@chris
Yes, I love the unix principle. No, I cannot imagine using more than one account. I'm not able to see the connection between those two points. (Transfered to the unix world, it seems to me like the opposite of "one job - one tool", but many tools that do the same - just in another flavour).

@Martin 🪷 Schmitz

well there are more and more specialists Fedi apps for special tasks .. like this one here i discovered today: https://rss-parrot.net/

and there are many other examples...

e.g. Why a special Fedi App just for sharing photos or discussing books?

So is this the way to go in the Fedi?

@chris
Excellent example. Personally I see absolutly no point in mirroring RSS-Feeds to the fediverse. I don't want them in my mastodon timeline. I make heavy use of a feed reader. THIS is, what "one job, one tool" means to me. 😀
Your mileage may vary.

I think Hubzilla embodies the Unix principle. In an Unix operating system, you can install and use different apps, and each app is doing one task. But you as a user don't have to create user accounts for each app. You can use one single user account for that Unix operating system.
The same happens with Hubzilla. You create one user account, and then you install and use addons, for each task a different one. Hubzilla is like an operating system, just a online one.

@Martin 🪷 Schmitz

I make heavy use of a feed reader. THIS is, what "one job, one tool" means to me.

So this feedreader is installed online? Than you would be willing to install or register for that right?

@Ema エマ

I think Hubzilla embodies the Unix principle.

Yes and No... Looking from the inside of HZ it is very true what you say... Looking from the AP World on it HZ appears in a monolithic design...

This is not understood at all... and causing a lot of misunderstanding by DEVs... they rather start a completely new project than digging deeper into the options they could find in a system like Hubzilla.

E.g. there is this App flohmarkt

https://codeberg.org/grindhold/flohmarkt

https://fediflohmi.ch/

by @grindhold

featured by @Lioh @Lioh and others

this app does not seam to have even a Group / Forum function but this would be basic for such a project. Hubzilla and also streams have all you need for such a flee-market fedi app.. they have also location infos... streams has even a search by location option... something which is missing in all other Fedi projects as well...

It would be much easier to code a HZ addon and work on a nice and fitting theme but DEVs do not get it and start a completely new project because they do not tread HZ as the one Unix operating system of their choice.

Why is that the case ?

Most of the confusion stems from "Write programs to work together". There are lots of fediverse projects that do one thing and some can even do it well, but they don't work together with anything else. Hubzilla takes a modular approach where you've got a common UI, a common permission system, and common framework for the modules to interact. So that they all work together. Which approach is the true Unix philosophy? What we once knew as Unix is now GNU-Linux and it's built around a graphical (not text) interface with big behemoth programs that sing, dance, and fly a kite and don't work with anything else. So it really doesn't matter. Interestingly, most of the programs that I do use are big behemoths that try to be operating systems and contain their own environments separate from anything else on the system, like firefox  and emacs and phpstorm and universal media player. Do you spend all your time on the computer piping text files through awk and sending them to lp or lpr?  Really?

@𝓒𝓱𝓻𝓲𝓼

It would be much easier to code a HZ addon and work on a nice and fitting theme but DEVs do not get it and start a completely new project because they do not tread HZ as the one Unix operating system of their choice.

Why is that the case ?

Possible reason #1: They've never heard of Hubzilla. The most likely reason.

Possible reason #2: They've only ever heard the name, but they don't know what Hubzilla is and what could be done with it. Including expanding it with third-party apps.

Possible reason #3: "Not worth it, nobody uses Hubzilla anyway."

@Jupiter Rowland So looking at the new website of HZ, do you feel we are doing a good job for promoting the options of HZ also for DEVs?

@𝓒𝓱𝓻𝓲𝓼 I'm not a dev, I don't know what makes a website work for them.

Besides, the website will be utterly worthless for anyone who hasn't even heard the name Hubzilla or of Hubzilla's mere existence. How should such people even search for a term which they've never heard or read in their lives?

@Jupiter Rowland Mike has talked so much about the project here and in forums... so the projects should be known by now to everybody who is truly interested in federated networks... so i guess there still must be an other cause...

