Open format discussion (Zot vs ActivityPub) #2

New Issue

ajeremias · 2020-01-17T18:48:49Z

ajeremias commented

2020-01-17 18:48:49 +00:00

It seems activityPub is having some flaws.. which don't provide to the final user the privacy and control.. that ZOT could give... but this is a long discussion happening in the self-hosting fediverse community..

Here's a cool introduction to Zot..
https://zotlabs.org/help/about/about#Technical_Introduction

Anyway hubzilla might be the software to change to be able to do this... it supports loads of stuff..

It seems activityPub is having some flaws.. which don't provide to the final user the privacy and control.. that ZOT could give... but this is a long discussion happening in the self-hosting fediverse community.. Here's a cool introduction to Zot.. https://zotlabs.org/help/about/about#Technical_Introduction Anyway hubzilla might be the software to change to be able to do this... it supports loads of stuff..

ajeremias closed this issue

2020-01-17 18:49:31 +00:00

servalan commented

2020-01-17 21:57:00 +00:00

I've used both protocols and as it turns out that actually existing Zot is not all that much better at keeping griefers out than ActivityPub. A lot depends not so much on the protocol as the way that the client app is written.

ajeremias referenced this issue

2020-01-17 22:05:33 +00:00

We need to look through all the ActivityPub codebases #1

ajeremias commented

2020-01-17 22:07:44 +00:00

Poster

I’ve used both protocols and as it turns out that actually existing Zot is not all that much better at keeping griefers out than ActivityPub. A lot depends not so much on the protocol as the way that the client app is written.

Can go deeper on this? I see many problems arising at least in mastodon, everytime a new feature is being implemented.. is the feature of migrating accounts compatible with other AP apps?

the next paragraphs are a answer to: #1 (comment)

FIXED: zot link: https://zotlabs.org/help/en/about/about

For the security/offline p2p minded we are planing to bridge to scuttlebutt useing the “pub code” have one of the core devs putting a proposal up for this here over the next weeks.

what is “pub code”?

Hubzilla is nice for geeks but not so nice for outreach to normal people. As we are doing a project to balance the power of geeks and empower normal people #activityPub is better for a #4opens project.

what hubzilla needs is some good ui person.. and i think a good ui/design person would be really helpful, for the project. Also I think the main features which i don’t believe AP is able to solve easily, and would be interesting for indymedia, like nomadic identity, clones and “ remote authentication is silent and invisible”

Also hubzilla supports AP, so it could federate to AP :)

> I’ve used both protocols and as it turns out that actually existing Zot is not all that much better at keeping griefers out than ActivityPub. A lot depends not so much on the protocol as the way that the client app is written. Can go deeper on this? I see many problems arising at least in mastodon, everytime a new feature is being implemented.. is the feature of migrating accounts compatible with other AP apps? the next paragraphs are a answer to: https://unite.openworlds.info/indymedia/indymedia-reboot/issues/1#issuecomment-84 FIXED: zot link: https://zotlabs.org/help/en/about/about > For the security/offline p2p minded we are planing to bridge to scuttlebutt useing the “pub code” have one of the core devs putting a proposal up for this here over the next weeks. what is “pub code”? > Hubzilla is nice for geeks but not so nice for outreach to normal people. As we are doing a project to balance the power of geeks and empower normal people #activityPub is better for a #4opens project. what hubzilla needs is some good ui person.. and i think a good ui/design person would be really helpful, for the project. Also I think the main features which i don’t believe AP is able to solve easily, and would be interesting for indymedia, like nomadic identity, clones and “ remote authentication is silent and invisible” Also hubzilla supports AP, so it could federate to AP :)

ajeremias reopened this issue

2020-01-17 22:07:59 +00:00

servalan commented

2020-01-17 22:19:00 +00:00

Mastodon's account migration I havn't tried. I've looked for some kind of explanation of how they do it, but failed. Compared to Zot it's a crude hack, but Mastodon does have cryptographic identity via the signing keys and so in principle could do something similar to Hubzilla or Zap.

OMN commented

2020-01-17 22:21:50 +00:00

OK guys the project we are doing here is based on the #4opens which is deisgined to over come lots of deep problems you might not have thought about https://unite.openworlds.info/Open-Media-Network/Open-Media-Network/wiki/4opens

This with outreach UI is the jugment on what gets used and what dose not, if you are brining ideas to the project please run them past the #4opens before pushing them hard here.

Zot fails - Open “industrial” standards – this is a little understood but core open, it’s what the open internet and WWW are built from. Here is an outline https://en.wikipedia.org/wiki/Open_standard

And hubzilla clearly fails good UI and is going to contine to fails this.

