weak ties bringing domains, use cases and experiences and doing their own innovation. isn't ad-hoc actually more permissive to idea generation? Rather than a homogenous group of people trying to figure out top down a sleek idea everyone will adhere to?difficult to fund, sure but that is a different problem *ducks the incoming object*@bob @RyunoKi @TerryHancock @Seirdy @forgefriends
       
 (DIR) Post #AIdWxdXyarDSOipsZ6 by humanetech@mastodon.social
       2022-04-20T08:55:57Z
       
       0 likes, 1 repeats
       
       @eylul @bob @RyunoKi @TerryHancock @Seirdy @forgefriends I'd say that one hand an amount of ad-hoc interoperability is required. One needs to be able to experiment, do new unique things. But at a certain moment things need to be brought back to a stage where some other can build upon. Make it part of the substrate. This isn't happening enough and hence with N apps we get N -1 separate integrations to worry about.Fragmentation isn't bad as long as you consider you are part of something bigger
       
 (DIR) Post #AIdWxfe4n9HeuDhmLo by humanetech@mastodon.social
       2022-04-20T08:57:37Z
       
       0 likes, 0 repeats
       
       @eylul @bob @RyunoKi @TerryHancock @Seirdy @forgefriends A "homogenous group of people" isn't really needed, and stuff like "official standards body" even undesirable. But more community interaction than 'everyone go it alone' would be beneficial I think.
       
 (DIR) Post #AIdWxiG54os90fV9rk by humanetech@mastodon.social
       2022-04-20T08:59:08Z
       
       0 likes, 0 repeats
       
       @eylul @bob @RyunoKi @TerryHancock @Seirdy @forgefriends I witness that in general our society puts pressure on us (distractions, time constraints, occupied with other things, need for an income, etc) makes it very hard to organize in such ways. Here we have to find solutions to get to a minimum level of collab.
       
 (DIR) Post #AIdX673Km9FGE0XppY by TerryHancock@mastodon.art
       2022-04-20T09:00:40Z
       
       0 likes, 0 repeats
       
       @humanetech @eylul @bob @RyunoKi @Seirdy @forgefriends I THINK what I hear you talking about is a generalization of the "embrace, extend, extinguish" problem we used to hear a lot of complaints about with Microsoft or other corporates co-opting standards.One problem I see is that if all this stuff gets merged into the "substrate", then it grows into an unwieldy mass, and supporting it is like trying to write a new web browser these days, where it's an overwhelming task.
       
 (DIR) Post #AIdX67fGV7aQ7eI7xA by humanetech@mastodon.social
       2022-04-20T09:05:27Z
       
       0 likes, 0 repeats
       
       @TerryHancock @eylul @bob @RyunoKi @Seirdy @forgefriends Depends what the goals are. I quoted:“Any decentralized [ecosystem] requires a centralized substrate, and the more decentralized the approach is the more important it is that you can count on the underlying system”From: https://www.thediff.co/p/the-promise-and-paradox-of-decentralizationHere meaning 'decentralized but interoperating'. Things grow unwieldy if we *don't* organize our substrate but still create our own flavour of fediverse apps all the time. Becomes unmanageable
       
 (DIR) Post #AIdX68PhiSjWRg1DKy by humanetech@mastodon.social
       2022-04-20T09:07:46Z
       
       0 likes, 0 repeats
       
       @TerryHancock @eylul @bob @RyunoKi @Seirdy @forgefriends Note that "the substrate" is not a 'block' of highly formalized specs that you find at some central website and then implement. It is the totality of organization structures, people processes, etc. that make interoperability at larger scales still manageable and sustainable in the ecosystem / technology landscape.
       
 (DIR) Post #AIdX695BEFuUWJQKzA by humanetech@mastodon.social
       2022-04-20T09:09:24Z
       
       0 likes, 1 repeats
       
       @TerryHancock @eylul @bob @RyunoKi @Seirdy @forgefriends Right now Fediverse hardly has any substrate. There's a spec you can start with, and then for broad interop you have to hop from codebase to codebase, bother devs that may have moved on or are busy, and reverse-engineer the protocol extensions and best-practices from that.
       