DEVs also meet at events like 37C3 and so on.. we never really presented the project there didn't we? I tried last year but I'm not a DEV either...

@𝓒𝓱𝓻𝓲𝓼

@Jupiter Rowland Mike has talked so much about the project here and in forums... so the projects should be known by now to everybody who is truly interested in federated networks... so i guess there still must be an other cause...

And how many Mastodon users do you think follow Hubzilla threads? How many have joined the Hubzilla forums?

There have been countless cases since November 2022 of developers joining Mastodon. Getting all giddy and excited about it. Building stuff against it, be it websites, be it mobile apps, whatever. Driving their developments to stable 1.x releases. And it was only after that that they found out that there's more to the Fediverse than Mastodon. Like, there's anything else in the Fediverse than Mastodon. Like Pleroma or Pixelfed or PeerTube.

Even then, the path to learning that Hubzilla exists was long. And I guess many of these still don't know what Hubzilla is, much less what can be done with it. Especially since pretty much the only non-Mastodon Fediverse talk on Mastodon is either Threads drama or Firefish being dead drama.

Believe me, I follow Mastodon closely, even though the vast majority of my Mastodon contacts is silenced.

@𝓒𝓱𝓻𝓲𝓼

So looking at the new website of HZ, do you feel we are doing a good job for promoting the options of HZ also for DEVs?

No, the website needs to be restructured.

1. The help system that comes with the Hubzilla software is good for a quick reference for admins and users but it is not very good for developers. Developers need a knowledge base that is much more comprehensive and easier to search.

2. We need a section that lists all of the addons and themes that are available for Hubzilla, to demonstrate to both users and developers that Hubzilla can be modified.

3. We need to have a section that lists the possible use cases, and which addons need to be installed for that use case, preferably with step-by-step instructions on how to set everything up.

4. It needs to be easier to find the support channels and forums, So better navigation on the site is required. We also need to create the developer support forum channel, which we are still working on.

And several other things. We have a lot to do on the website to make Hubzilla attractive to both users and developers.

@Jupiter Rowland Possible reason #4: Very few people care about existing code. Given a lot of folks doing this kind of development in their spare time for fun and for the sake of it, starting "greenfield" has many motivational aspects that seem interesting: Play with newer technology (opposed to old existing projects on "dusty" old stacks). Build a "clean" architecture and stack (opposed to old existing projects where you usually get stuck in that ball of mud most "real-world" code tends to become having seen some iterations, teams and hacks to work around issues missed early in development). But also: Be an authority, a "project lead", ... (opposed to existing, old projects where collaboration most likely would mean working with others, accept decisions by an existing team that didn't just wait for you to come and explain to them how things need to be done "right"). Looking through a load especially of the younger AP implementations, this feels to shine through in virtually each one of these.

@𝓒𝓱𝓻𝓲𝓼

@Kristian That is a very valid psychological argument.
@Scott M. Stolz Kristian's comment is a massive marketing argument for Hubzilla that we should put out there: The add-on system lets you start your own "project within a project" - By developing apps and addons that expand Hubzilla functionality, you can be the mostly un-influenced* authority for a local, independent piece of code.
* well, you are tied to php, and whether the use of the comanche framework is an advantage oder impedement might be seen differently, but in principle, this is something rarely any other Fediverse project offers...

@Witcraft (Hubzilla)

Kristian's comment is a massive marketing argument for Hubzilla that we should put out there: The add-on system lets you start your own "project within a project" - By developing apps and addons that expand Hubzilla functionality, you can be the mostly un-influenced* authority for a local, independent piece of code.

In theory, yes, but practically, only in edge-cases in which you need Hubzilla's sheer CMS firepower along with whatever you want to add. If you don't, (streams) would be the better choice because it comes with little more than the bare necessities, its development is quicker, and it has better ActivityPub support.

@Jupiter Rowland

In theory, yes, but practically, only in edge-cases in which you need Hubzilla's sheer CMS firepower along with whatever you want to add. If you don't, (streams) would be the better choice because it comes with little more than the bare necessities, its development is quicker, and it has better ActivityPub support.

This is something that I have personally been considering. I have several use cases that radically change how the platform operates. I can build them on Hubzilla or Streams because both can be modded.