Outline on what we are building is here https://unite.openworlds.info/Open-Media-Network/Open-Media-Network/wiki its a simple project with almost all the pices in place and much of the code allready writern.

The inital social side of the project and workflow is alrready astablished in the indymedia project so we do not have to build much new here.

OBviusely the project will grown into something new from this starting point, lets first get to the starting point in the best way we can.

Ps. the can be meany codebases, we are planing to build 2 data is just data #geekproblem

OK guys the project we are doing here is based on the #4opens which is deisgined to over come lots of deep problems you might not have thought about https://unite.openworlds.info/Open-Media-Network/Open-Media-Network/wiki/4opens This with outreach UI is the jugment on what gets used and what dose not, if you are brining ideas to the project please run them past the #4opens before pushing them hard here. Zot fails - Open “industrial” standards – this is a little understood but core open, it’s what the open internet and WWW are built from. Here is an outline https://en.wikipedia.org/wiki/Open_standard And hubzilla clearly fails good UI and is going to contine to fails this. Outline on what we are building is here https://unite.openworlds.info/Open-Media-Network/Open-Media-Network/wiki its a simple project with almost all the pices in place and much of the code allready writern. The inital social side of the project and workflow is alrready astablished in the indymedia project so we do not have to build much new here. OBviusely the project will grown into something new from this starting point, lets first get to the starting point in the best way we can. Ps. the can be meany codebases, we are planing to build 2 data is just data #geekproblem

👍 1

ajeremias commented

2020-01-17 22:23:05 +00:00

Poster

aren't we getting a ui person for this project? we need that.. ?

😕 1

ajeremias closed this issue

2020-01-17 22:23:57 +00:00

servalan commented

2020-01-17 22:28:50 +00:00

The current popular fediverse servers have various problems. This can mean that sometimes it can be tricky to keep griefers out. But, like a lot of internet systems, there's enough configurability that they work well enough most of the time.

Personally I think the local and federated timelines aren't needed. They're not part of ActivityPub and are just legacy from StatusNet. Also the relaying of posts is afaik indiscriminate, meaning that adversaries can use your instance for their purposes. This isn't like Tor where there's multiple cryprographic layers and so the "in plain sight" nature of such relaying may create problems in future.

To get technical, Pleroma lacks enforcement of post digest checks and so replay attacks may be possible. I expect they will close that loophole before long, otherwise it will be badly exploited.

The current popular fediverse servers have various problems. This can mean that sometimes it can be tricky to keep griefers out. But, like a lot of internet systems, there's enough configurability that they work well enough most of the time. Personally I think the local and federated timelines aren't needed. They're not part of ActivityPub and are just legacy from StatusNet. Also the relaying of posts is afaik indiscriminate, meaning that adversaries can use your instance for their purposes. This isn't like Tor where there's multiple cryprographic layers and so the "in plain sight" nature of such relaying may create problems in future. To get technical, Pleroma lacks enforcement of post digest checks and so replay attacks may be possible. I expect they will close that loophole before long, otherwise it will be badly exploited.

hamishcampbell commented

2020-01-17 22:37:31 +00:00

Please focues on the project we are doing - its a reboot of early indymedia, do some background searching if you need to, but must worn you, the mainstrem history is not very good. You can still find all the email list archives on wayback masion if madly intrested ;)

ajeremias commented

2020-01-17 22:40:45 +00:00

Poster

Please focues on the project we are doing - its a reboot of early indymedia, do some background searching if you need to, but must worn you, the mainstrem history is not very good. You can still find all the email list archives on wayback masion if madly intrested ;)

I don't find this a very open call for participation.. we are all here to learn and to understand why such choices are made..

> Please focues on the project we are doing - its a reboot of early indymedia, do some background searching if you need to, but must worn you, the mainstrem history is not very good. You can still find all the email list archives on wayback masion if madly intrested ;) I don't find this a very open call for participation.. we are all here to learn and to understand why such choices are made..

hamishcampbell commented

2020-01-17 22:47:48 +00:00

@ajeremias have a look through the wiki pages, this project like indymedia it self is old and embeded in meany coultures which you are just steping into, step gently is all am asking :)

For example you might find the scuttalbutt side more to your likeing, the should be a outline up here in the next few week on this side.