 (DIR) Post #AIdX6B3TtXjudce0cC by humanetech@mastodon.social
       2022-04-20T09:38:39Z
       
       0 likes, 0 repeats
       
       @TerryHancock @eylul @bob @RyunoKi @Seirdy @forgefriendsFunny but sad thing is that in this fleety medium all we discussed in this thread will be history tomorrow. Whether anything comes from it is fully up to  serendipity. Luck-based.There's no way, no where to follow-up. Leave an action item in a backlog, make a note, update some docs. There's no group / community to inform with any insight. There's no substrate to speak of. Best would be to post at #SocialHub which is not really active.
       
 (DIR) Post #AIdX6CyEm0jWZwCqie by humanetech@mastodon.social
       2022-04-20T09:42:28Z
       
       0 likes, 0 repeats
       
       @TerryHancock @eylul @bob @RyunoKi @Seirdy @forgefriends This is where the grassroots fedi differs a lot from corporate world and where it can be a disadvantage. A big corporate may push things into adoption. Ensure the attention and dedication exists to see things maturing further.I said before that a risk is that the fediverse might just manage to become interesting.. for exploitation, before it manages to become strong to stand its ground in the subsequent corporate onslaught. Or stall.
       
 (DIR) Post #AIdbelQzXmch82OAe8 by eylul@social.eyluldogruel.com
       2022-04-20T09:55:49Z
       
       0 likes, 0 repeats
       
       @humanetech now that is a thing I agree 100% and is a whole different rabbit hole. xD @bob @RyunoKi @TerryHancock @Seirdy @forgefriends
       
 (DIR) Post #AIdbem7B0wMpEs7rOq by humanetech@mastodon.social
       2022-04-20T10:02:18Z
       
       0 likes, 0 repeats
       
       @eylul @bob @RyunoKi @TerryHancock @Seirdy @forgefriends Well, its here I found some interesting stuff to ponder. So the nature of fedi/FOSS grassroots culture, the ad-hoc interop and the observed implications for the ecosystem led me to see it hampers technology adoption.. a broken technology innovation lifecycle.And to fix it we need to look at the unique dynamics, motivations and expectations that are quite different than in traditional (corporate) settings.https://discuss.coding.social/t/challenge-fixing-the-fediverse-technology-adoption-lifecycle/38#alternative-adoption-models-4
       
 (DIR) Post #AIdbemjohHH9AiCicy by humanetech@mastodon.social
       2022-04-20T10:05:38Z
       
       0 likes, 1 repeats
       
       @eylul @bob @RyunoKi @TerryHancock @Seirdy @forgefriends For instance in FOSS projects often start from a completely different angle e.g. to scratch an itch of the developer to try out some new technology, to experience 'The Joy of Coding'. Intrinsic motivations. Not a drive for productivity and ROI. And in our tools we can help retain these motivations and in ways that also set expectations properly (so people don't get disappointed, or burned out, etc.)
       
 (DIR) Post #AIdc1hAVZQMiYhc5NQ by eylul@social.eyluldogruel.com
       2022-04-20T10:03:28Z
       
       0 likes, 0 repeats
       
       @humanetechthat risk is there independent of centralization though. Just look at HTML. (if anything it gives a nice centralized target for the commercial entity)the solution there is if anything prioritization of non-profit technologies, good anti-trust laws that are actually used. We cannot work around the issues of capitalism with out-teching commercial products alone, not sustainably. @TerryHancock @bob @RyunoKi @Seirdy @forgefriends
       
 (DIR) Post #AIdc1hm5JiQIRFC5wm by humanetech@mastodon.social
       2022-04-20T10:12:36Z
       
       0 likes, 0 repeats
       
       @eylul @TerryHancock @bob @RyunoKi @Seirdy @forgefriends Agreed. There are many aspects and we face kind of a wicked problem altogether. Role of anti-trust and other regulation is very important, but also very slow to catch up (maybe even eternally outpace by tech racing ahead)Decentralization and open standards create a more level playing field where ideally everyone small or large can be part of. I we manage to uphold decent interoperability levels we also get network effects in our favor.
       