Most of my projects need CMS capabilities, but Hubzilla's CMS capabilities are too simple for my needs, in most cases. And the fact that the content must be tied to a channel means I have to manually code pages for the main website anyway, which eliminates the benefit of having CMS features in Hubzilla, at least for my use cases.

Hubzilla's existing CMS and wiki apps are okay for simple websites, and I will use them for simple websites, but more complex stuff will use a content & knowledge management system of my own design.

Since I am using my own CMS addon, does that mean Streams would be better? Not necessarily, since Streams is so stripped down that some of the features I need for my CMS have been removed. I would have to recreate some of the code that already exists in Hubzilla.

But there is an additional factor and that is that Streams usually has the bleeding edge of technology, where production websites usually prefer something a little more stable and that maintains backwards compatibility, and Hubzilla seems to do that better than Streams. So there is an argument that Hubzilla would be better for production websites, and then we backport new stuff from Streams into Hubzilla once it has been tested and used in the wild for awhile.

So, the question comes up, will I use Hubzilla, or will I use Streams, or will I fork one of them to create something new?

Well, that depends on the website project, and I might wind up doing all three at some point.

I am interested in promoting OpenWebAuth (which I am promoting as Magic Signon), Nomadic Identity, ActivityPub, and Zot and Nomad. And I think Zot and Nomad are important to maintain because having a competing protocol that we control can be used to put pressure on ActivityPub and the AT protocol and others and can be used to demonstrate new features in the wild. Plus, I want the option to have sites that can talk to each other via Zot that are not connected to ActivityPub, basically creating a private network with its own realm.

So, it depends on the use case. Sometimes Hubzilla is the best choice, sometimes Streams is the best choice, and sometimes our changes are so radical that it just makes sense to fork one of them.

I plan on promoting Hubzilla, Streams, and Neuhub, and am hoping to foster collaboration between the projects. And if it makes sense, stuff created on one could be ported to another.

@Witcraft (Hubzilla)

Kristian's comment is a massive marketing argument for Hubzilla that we should put out there: The add-on system lets you start your own "project within a project" - By developing apps and addons that expand Hubzilla functionality, you can be the mostly un-influenced* authority for a local, independent piece of code.

Quite frankly, the addon system is the only reason why I don't have to fork Hubzilla or Streams, for most projects.

Ideally, it would be similar to WordPress, where you have a well-documented plugin system where people can create what they want without permission from a central authority.

@Scott M. Stolz

Since I am using my own CMS addon, does that mean Streams would be better? Not necessarily, since Streams is so stripped down that some of the features I need for my CMS have been removed. I would have to recreate some of the code that already exists in Hubzilla.

(streams) is only better than Hubzilla if you either need something sleek and lightweight, or if you want to interact with Mastodon and other ActivityPub users a lot while making use of nomadic identity, including the option of moving.

If your goal is basically a kind of Web portal, and you aren't going to use any of the "social" features, then you can stay with Hubzilla.

If what you're planning will work perfectly reliably with Pubcrawl off, you can stay with Hubzilla.

If, on the other hand, you require connectivity with OStatus, Diaspora* etc. along with ActivityPub, you have to stay with Hubzilla.

If you need a GUI for CalDAV, you have to stay with Hubzilla. (streams) has CalDAV, but no GUI for it.

@Jupiter Rowland

(streams) is only better than Hubzilla if you either need something sleek and lightweight, or if you want to interact with Mastodon and other ActivityPub users a lot while making use of nomadic identity, including the option of moving.

but then, what are all these mastodon accounts doing in my hubzilla timeline?

@Alexander Goeres Hubzilla can interact with Mastodon pretty well. The vast majority of my connections is on Mastodon.

But (streams) and its Nomad protocol have better and deeper integration of ActivityPub, even into nomadic identity, than Hubzilla and its Zot6 protocol. I can't give you any specifics, though. But Nomad and (streams) are optimised for integrating ActivityPub into something nomadic. Zot6 is only Zot6, and on Hubzilla, everything that isn't Zot6 is bolted on on the outside.