@ajeremias have a look through the wiki pages, this project like indymedia it self is old and embeded in meany coultures which you are just steping into, step gently is all am asking :) For example you might find the scuttalbutt side more to your likeing, the should be a outline up here in the next few week on this side.

ajeremias commented

2020-01-17 22:49:24 +00:00

Poster

sorry thats ma style... i've been wanting to have this discussion about AP and ZOT since dayzz.. and it seemed appropriate.. why not?

servalan commented

2020-01-17 22:52:25 +00:00

I think scuttlebutt will be good enough for more private networks. Being able to be intermittently online and having no admins is an advantage under some conditions.

servalan commented

2020-01-17 22:55:39 +00:00

ajeremias: It's reasonable to consider them both, but for a media project I think the current UX issues of Zot apps make it a less viable candidate.

👍 2

hamishcampbell reopened this issue

2020-01-18 08:14:46 +00:00

hamishcampbell added the

geekproblem

label 2020-01-18 08:17:15 +00:00

hamishcampbell commented

2020-01-18 08:40:26 +00:00

The #geekproblem is one of the main resions the orgional indymedia project faild so addressing and keeping focues on this issue to is key to a reboot.

hamishcampbell changed title from ~~Zot vs ActivityPub~~ to Open format discushern (Was Zot vs ActivityPub)

2020-01-18 09:09:31 +00:00

ajeremias commented

2020-01-18 09:27:21 +00:00

Poster

I think that the original indymedia project failed because of attitudes like this.. how do you dare to put out ur medals and say you have a higher saying in this project, just because your are older on it? this is outraging me. anyway im out. and what the f*** is this "geek problem" medal? are you just downgrading this question? all this git repo is a #geekproblem.. thats what we are trying to solve here.

😕 1

hamishcampbell added the

Discussion

label 2020-01-18 09:41:07 +00:00

ajeremias commented

2020-01-18 11:00:28 +00:00

Poster

it seems there is a edit problem
#3 (comment)

Publish (object to a stream of objects) – to publish a story.
Subscribe (to stream of objects) – to a person or organization, a page, a group, a subject etc.
Moderate (stream or object) – you can say I like/not like (push/pull or yes/no) this etc.
Rollback (stream) – you can remove from your flow untrusted historical content by publishing flow/source.
Edit (meta data of object/stream) – you can edit the metadata in any site/instance/app you have a login on. This is the back-end of the project to build a DIY trust based grassroots semantic web. The front-end can be anything you like, for example it can be regional/city/subject based indymedia sites. The data cauldron and the golden ladle
(https://unite.openworlds.info/Open-Media-Network/Open-Media-Network/wiki)

How do you want to achieve this, if you didn't analyze deeply if AP or ZOT are able to do this? #geekproblemISthePROBLEM! or maybe im missing some of your documentation.

it seems there is a edit problem https://unite.openworlds.info/indymedia/indymedia-reboot/issues/3#issue-27 > Publish (object to a stream of objects) – to publish a story. > Subscribe (to stream of objects) – to a person or organization, a page, a group, a subject etc. > Moderate (stream or object) – you can say I like/not like (push/pull or yes/no) this etc. > Rollback (stream) – you can remove from your flow untrusted historical content by publishing flow/source. > Edit (meta data of object/stream) – you can edit the metadata in any site/instance/app you have a login on. This is the back-end of the project to build a DIY trust based grassroots semantic web. The front-end can be anything you like, for example it can be regional/city/subject based indymedia sites. The data cauldron and the golden ladle > (https://unite.openworlds.info/Open-Media-Network/Open-Media-Network/wiki) How do you want to achieve this, if you didn't analyze deeply if AP or ZOT are able to do this? #geekproblemISthePROBLEM! or maybe im missing some of your documentation.

muppeth commented

2020-01-18 12:13:02 +00:00

sorry for typos and stuff but wrote it on the go on mobile

Of course its a geekproblem if i understand what i mean by it.
However i think you are at this stage of the project if i understand correctly. Its important to think in long run what issues you might have in the future. I did that with my project ad despite AP popularity we have choosen zot. And it is because a lot of issues AP does not support, supports poorly or will never support, is already up and running on zot.
And you have to solve those issues before you even look at UI/UX. All quoted above requirenments (posting, subscribing, editing and deleting) is already implemented and working well in hubzilla. Additionally you have much fine grained ACL possibilities and can have multiple profiles per account. At the end not only you need to implent core backend functionalities but also frontend while in zot you only need frontend improvements.

Additionally, even if you implement all those into AP which will probably cost you lots of money, you need all other AP server implementations and clients to implement. That wont be easy. On contrary, hubzilla and zap already have all that and the core team behind both is very flexible and enthusiastic.

For some reason AP has taken off which is great. The geekproblem as you say (i guess you try to undermine it somehow as not important) is at this srage of your project critical. If you make bad choices now, you will need to deal with them in long run too.
I'm not saying zot is perfect but since i accociated indymedia with rather privacy aware project, i would assume the choice of tech is also privacy oriented.