 (DIR) Post #AIdc1iQqs926TgGeUS by RyunoKi@layer8.space
       2022-04-20T10:40:22Z
       
       0 likes, 0 repeats
       
       @humanetechHm, the small parties tend to get acquired by the big ones, though :-(@eylul @TerryHancock @bob @Seirdy @forgefriends
       
 (DIR) Post #AIdc1jcETJjM9GbFrs by humanetech@mastodon.social
       2022-04-20T10:32:46Z
       
       0 likes, 0 repeats
       
       @eylul @TerryHancock @bob @RyunoKi @Seirdy @forgefriends If y'all are interested to discuss more on these subjects I gladly invite you to the #SocialCoding Foundations chatroom on Matrix, which I created for the purpose.https://matrix.to/#/#socialcoding-foundations:matrix.org
       
 (DIR) Post #AIdc5ubTYs1DNvEo9A by RyunoKi@layer8.space
       2022-04-20T10:41:09Z
       
       0 likes, 0 repeats
       
       @humanetechSwitching from microblogging to chat, huh?I'm in!@eylul @TerryHancock @bob @Seirdy @forgefriends
       
 (DIR) Post #AIdigYPncSt10adESu by liwott@nerdica.net
       2022-04-20T11:35:06Z
       
       0 likes, 0 repeats
       
       @humanetechFunny but sad thing is that in this fleety medium all we discussed in this thread will be history tomorrow.The world needs a delightful list of fedi meta-threads 😁
       
 (DIR) Post #AIdigZ2nHU4uxWsNFI by RyunoKi@layer8.space
       2022-04-20T11:55:00Z
       
       0 likes, 0 repeats
       
       @liwottNo worries. We have chat in parallel.Does that make it a stream? 🤔@humanetech
       
 (DIR) Post #AIdvPBXWiOcYXjIJKy by TerryHancock@mastodon.art
       2022-04-20T13:55:47Z
       
       0 likes, 0 repeats
       
       @humanetechIs there not a "thread roll" application for Fedi?Fedilab has a decent thread view. Surely turning a thread into a markdown document is possible? That's not a "hard problem", just a "small sharp tool".@eylul @bob @RyunoKi @Seirdy @forgefriends
       
 (DIR) Post #AIdvPCEQ8uvqglMZCC by RyunoKi@layer8.space
       2022-04-20T14:17:32Z
       
       0 likes, 0 repeats
       
       @TerryHancockI'd be happy to have an indicator for a toot being a reply vs. top-level.Haven't checked Fedilab @apps for an open issue on this, though.With respect to the topic: do you imagine an unroll bot like on birdsite?@humanetech @eylul @bob @Seirdy @forgefriends
       
 (DIR) Post #AIdwOGvZIX1MuwfU4u by TerryHancock@mastodon.art
       2022-04-20T14:28:33Z
       
       0 likes, 0 repeats
       
       @RyunoKiWhat I imagined was an a simple app to convert a thread to a permanent document, given a link.But the "thread roll" bot on Twitter is what I thought of when the issue was raised.I thought maybe something already existed, but my search lead to Twitter thread rolls discussing Mastodon or Fediverse (and not saying nice things, either  :artcathmm: ).@apps @humanetech @eylul @bob @Seirdy @forgefriends
       
 (DIR) Post #AIdx9bRErBga8o3Cme by TerryHancock@mastodon.art
       2022-04-20T14:37:05Z
       
       0 likes, 1 repeats
       
       @RyunoKiOr I guess, a reader like FediLab could have a "save thread" or "export thread" button in the thread view.(I don't see one right now. Assuming I'm not just being blind to it).@apps @humanetech @eylul @bob @Seirdy @forgefriends
       
 (DIR) Post #AIdyDM1gCWitSARrTU by RyunoKi@layer8.space
       2022-04-20T14:49:01Z
       