If you have a channel with clones and a whole bunch of Mastodon contacts, and you have to move your main instance, (streams) will probably cause you less trouble than Hubzilla.

There may also be other compatibility issues between ActivityPub and Hubzilla that have been solved on (streams).

Mikes goal is to implement all we can do with ZOT / Nomad also into AP so that we don't need an other protocol anymore .. while reading the last weeks posts from Mike we can guess that it really moves in this direction an apart form ZOT ... at the end there will me NO Nomad anymore... and Hubzilla will stay there as a multi function modular behemoth, seen from other fedi apps as monolith, which can do things no other fedi app can do but with a kind of closed up ecosystem like e.g. Wordpresse or Nextcloud has today....

If you just want to post in a decentralized network you don't need all that... but the future is not just about posting in a network but to organize your life more and more online but also responsible and confident in connection with others and for that HZs design is THE tool of the future... we can extend it in every needed and wanted direction - sing, dance, and fly a kite - and as an Account holder you need just ONE account to do it all on ONE server connected with all the other Server out there in a decentralized network ...
Does this fit to the Unix philosophy? It doesn't matter. We just have to give other DEVs a more easy chance to understand and to step into the system.

@𝓒𝓱𝓻𝓲𝓼

Mikes goal is to implement all we can do with ZOT / Nomad also into AP so that we don't need an other protocol anymore

But that is going to take time, and there still is resistance in ActivityPub to Mike's ideas. On one hand, I see some people within ActivityPub who really want to implement nomadic identity and privacy and other features Zot has had for a long time. But then I always see others state "we don't need that feature" and "it should be done as a separate protocol." So the ActivityPub world is not in agreement on whether Mike's technology should be added to ActivityPub.

In the mean time, I am not waiting. I am implementing projects using Zot, which already has the features I need. At least my projects will work together and have advanced features.

And even if ActivityPub does eventually adopt everything I need for my projects, I still think that Zot and/or Nomad need to stick around and be maintained, even if someone forks it from Mike and takes over. Competition is important. I would rather not have only one protocol to choose from. I don't care if it is popular or not, since I can just use it between my websites and communities. And if my websites have advanced features that ActivityPub websites & apps don't, that just gives my websites & apps a competitive advantage over them.

Because Zot and Nomad are much smaller protocols, they can rapidly roll out new features, if they wanted to, whereas ActivityPub would take years to implement the same features. And Zot & Nomad's very existence puts pressure in ActivityPub to implement new features. It would be a mistake to completely abandon Zot and/or Nomad.

@Scott M. Stolz There are only two reasons why Nomad even exists.

One, Zot can't do what Mike wanted nomadic identity to do. Especially not in connection with the ActivityPub part of the Fediverse.

Two, ActivityPub can't do that either due to some people's stubbornness.

If ActivityPub gains all features Mike needs in it, he'll happily transition (streams) to ActivityPub and drop Nomad like a hot potato. Nomad will have outlived its purpose at that point.

@Jupiter Rowland I understand that. But there are actually some features I want Zot or Nomad to have that I know would never be introduced to ActivityPub since they are too niche. So maybe it would not be useful for Mike, but it may be useful to keep around for other things.

For example, I may want to sync data between websites or between user accounts, and if I forked Zot or Nomad, I could add syncing features to Zot or Nomad.

Just because Mike gives up on it, that does not mean others will. Although it would be nice if we could convince Mike to allow Nomad to morph into a specialty protocol with features that would never make it into ActivityPub.

Any rumours of Nomad's death are greatly exaggerated.

"For example, I may want to sync data between websites or between user accounts, and if I forked Zot or Nomad, I could add syncing features to Zot or Nomad."

Fun factoid. Sync is built in.

@Mike Macgirvin 🖥️

Fun factoid. Sync is built in.

Could it be adapted to handle syncing a project management system, and exchanging tasks between other users. Or syncing a wiki or articles? Because that is the use case I have in mind at the moment.

Articles and cards and webpages and wikis are just items with a different item_type. Wikis are a bit weird because they also set the revision. import_items() handles all of these. If that's not happening, fire up a debugger and figure out why not.