*sorry for typos and stuff but wrote it on the go on mobile* Of course its a geekproblem if i understand what i mean by it. However i think you are at this stage of the project if i understand correctly. Its important to think in long run what issues you might have in the future. I did that with my project ad despite AP popularity we have choosen zot. And it is because a lot of issues AP does not support, supports poorly or will never support, is already up and running on zot. And you have to solve those issues before you even look at UI/UX. All quoted above requirenments (posting, subscribing, editing and deleting) is already implemented and working well in hubzilla. Additionally you have much fine grained ACL possibilities and can have multiple profiles per account. At the end not only you need to implent core backend functionalities but also frontend while in zot you only need frontend improvements. Additionally, even if you implement all those into AP which will probably cost you lots of money, you need all other AP server implementations and clients to implement. That wont be easy. On contrary, hubzilla and zap already have all that and the core team behind both is very flexible and enthusiastic. For some reason AP has taken off which is great. The geekproblem as you say (i guess you try to undermine it somehow as not important) is at this srage of your project critical. If you make bad choices now, you will need to deal with them in long run too. I'm not saying zot is perfect but since i accociated indymedia with rather privacy aware project, i would assume the choice of tech is also privacy oriented.

❤️ 1

servalan commented

2020-01-18 12:35:29 +00:00

Unless a lot of designers volunteer to overhaul the UI of Zap/Hubzilla I would consider it to be a non-starter. ActivityPub has flaws but is good enough, with a written protocol specification.

A major reason not to choose Zot right now is the mobile experience. You can be sure that the primary way that people interact with any media will be via mobile. There is Nomad, but my guess is that most people will not consider this to be acceptably usable and reject it quickly, whereas fediverse apps have good UX already.

Unless a lot of designers volunteer to overhaul the UI of Zap/Hubzilla I would consider it to be a non-starter. ActivityPub has flaws but is good enough, with a written protocol specification. A major reason not to choose Zot right now is the mobile experience. You can be sure that the primary way that people interact with any media will be via mobile. There is Nomad, but my guess is that most people will not consider this to be acceptably usable and reject it quickly, whereas fediverse apps have good UX already.

❤️ 1

muppeth commented

2020-01-18 12:59:32 +00:00

And with AP what implementation do you have to replace indymedia UI? Not sure i understand your project specs then.

😕 1

muppeth commented

2020-01-18 13:02:16 +00:00

As far as i see you are planning to build new ui for pleroma so what is the difference?

😕 1

muppeth commented

2020-01-18 13:15:06 +00:00

Actually looking at your idea from the other issue (pleroma) it would be super easy to implement such front page view in hubzilla. More over, users thatnks to remote auth could view the front page as you design it or use their network stream/forum view to see rhe post from within their account (their chosen ui). Not to mention each group or category within your indy insrance could have a custom ui's by simply modifying templates to their needs (for example if they dont need cloud tag they could remove it etc) or modyfying theme schema colors and backgrounds (very easy without need to know css as there is ui for this elements defined by theme author which meams anything in the the can be 'variabled').
And yeah currently there is only nomad butwe are working on nomad 2.0 now and as i stated before viewing indy news from tusky is going to be the same stream mess as any other person writting posts. Not to mention threaded comments, following/unfollowing discussions etc.

You guys go for whatever, i just want to show you that there are things to consider.so thinking out of the box a little might be helpful. Fact everyone went AP doesnt mean its the best thing to do. And in long run might be a bad choice overall.

Actually looking at your idea from the other issue (pleroma) it would be super easy to implement such front page view in hubzilla. More over, users thatnks to remote auth could view the front page as you design it or use their network stream/forum view to see rhe post from within their account (their chosen ui). Not to mention each group or category within your indy insrance could have a custom ui's by simply modifying templates to their needs (for example if they dont need cloud tag they could remove it etc) or modyfying theme schema colors and backgrounds (very easy without need to know css as there is ui for this elements defined by theme author which meams anything in the the can be 'variabled'). And yeah currently there is only nomad butwe are working on nomad 2.0 now and as i stated before viewing indy news from tusky is going to be the same stream mess as any other person writting posts. Not to mention threaded comments, following/unfollowing discussions etc. You guys go for whatever, i just want to show you that there are things to consider.so thinking out of the box a little might be helpful. Fact everyone went AP doesnt mean its the best thing to do. And in long run might be a bad choice overall.