       0 likes, 0 repeats
       
       @TerryHancockI think, an app would be more appropriate than a web service. The latter would need to authorise itself on behalf of a person before it can work.At least if non-public messages are not to be missed.How would you imagine the result should be handled?Like, automatically saved?Offered as share?Shown as view?Something else?@apps @humanetech @eylul @bob @Seirdy @forgefriends
       
 (DIR) Post #AIe0wPg9fNj5iozgBs by humanetech@mastodon.social
       2022-04-20T15:19:14Z
       
       0 likes, 0 repeats
       
       @RyunoKi@TerryHancock I used @Chartodon before from the top of the thread. By @ColinTheMathmo
       
 (DIR) Post #AIe1BwU3yrFfoBk7Xc by TerryHancock@mastodon.art
       2022-04-20T15:22:21Z
       
       0 likes, 1 repeats
       
       @RyunoKiWell let's tell a "user story":"I'm a small-time dev with a git forge repo where I dabble with projects."I see a thread on Fedi about something I think would be cool to write. It's a big thread, with a lot of ideas in it that I think would be useful."It also acquaints me with people who might be interested in helping somehow."How can I quickly turn that into a useful stub project on my Git forge repo?"@apps @humanetech @eylul @bob @Seirdy @forgefriends
       
 (DIR) Post #AIe1fx0eagV67tIPZY by TerryHancock@mastodon.art
       2022-04-20T15:27:46Z
       
       0 likes, 0 repeats
       
       @RyunoKiOr..."I'm a PhD humanities student, and I got into a complex discussion thread on Mastodon."I'd like to reference this discussion in my dissertation, with something better than 'private communications...'"How can I turn the thread into a readable format for future reference, with appropriate credit for (permission from?) participants in the discussion?"Seems like thread-to-markdown would help with both of those cases?@apps @humanetech @eylul @bob @Seirdy @forgefriends
       
 (DIR) Post #AIe2CnFbtWDhPrx6YK by humanetech@mastodon.social
       2022-04-20T08:22:39Z
       
       0 likes, 0 repeats
       
       @bob @TerryHancock Agree on the scepticism re:linked data. I feel the open world model, universally expressed machine-readable semantics is big part of the reason why Semantic Web failed. It's trying to model something impossible to model: the complexity of everyday reality.That's why I mentioned a 'Sweet spot'. Something inbetween anyone inventing their own format, and a 'universal truth'. Closed-world domain-specific (bounded context) extensions, where there's consent among implementers.
       
 (DIR) Post #AIe2CnrtbAqRKbrgEC by jens@social.finkhaeuser.de
       2022-04-20T08:51:44Z
       
       0 likes, 0 repeats
       
       @humanetech @bob @TerryHancock Interestingly, this echoes something in the E2EE for messaging discussion. Some people lament that interoperability would break E2EE or vice versa, which is not necessary. Email shows that it's not.The trick is to have a fairly simple protocol that models the end-to-end messaging parts, which is inherently capable of transporting structured data. Email does this with MIME, ActivityPub with JSON.Then, interoperable standards are a question of defining ...
       
 (DIR) Post #AIe2CoautmrDaEvdOy by apparentlymart@mastodon.online
       2022-04-20T14:41:05Z
       
       0 likes, 1 repeats
       
       @jens @humanetech @bob @TerryHancock the earliest incarnations of what became "activity streams" (which in turn became the foundation of ActivityPub) were layered on email+MIME and Atom respectively, in both cases with the intent of "progressive enhancement" of an existing transport/channel, giving a potential fallback to a generic presentation for clients that didn't support the extra structured data.
       
 (DIR) Post #AIe2Cpz3jVlNtPEQfg by jens@social.finkhaeuser.de
       2022-04-20T08:53:26Z
       
       0 likes, 0 repeats
       
       @humanetech @bob @TerryHancock ... data structures to send/receive.This does raise the spectre of a complexity problem, but unlike a single complex standard, a simple transport with simple orthogonal "data structure exchange protocols" tend to be easier to keep track of. The complexity lies in their interaction, which in the worst/best case can be strictly limited.
       
