FeedLand roadmap

[scripting]

There's a lot of new stuff here and I don't go into great detail, but I wanted to post it here first, see what feedback there is, if any.

1. The next version will not use Twitter identity. When you sign up you'll specify both a name and an email address. Both must be unique. An email confirms. Click the link and you're sent back with the credentials your browser needs to access your account. The usual dance.

2. I do not plan to transition feedland.org to use this feature. Recall that we haven't been accepting new members since December 12. Everyone who uses it has a Twitter identity and it's working, and I don't want to screw with that. As long as Twitter is willing to let use their identity service, we'll keep using it on the first FeedLand server.

3. Here's the big news: The new FeedLand server software will be available as open source, so anyone will be able to run a FeedLand instance. It's a Node.js application. Uses MySQL. You may want to hook up an S3 bucket for special features like RSS feeds for Likes. Obviously you'll need a way to send emails.

4. The open source model is WordPress. I want to make it as easy to set up a FeedLand instance as it is to start a new WordPress server. Also planning to use their open source license.

5. People can set up commercial services to host FeedLand for individuals and groups. Every instance is set up to do that. It's not a resource hog. I'm spending about $25 a month to host feedland.org for almost 1000 users.

6. The client, which runs in the browser, will not be open source. I don't want to spawn a bunch of incompatible forks. I want FeedLand and its API to be solid. By maintaining control of the client, which btw need not be the only client, I can help be sure that we're starting a developer community with some basic rules about interop. If you want to run FeedLand it has to behave like FeedLand. I've been down this road and watched others go down this road. I think this is the right way to start.

7. I don't have dates for any of this. But at this point the path is pretty certain, so I felt it was time to say where FeedLand is going.

[scripting]

It's okay to Like this post if you understand and agree.

Or if you have questions, please ask. 😄

[mistersugar]

I appreciate seeing this roadmap. Thank you for explaining the future of Feedland. Point number 5 got my attention; I imagine I will be able to set up an instance for my colleagues at Duke or for the science blogger community, or for a journalism or policy class (you've got me thinking about how this can be a good teaching tool).

[scripting]

@mistersugar --

re setting up "an instance for my colleagues at Duke or for the science blogger community" -- that is exactly the kind of use I have in mind.

you will need some new skills to do this, and this will not be a "poets" product -- it's industrial strength.

[scripting]

I just posted the roadmap on my blog.

http://scripting.com/2023/01/15/160957.html

Feel free to comment on it publicly, on your blog, or elsewhere.

[scripting]

BTW, now that the existence of an open source server product is announced --

I've been updating the client software regularly over the last few weeks so that it works with both versions, the one that uses Twitter identity (the one we all use) and the experimental one which will become the open source product.

So far I don't think anything has broken on this side which I continue to use to read the news.

But now I can report changes that might impact users of the current product and ask that you confirm that they still work as before (as they should). I am Mr No Breakage and I don't expect that to change now.

So you may see some more activity here...

[troutcolor]

Hi Dave,
Appreciate the road map. Looks grand.

I can see the potential for a FeedLand instance built round or for a community. Gathering interesting feeds with like minded people but being able to build your own slice of these feeds would be amazing. I am not sure if “my people” are rss savvy enough at the moment for this but I love the idea.

I am wondering about “this will not be a ‘poets’ project” bit? Is the “easy as WordPress” a longer term goal? Setting up WordPress is a very low bar.

Cheers

John

[scripting]

@troutcolor --

That's exactly the idea.

I envision a department of a university maintaining a FeedLand server to gather new writing from other departments at other universities. It would be of interest to everyone in the discipline and a great resource for students and the general public. I would love it if one computer science department did something like this to help me keep up to date on developments I should be aware of. It's amazing to me that in 2023 there aren't hundreds of such FeedLands.

Similarly with a workgroup at a company or a news organization.

Re poets, how one characterizes the software makes no difference to what the software is or how it is documented, but it does set expectations. I had a bad experience being the technical advisor to people who said they were poets. While the people here were always diligent and researched their questions carefully before bringing me into it, many others thought "poet" meant lazy and wanted me to do their work for them. I realized it wasn't communicating what I meant it to. I am never signing up for that again. It's a ridiculous use of my time.

I don't understand what you mean that "WordPress is a very low bar" -- does that mean it's hard to set up or easy? I can't tell.

[troutcolor]

Hi Dave,
Sorry, might be a uk idiom. WordPress is very easy indeed to set up on cheap shared hosting. A lot of these services have a one click install for WordPress. My own experience has mostly been on host with an existing Database, basically upload the files and step through a few settings on the web. A few minutes.
I am defiantly on the poet side of the technical fence. I’ve set WordPress up from scratch on a raspberry pi a few times and it is not much harder, especially as there are lots of instructions to get Apache and MySQL running. FWIW I’ve also set up river5 and 1999 on Raspberry Pi and that was only a wee bit trickier.
I think it is probably the shared hosting options that make WordPress so easy.

[scripting]

@troutcolor -- thanks for explaining.

I have been writing the setup document for the last couple of days. I also have read the wordpress docs, but haven't installed their server yet. But it will be similar, and it will be possible to package it up and make it one click, and hosting will totally make sense. You'll see soon enough how it installs, and I hope you'll review it and tell me how we could do it better. I really look forward to that. It's hard writing this stuff in a vacuum. ;-)

Also when you use the term poet, or Anton, that's fine with me. It's just something I want to disclaim. I've become associated with that idea, and given the way people interpret it, I no longer want that association. I expect I'll have to discuss this quite a bit. ;-)

Really looking forward to the next step.

[troutcolor]

I just used the term poet to lower any expectations.
I’d be more than happy to attempt any set up locally.

[scripting]

@troutcolor -- you will be doing that quite soon.

[davisshaver]

I'll be glad to take a trial run through the setup docs once they're ready! I could see us setting up an instance for LebTown members and seeing how that goes.

[scripting]

@davisshaver --

great to see you here! and it would be great if you could help with testing the open source server.

hope all is well...