❤️ 1

hamishcampbell commented

2020-01-18 14:26:52 +00:00

Yep the OMN that the indymedia reboot is based on is completely format agnostic. we are doing activertypub first as it has a HUGE advantage in adoption and UI currently its a no brainer to go this path first. If we implemented zot we would also have to implement activertypub so we were not in a tiny ghetto and cut off from all those apps and users, this would be shooting the project in the foot and self gettoiseing which is not a good look.

This self mutilation would not be the first time this was forced on an open project sadly. I remember when indymedia UK implemented RSS they had to build their own actavist flavor which was incompatible with standards based RSS of course nobody else implemented this flavor so indymedia’s RSS was utterly pointless. Shooting in the foot comes to mind and this is a good easey to understand example of the #geekproblem

Another #geekproblem we need to think about is signal to noise, this destroys open projects as fast as failer in standards adoption https://en.wikipedia.org/wiki/Signal-to-noise_ratio

Yep the OMN that the indymedia reboot is based on is completely format agnostic. we are doing activertypub first as it has a HUGE advantage in adoption and UI currently its a no brainer to go this path first. If we implemented zot we would also have to implement activertypub so we were not in a tiny ghetto and cut off from all those apps and users, this would be shooting the project in the foot and self gettoiseing which is not a good look. This self mutilation would not be the first time this was forced on an open project sadly. I remember when indymedia UK implemented RSS they had to build their own actavist flavor which was incompatible with standards based RSS of course nobody else implemented this flavor so indymedia’s RSS was utterly pointless. Shooting in the foot comes to mind and this is a good easey to understand example of the #geekproblem Another #geekproblem we need to think about is signal to noise, this destroys open projects as fast as failer in standards adoption https://en.wikipedia.org/wiki/Signal-to-noise_ratio

👍 1

hamishcampbell commented

2020-01-18 14:35:06 +00:00

I just want to look at hubzilla and its obviously the wrong project to build what we are planning. If you are unsure go and read up what we are doing.

OMN - MORE IMPORTANT FOR WHAT IT DOES NOT DO

I just want to look at hubzilla and its obviously the wrong project to build what we are planning. If you are unsure go and read up what we are doing. OMN - MORE IMPORTANT FOR WHAT IT DOES NOT DO

👎 1

wakest commented

2020-01-19 19:31:34 +00:00

@hamishcampbell you're response to new people here is very disappointing and to put it in you're own words "on that subject lets poor on the positive and not feed the problem". I (personally) agree that AP is the right choice. Responding with words such as "its obviously ... wrong" is not constructive and comes off as really rude to someone who "obviously" cares enough to come here and talk about these problems together. If this is indeed an open project we need to be more open about how we respond to people who have ideas that are different then are own.

hamishcampbell commented

2020-01-19 20:21:32 +00:00

sorry, its hard work juaggaling all the input need a few more peole to try and hold focues #4opens I get sttressed.

❤️ 1

ajeremias commented

2020-01-19 20:53:25 +00:00

Poster

Can we at minimum stop diminishing the role and the work of the geeks on this problem? CAN WE NOT USE #GEEKPROBLEM? I find it very offensive.

❤️ 1

ajeremias commented

2020-01-19 21:00:15 +00:00

Poster

Also, on if AP or ZOT makes more sense for this project, I'm still full of doubts, because if this project is about implementing one more gui, on top of all guis which AP already has.. like pixelfed, peertube, mastodon, plume, etc... then i dont understand the point of this project, because this networks are already doing a very good job. i think this project may make sense if we join together our forces to help out a project that we want to support, and then i would either go towards helping out on mobilizon or hubzilla...

btw.. hubzilla implements AP already.

so please can someone englight me on what this project is about? that scheme is full of enigmas... thanks.

more on what hubzilla is.. if you don't known.
https://medium.com/we-distribute/the-do-everything-system-an-in-depth-review-of-hubzilla-3-0-692204177d4e

Also, on if AP or ZOT makes more sense for this project, I'm still full of doubts, because if this project is about implementing one more gui, on top of all guis which AP already has.. like pixelfed, peertube, mastodon, plume, etc... then i dont understand the point of this project, because this networks are already doing a very good job. i think this project may make sense if we join together our forces to help out a project that we want to support, and then i would either go towards helping out on mobilizon or hubzilla... btw.. hubzilla implements AP already. so please can someone englight me on what this project is about? that scheme is full of enigmas... thanks. more on what hubzilla is.. if you don't known. https://medium.com/we-distribute/the-do-everything-system-an-in-depth-review-of-hubzilla-3-0-692204177d4e

ajeremias changed title from ~~Open format discushern (Was Zot vs ActivityPub)~~ to Open format discussion (Zot vs ActivityPub)

2020-01-19 21:00:41 +00:00

ajeremias removed the

geekproblem

label 2020-01-19 21:00:49 +00:00

hamishcampbell added the

geekproblem

label 2020-01-20 11:43:37 +00:00

hamishcampbell commented

2020-01-20 11:43:37 +00:00

The one we are using has all the open tools implemented as we are building a network not a portal, we are radically decentalsing power away from the geeks to the users. In this the project is "more important for what it does not do" by not doing many things as a portal we open/empower normal people to build their own solutions as a network.