 (DIR) Post #AIe2CqMoJClJ54LOgy by apparentlymart@mastodon.online
       2022-04-20T14:45:46Z
       
       0 likes, 0 repeats
       
       @jens @humanetech @bob @TerryHancock unfortunately that principle was basically lost in the final JSON incarnation, since the focus shifted toward providing a vocabulary for point-to-point integration between silos via partnerships, rather than on defining general functionality for the general internet. One of my big regrets about it and a big part of why I burned out of there at the time. 😖
       
 (DIR) Post #AIe2CsPMifzhPZYSx6 by apparentlymart@mastodon.online
       2022-04-20T14:46:43Z
       
       1 likes, 0 repeats
       
       @jens @humanetech @bob @TerryHancock Hopefully there's new momentum to get to something like this within the Fediverse now though; perhaps Mastodon's interpretation of ActivityPub is now the "fallback presentation" for the next generation, as email/MIME and Atom were for the early attempts.
       
 (DIR) Post #AIe3RRRHN96q49xWU4 by liwott@nerdica.net
       2022-04-20T15:44:01Z
       
       0 likes, 0 repeats
       
       In the PhD dissertation case, maybe a thread-to-latex tool would be even better?But of course the implementing the output format is probably the most trivial part !@TerryHancock@humanetech @Seirdy @eylul @apps @bob @forgefriends @RyunoKi
       
 (DIR) Post #AIe3RS9wh4q2IgrC6a by RyunoKi@layer8.space
       2022-04-20T15:47:33Z
       
       0 likes, 0 repeats
       
       @liwottMight be interesting down the road to you, @HeikeW @humanetech @Seirdy @eylul @apps @bob @TerryHancock @forgefriends
       
 (DIR) Post #AIeD9AvL2HJvArwoLY by TerryHancock@mastodon.art
       2022-04-20T06:54:56Z
       
       0 likes, 0 repeats
       
       @humanetech Interesting talk. Of course, many of these "fresh new ideas" were tried and largely found wanting, which is why we're not doing them so much.There are also certain fundamental problems that have been left unsolved.Like computing diffs and patches on data that it is not textual -- or more precisely, organized into ordered linear arrays of records.Because of this limitation, collaborative ART projects are still stuck with lock-based collaboration, as in the 1970s.@eylul
       
 (DIR) Post #AIeD9Be0MD37POqTy4 by TerryHancock@mastodon.art
       2022-04-20T06:58:10Z
       
       0 likes, 0 repeats
       
       @humanetechAnd THAT limitation has contributed to the continued addiction to text-based formats whenever collaboration is desirable.But text-based formats absolutely suck for things like representing 3D models or animation or even drawings.We can serialize things into text, like SVG -- but typically these representations aren't UNIQUE. Nor does applying conventional version control work well on them -- the preservation of the text structure doesn't reliably preserve what matters.@eylul
       
 (DIR) Post #AIeD9CLximD9bjPaU4 by TerryHancock@mastodon.art
       2022-04-20T07:01:12Z
       
       0 likes, 0 repeats
       
       @humanetech What I mean is that I can show you drawing A and drawing B, which look nearly identical to a human, but have completely different SVG.The computer can't merge them sensibly.Blender files are even worse, even though, conceptually it shouldn't really be that hard. A human can merge two Blend files originating from a single original (i.e. what diff3 does with text) fairly easily, as long as they don't genuinely conflict.@eylul
       
 (DIR) Post #AIeD9D8WoD3k2M8NBQ by humanetech@mastodon.social
       2022-04-20T07:02:17Z
       
       0 likes, 0 repeats
       
       @TerryHancock @eylul Agreed. The question is where the limitations come from. There are many factors at play, most likely. The complexity of technologies being one of them.Bret Victor has another great video that relates to what you say about text formats, and I think it will greatly appeal to both of you. Here it is:https://vimeo.com/66085662Once again it shows how out-of-the-box thinking may lead to considering entire new paths to explore.
       