@Mike Macgirvin 🖥️ So we could probably use the same mechanism to sync projects and tasks. I'll have to investigate that when I convert my project management system to use a federated model.

(Context: I did not like any of the options out there, so I wrote my own project management system and have been using it for years now. I am considering using Streams or Hubzilla as a base and creating a federated version of it.)

which can do things no other fedi app can do but with a kind of closed up ecosystem like e.g. Wordpresse or Nextcloud has today....

We just have to give other DEVs a more easy chance to understand and to step into the system.

Thinking about that i want to say now that it is even different...
With the HZ webpage app we can build a lot of things even without the need for provided addons/plugins.

Every Account holder who is trusted and granted the admin setting "allow code" can use all kinds of snippe, scrips and css code to build special tolls for web publishing.

Every Account owner can style the channel and the all the other webpages of a account as he likes and make it look like in individual and unique place in the web - connected to all the other websites/channels out there... and move it all to an other HZ Server.

This woks today out of the box without any need of enhancements by Devs or further development.

Not many people have explored Hubzilla in this directions.

So it is not just about DEVs for more development but also about showing and teaching "regular" Account holders what they can do with Hubzilla today.

All on a system which can be very easy self hosted...

Looking on it all together I don't know of any other web publishing tool with such options...

May I be wrong here? If so please tell me.

If I'm not wrong: Why has someone like me to write about this? Why this all is not communicated by all this institutions which fight for freedom on the web?

I really don't get it and it starts to scare me if i philosophize about the reasons.

@chris@im.allmendenetz.de I've often wondered about it myself. I think it's analogous to questions such as why did AP catch on even though it wasn't the first, or why did mastodon grew faster than Misskey, Pleroma and others? One possible answer for the second question, with regards to Misskey, could be superior documentation (at least in English) and a much better presentation to end users in layman's terms early on. Post language filtering capabilities and somewhat better moderation interface could have also been factors for faster adoption.

@𝓒𝓱𝓻𝓲𝓼

If I'm not wrong: Why has someone like me to write about this? Why this all is not communicated by all this institutions which fight for freedom on the web?

I really don't get it and it starts to scare me if i philosophize about the reasons.

For us Hubzilla users, it's perfectly normal common knowledge.

But most of the world outside the Fediverse isn't even aware of the existence of the Fediverse. With which I mean Mastodon.

For most of those few who have heard of the Fediverse, the Fediverse is Mastodon. In fact, this even applies to many many Mastodon users, even such who have been around for months.

Most of those who have heard that there's more to the Fediverse than Mastodon don't know anything beyond that. They don't know any names of Fediverse projects other than Mastodon.

Next, most of those who have heard of other actual Fediverse projects that aren't Mastodon don't know anything about these projects except for their names. And even then, they quite likely only know a few projects based on ActivityPub. The only projects of which they've got a rough idea what they do are maybe Pixelfed and PeerTube which are being advertised quite a lot as Instagram and Youtube in the Fediverse.

Not many of these have even heard of Hubzilla. Those who do only know the name.

Most of those who know more than the name "know" that Hubzilla allows for more characters than Mastodon, and that's it. Or maybe that it does some weird freaky shit (from a Mastodon point of view).

There may be about one or two dozen Fedizens who primarily use Mastodon, and who have practical experience with Hubzilla. But they've noped out of Hubzilla because they couldn't handle the UI/UX which, surprisingly, turned out vastly different from Mastodon's. And they failed to connect to their Mastodon contacts because nobody had told them that they have to turn Pubcrawl on first. So all they know about Hubzilla is that it sucks.

Those very very few who did get through Hubzilla's UI/UX usually fully made the switch, and they've never looked back.

In the meantime, even the Internet freedom fighters who make it to the Fediverse never even look beyond Mastodon, much less move beyond Mastodon. Even if many of them could have a total ball with Hubzilla.

Some, including Sascha Lobo, still aren't convinced that the Fediverse itself is worth a look, even though that's often a case of "but muh follower numbers" and thus hanging on to 𝕏 for their dear lives. So why should they try Hubzilla if someone like Sascha Lobo tells people to even steer clear of Mastodon with Mastodon standing in for the whole Fediverse?