Top down geek standards, top down geek working, top down geek solutions are clearly the #geekproblem we are trying to overcome in this reboot. it's important to have this in mind for all thinking #4opens otherwise the signal to noise issue will degraded and kill the seed before it has a chance to grow.

Also the is no blocking happening you can build a zot version based on hubzilla, that is a good outcome, BUT its not the project you have come onbord and we are building here. we have two standerds based paths activertypub and scuttalbutt and need to prioratise and focues the resorces to get these projects into existence to plant the first seeds of the network so more seeds can be planted and grow. We plan a divers forist not a mon colture as a outcome.

A part of the #'geekproblem is fighting over "open" standerds. I put in a illustration of this about indymedia and its implemetation of RSS.

"dont hate the media be the media"

The one we are using has all the open tools implemented as we are building a network not a portal, we are radically decentalsing power away from the geeks to the users. In this the project is "more important for what it does not do" by not doing many things as a portal we open/empower normal people to build their own solutions as a network. Top down geek standards, top down geek working, top down geek solutions are clearly the #geekproblem we are trying to overcome in this reboot. it's important to have this in mind for all thinking #4opens otherwise the signal to noise issue will degraded and kill the seed before it has a chance to grow. Also the is no blocking happening you can build a zot version based on hubzilla, that is a good outcome, BUT its not the project you have come onbord and we are building here. we have two standerds based paths activertypub and scuttalbutt and need to prioratise and focues the resorces to get these projects into existence to plant the first seeds of the network so more seeds can be planted and grow. We plan a divers forist not a mon colture as a outcome. A part of the #'geekproblem is fighting over "open" standerds. I put in a illustration of this about indymedia and its implemetation of RSS. "dont hate the media be the media"

ajeremias commented

2020-01-20 13:57:50 +00:00

Poster

This project is such a crap... you keep on attacking and diminishing the geeks as if they are the worst problem in the world. and you are just trying to get people on board at your command. guess what.. we are anti-hierarchies and we won't follow you. so get a life.. bye.

muppeth commented

2020-01-20 14:47:59 +00:00

@hamishcampbell I think you are looking at it wrong asn somehow feel like you need to take defensive position. I will speak for myself here but the intention is not to introduce geek only platform. Believe me. At disroot we have chosen software to use on our platform based on sexiness and eas of use, which sometimes resulted in diminished functionality. We decided to keep registration open despite people suggesting paid only accounts or invite only to be as much inclusive. We've written tutorials and howtos etc. So my goal here is definitely not to push eg. i3 as a go to desktop for newbies. I'm not fanatic, just want to share possibilities and potential of hubzilla (took me few years to discover it also).

I'm expressing my opinion as a person using and operating federated, decentralized platforms for quite a while now. And my opinion is that if you are trying to build such a platform you are thinking about (I read the user stories) in my opinion Hubzilla is the best aproach because it covers all the use cases and was build exactly to be a publishing/CMS platform rather then just a micro-blogging. Pleroma (which u are planning to use) is rather very narrow single scope application for a specific use case (micro blogging).
For instance:

You can build a indymedia news portal quite easily. This allows people to have easy way to view all the news, categorised, with sidebars showing recent news feed etc. When looking at the portal user only sees the news posted to it,so unlike pleroma/mastodon/tusky you dont get the noise of the other posts from all over the place. For instance if I'm not on your server, I cant see indymedia posts in the local timeline, which means it will be hard for me to follow indymedia specific topics. Of course you can use a hashtag but IMO this is already creating workaround which non-geeks will not find easy to use. Another thing which you specify is permissions, moderation etc. With current implementation of hubzilla, building a portal for indymedia news is already there. You can assign multiple people to be able to moderate everything posted to it, you have possibility to remove and edit posts and comments (in case things change and people want to adjust stuff. think about an action with changing timeline etc where newest updates are edited in the top post). thanks to OWA (Open Web Authentication) users can view indymedia page as it suppose to look like (designed by you) no matter the UI choice of their profiles (my hubzilla channel can be totally different then yours but when I interact it yours I visit your page). Additionally people then can interact with your page using their credentials and you can define roles and permissions people get (if you want to decide who can post or comment based on some criteria). At the end you get a page like indymedia currently but totally decentralized and federated with possibliity to migrate accounts, create syncronious copies of the accounts (think of a gov. shutting your node but you have copy of it hosted on all the other nodes). Additionally people that dont want to look at indymedia via your portal can still interact with it the way they want, either their Netowrk stream, public posts, or as in mastodon, federated timeline, tags etc. So both worlds come together. For people who do not have an account on either federated network, they can still browse through your portal as if its just a website.
Guy with a blog
Hubzilla allows for addons turning your social network into a CMS essentially (with all the goodies of federated network built in). Usrs can modify the way their site looks, and can post either articles, wikis, notes, pictures etc.
Guy selling his video game
Hubzilla has a shopping cart addon.
Person posting photos from a trip
Hubzilla has a gallery addon, plus it supports WebDav standard which means you can easily mount your hubzilla cloud drive n your laptop/phone and either upload automatically, backup, copy you name it. Its essentially your file sharing cloud account.

The only thing missing is your imagination to create nice looking (if you dont like the current ones) themes to hot all those use cases. The heavy lifting is already done. So in my opinion if you anyway need to write a big chunk of backend for pleroma to make it possible (which also means probably AP as well to stay compatible and not start making features only available for pleroma) and you need to create UI's for all the use cases anyway, you might as well help existing project like Hubzilla to achieve that.

This was one of the reasons why we did not just go for mastodon with disroot what it was a hip thing to do. We realized the potential of hubzilla and the fact it does solve already a lot of things that arent solved in AP world.
I am here solely to show other possibilities that might be much more suitable for your use case. Specially as I see you dont have an army of people that want to dive into creating backends/frontends for it.

Below you have different views on the same page depending on your use:

how it looks like on mastodon
how it looks like on the portal page (disroot-forum.png)
how it looks like from just network stream (disroot-network-stream.png)
how it looks like form my account (so the theme can be totally different Screenshot from 2020-01-20 15-44-48.png)

The templates and themes of course can be modified and users not only can build their own templates and schemas, you can have multiple themes for users to choose from and they can choose different template for each of their channels (for example all the use cases you presented could be one account, with just different profiles depending on what they want to do / achieve).

@hamishcampbell I think you are looking at it wrong asn somehow feel like you need to take defensive position. I will speak for myself here but the intention is not to introduce geek only platform. Believe me. At disroot we have chosen software to use on our platform based on sexiness and eas of use, which sometimes resulted in diminished functionality. We decided to keep registration open despite people suggesting paid only accounts or invite only to be as much inclusive. We've written tutorials and howtos etc. So my goal here is definitely not to push eg. i3 as a go to desktop for newbies. I'm not fanatic, just want to share possibilities and potential of hubzilla (took me few years to discover it also). I'm expressing my opinion as a person using and operating federated, decentralized platforms for quite a while now. And my opinion is that if you are trying to build such a platform you are thinking about (I read the user stories) in my opinion Hubzilla is the best aproach because it covers all the use cases and was build exactly to be a publishing/CMS platform rather then just a micro-blogging. Pleroma (which u are planning to use) is rather very narrow single scope application for a specific use case (micro blogging). For instance: 1. You can build a indymedia news portal quite easily. This allows people to have easy way to view all the news, categorised, with sidebars showing recent news feed etc. When looking at the portal user only sees the news posted to it,so unlike pleroma/mastodon/tusky you dont get the noise of the other posts from all over the place. For instance if I'm not on your server, I cant see indymedia posts in the local timeline, which means it will be hard for me to follow indymedia specific topics. Of course you can use a hashtag but IMO this is already creating workaround which non-geeks will not find easy to use. Another thing which you specify is permissions, moderation etc. With current implementation of hubzilla, building a portal for indymedia news is already there. You can assign multiple people to be able to moderate everything posted to it, you have possibility to remove and edit posts and comments (in case things change and people want to adjust stuff. think about an action with changing timeline etc where newest updates are edited in the top post). thanks to OWA (Open Web Authentication) users can view indymedia page as it suppose to look like (designed by you) no matter the UI choice of their profiles (my hubzilla channel can be totally different then yours but when I interact it yours I visit your page). Additionally people then can interact with your page using their credentials and you can define roles and permissions people get (if you want to decide who can post or comment based on some criteria). At the end you get a page like indymedia currently but totally decentralized and federated with possibliity to migrate accounts, create syncronious copies of the accounts (think of a gov. shutting your node but you have copy of it hosted on all the other nodes). Additionally people that dont want to look at indymedia via your portal can still interact with it the way they want, either their Netowrk stream, public posts, or as in mastodon, federated timeline, tags etc. So both worlds come together. For people who do not have an account on either federated network, they can still browse through your portal as if its just a website. 2. Guy with a blog Hubzilla allows for addons turning your social network into a CMS essentially (with all the goodies of federated network built in). Usrs can modify the way their site looks, and can post either articles, wikis, notes, pictures etc. 3. Guy selling his video game Hubzilla has a shopping cart addon. 4. Person posting photos from a trip Hubzilla has a gallery addon, plus it supports WebDav standard which means you can easily mount your hubzilla cloud drive n your laptop/phone and either upload automatically, backup, copy you name it. Its essentially your file sharing cloud account. The only thing missing is your imagination to create nice looking (if you dont like the current ones) themes to hot all those use cases. The heavy lifting is already done. So in my opinion if you anyway need to write a big chunk of backend for pleroma to make it possible (which also means probably AP as well to stay compatible and not start making features only available for pleroma) and you need to create UI's for all the use cases anyway, you might as well help existing project like Hubzilla to achieve that. This was one of the reasons why we did not just go for mastodon with disroot what it was a hip thing to do. We realized the potential of hubzilla and the fact it does solve already a lot of things that arent solved in AP world. I am here solely to show other possibilities that might be much more suitable for your use case. Specially as I see you dont have an army of people that want to dive into creating backends/frontends for it. Below you have different views on the same page depending on your use: - how it looks like on mastodon - how it looks like on the portal page (disroot-forum.png) - how it looks like from just network stream (disroot-network-stream.png) - how it looks like form my account (so the theme can be totally different Screenshot from 2020-01-20 15-44-48.png) The templates and themes of course can be modified and users not only can build their own templates and schemas, you can have multiple themes for users to choose from and they can choose different template for each of their channels (for example all the use cases you presented could be one account, with just different profiles depending on what they want to do / achieve).