 (DIR) Post #AIeD9E51IjqaxlV5M0 by jens@social.finkhaeuser.de
       2022-04-20T07:10:48Z
       
       0 likes, 0 repeats
       
       @humanetech @TerryHancock @eylul Well, this is what CRDTs are good for, diffs/patches for structured data. That's essentially what @alcinnz and I are working on, a framework that makes this more useful.Since there is multi authoring support, and synchronization will be added soon enough, this will lend itself to local first, distributed apps.
       
 (DIR) Post #AIeD9EXjZyoeOow16u by TerryHancock@mastodon.art
       2022-04-20T07:02:53Z
       
       0 likes, 0 repeats
       
       @humanetech However, because we have a human doing a machine's job, we waste a lot of labor. In fact, simply throwing away drawing B and having a human make the desired changes to drawing A is often the less laborious option.As things stand.@eylul
       
 (DIR) Post #AIeD9EraOAhBOODs3M by TerryHancock@mastodon.art
       2022-04-20T07:18:50Z
       
       0 likes, 0 repeats
       
       @jens @humanetech @eylul @alcinnz That's interesting. As I mentioned (in a catty-corner thread? branchlet above?) I found a paper about a diff3 for DAGs, which seems like it might fit VCS for object-database formats. (?)Which would cover Blender files, for example -- arrays of linked objects. They CAN become cyclic, but Blender does cycle-breaking, IIRC.This is probably dragging off topic, but I'm deeply interested in finding an open-source solution for this for multimedia projects.
       
 (DIR) Post #AIeD9FY7q0itWK7qMK by TerryHancock@mastodon.art
       2022-04-20T07:26:50Z
       
       0 likes, 0 repeats
       
       @jens @humanetech @eylul @alcinnz Since I've mention this paper a few times, it is:"A DAG Comparison Algorithm and Its Application to Temporal Data Warehousing"Johann Eder, Karl Wiggisser J.F. Roddick et al. (Eds.): ER Workshops 2006, LNCS 4231, pp. 217–226, 2006.c Springer-Verlag Berlin Heidelberg 2006I have a PDF I got from:https://www.researchgate.net/publication/221268571_A_DAG_Comparison_Algorithm_and_Its_Application_to_Temporal_Data_WarehousingI'm not sure if anyone can download it from there or not.I also don't know if this is THE solution, but it looked promising to me.
       
 (DIR) Post #AIeD9GQiZ2OMFdfRS4 by TerryHancock@mastodon.art
       2022-04-20T08:07:59Z
       
       0 likes, 1 repeats
       
       @jens @humanetech @eylul @alcinnz Just adding one more bit, circling back to the original discussion:This visual merging problem is an enabler. Not a goal in itself.If we wanted to have a Blender 3D model VCS/DAMS for open-source film-making, we'd need this CS problem to be solved.Most programming concentrates on applying existing CS solutions to end-user problems.Text is a preferred format for programming, because so many text problems are already solved.
       
 (DIR) Post #AIeD9GetiJjaxcIlYO by TerryHancock@mastodon.art
       2022-04-20T07:09:27Z
       
       0 likes, 0 repeats
       
       @humanetech So, at one point, I found a paper that explained how to do diff3 on a "directed acyclic graph" with considerable success.(Interestingly, diff3 on text can't be formally proved to work universally -- because in fact, it doesn't. It just works a useful amount of the time. Similarly, the graph technique probably isn't foolproof, but just useful).But now what?I'm an artist who dabbles in Python. I'm not a computer scientist. Solvable. But not by me.@eylul
       
 (DIR) Post #AIeD9IXshNJIoR2BtY by TerryHancock@mastodon.art
       2022-04-20T08:24:03Z
       
       0 likes, 1 repeats
       
       @jens @humanetech @eylul @alcinnz So software evolution is highly path-dependent. Once programmers figure out how to do something, they focus on applying that to problems.They don't spend time on all the untrod paths. That puts you back in the position of solving hard problems.So our "source" files get called "blobs" -- incomprehensible by the the software, by definition. Git stores them in LFS. And that's it.Visual programming metaphors would lead to programmers having the same problem.