disroot-mastodon-view.png

151 KiB

disroot-forum.png

725 KiB

disroot-network-stream.png

709 KiB

Screenshot from 2020-01-20 15-44-48.png

1.0 MiB

❤️ 1

muppeth commented

2020-01-20 14:59:11 +00:00

Just to make something clear. My intention was not to show how ugly it looks on mastodon but that you can view the same thing from different platforms.
But hubzilla does give additional features (mentioned above)

Just to make something clear. My intention was not to show how ugly it looks on mastodon but that you can view the same thing from different platforms. But hubzilla does give additional features (mentioned above)

hamishcampbell commented

2020-01-20 15:12:08 +00:00

Thanks for the feedback. But the project you've described is a different project from the one we're trying to launch, which is clearly built ontop of the activitypub ecosystem.
Hope all the best with your work.

Thanks for the feedback. But the project you've described is a different project from the one we're trying to launch, which is clearly built ontop of the activitypub ecosystem. Hope all the best with your work.

muppeth commented

2020-01-20 15:17:52 +00:00

according to https://unite.openworlds.info/attachments/8d1f8465-1598-4b78-a2b1-039b5576a8e7
and
https://github.com/Openmedianetwork/OMN/wiki/user-storys

it is (which is what I based my opinion on and feedback). I havent found proper description perhaps. I do have a feeling you are not interested in any feedback or suggestion and your mind is made up, so I wont bother you anymore because there is no point. I wish you all the best and see you on the federated network one way or another.

according to https://unite.openworlds.info/attachments/8d1f8465-1598-4b78-a2b1-039b5576a8e7 and https://github.com/Openmedianetwork/OMN/wiki/user-storys it is (which is what I based my opinion on and feedback). I havent found proper description perhaps. I do have a feeling you are not interested in any feedback or suggestion and your mind is made up, so I wont bother you anymore because there is no point. I wish you all the best and see you on the federated network one way or another.

👍 1

ajeremias removed the

geekproblem

label 2020-01-22 09:02:37 +00:00

hamishcampbell commented

2020-01-22 09:28:40 +00:00

Can we please stop removeing the #geekproblem tag as this project has been set up to directly medaite the #geekproblem we can discues the issue if you like.

hamishcampbell added the

geekproblem

label 2020-01-22 09:28:40 +00:00

OMN closed this issue

2020-01-23 08:57:16 +00:00

wakest commented

2020-01-25 21:32:36 +00:00

@hamishcampbell do you not understand that Hubzilla is federating with ActivityPub? I am following people from my Mastodon instance on Hubzilla, Zap and Friendica (All originally developed by the same person (Mike Macgirvin)