It's time to update the main Fedora website!

Project Discussion
1

The current getfedora.org site has been in place since 2014, with a couple of minor refreshes. It’s time for a big refresh.

Key Goals

  1. New decade, new look!
  2. Refreshed design which covers stakeholder needs in line with updated Fedora Mission and strategy.
  3. Unified site for end users and for people looking to get more involved with the project.
  4. Keep use-case focused paths clear for first-time users, while also making all the options Fedora has to offer easily discoverable and available.
  5. Easy to maintain and keep updated. (It’d be nice to for individual teams to own and maintain content related to what they work on.)

Notes

The current https://getfedora.org page was launched in December 2014, six and a half years ago. The main URL https://fedoraproject.org/ is just a redirect, because the companion site meant to go there never reached implementation. This site was based on the Fedora.next initiative, and has not been updated to match the new Mission, Vision, or guiding policy. It’s time for an update!

The recent Fedora Council meeting with Fedora Server had a lot in common with issues recently discussed by the Fedora Workstation Working Group. Both teams are struggling with where to direct people as the “home” of the project. The Get Fedora page is too focused on downloads only, the wiki is too chaotic, the docs site too much like walking into a big uncatalogued library, and taiga and pagure don’t present enough of a “landing”, especially for new folks.

When first designed, the plan was for Get Fedora to be specifically (and only!) an end-user brochure, with Fedora Hubs – and awesome new active tool for contributor engagement – to be at https://fedoraproject.org/ (the current top-level redirect there was meant to just be a temporary thing). Without depressing ourselves too much, let’s just say… that didn’t work out. Without that, the redirect just stayed forever, and people have started thinking of Get Fedora as the main Fedora Project page, when really the main page just… doesn’t exist. (The front page of the wiki is the closest we have, but it’s only one or two clicks from there to vast wastelands of incorrect and outdated wiki pages.)

The new design should be based at https://fedoraproject.org, and https://getfedora.org reversed to point either at that URL or a new download-focused subset. (Like https://getfirefox.com today.) (This will help with our search engine juice as well.)

Additionally, when we did that brochure page, we intentially focused on the three Editions, which were meant to be an easy way to steer users to the right thing by use case. This plan was absolutely a success, as we can see significant Fedora growth after it was implemented. We had hoped for other Fedora Spins and Labs to thrive through separate promotion, but that really hasn’t worked out to everyone’s satisfaction. (And we decided in 2018 to do away with the Spins/Labs distinction, as people found it more confusing than helpful… but the sites still remain separate.)

We’d like a new design which ties this all together, serving as a home for our various Linux OS offerings and as a center for both users and contributors. It should make it easy to just “get Fedora”, but also easy to find out more about the project, our various teams, and how to get involved. And it should be exciting, inspiring, and welcoming.

Stakeholders

  • Key Stakeholders:
    • Edition WGs (Workstation, Server, IoT, CoreOS)
    • Desktop Spin Teams (KDE, Cloud, Xfce, Cinnamon, Silverblue, etc.)
    • Other Spin/Lab teams (Python, Design, Astronomy, etc.)
    • Alt Arch Teams (Aarch64, Power, S390x)
    • Fedora Council
  • Other Stakeholders:
    • Docs
    • Ambassadors
    • Marketing (sigh)
    • Outreach Revamp
    • Join Fedora
    • Fedora Linux users
    • Others?
  • Implementors:
    • Design Team
    • Websites & Apps Team
    • Infrastructure

Rough Plan

  1. Initial small stakeholder meetings (Council, Design, and Web & Apps team representatives plus representatives from the key stakeholder groups.)
  2. Initial broad design + larger stakeholder information gathering + interate on general design
  3. Exploration of specific designs, feedback from stakeholders, testing of ideas.
  4. Select and implement final design
  5. Launch

Timeline

It’d be super-exciting to have this in place for the Fedora Linux 35 launch in October. But we could also plan for F36 in April 2022 – or launch out of sync with the release schedule.

Next Steps

  1. Discuss with Council (and everyone here!), refine this.
  2. Take to Mindshare, Design, and Websites and Apps Teams.
  3. Get someone from the new Program Management Team to help organize and coordinate this.
  4. Organize stakeholder meetings.
  5. Follow rough schedule above.
15 Likes
2

I totally agree with this! Very exciting. Can’t wait to see the changes while being part of it. This is going to be a great project, especially for the Websites and Apps team.

5 Likes
3

I’ll note that we’re not the only ones with this problem. openSUSE also identified that they have “too many front doors”. It’s definitely a harder problem than it might seem superficially. I agree that now is as good a time as any to tackle it.

I like having the downloads stay on getfedora with a more “welcome to the project” page on fp.o. But honestly, either option would be a big improvement!

The F35 launch feels impossibly optimistic to me, especially given the amount of research and design that will need to happen before we start implementing it. If we also have pieces that CPE will need to do, getting it into their Q3 plans gives us less than a month to have it well-defined enough to have them start working on it.

F36 Beta (mid-March 2022) feels like a reasonable deadline to aim for. If it’s ready by then we can soft-launch it with promotion happening concurrent with the F36 GA. That also gives us some room to slip to the GA. Of course, I may be too pessimistic here (arguably, that’s my job).

My biggest concern with all of this is that it’s sustainable. When the team who puts it all together goes off to do other neat things, it needs to be easy for the rest of the community to make changes.

4 Likes
4

It would be nice if we can condense or do some work with the too many downloads sites we have:

5 Likes
5

Awesome initiative. Should we define it as Fedora Objective? Or do you want to set it up outside of the Objectives framework?

For the downloads site I personally would love to have a mode condensed version of the site where i can get the idea of the entire system of Fedora deliverables (arches, spins, labs, editions, containers, rpm-ostree, amazon, flatpacks?..) on one page without clicking through several pages to reach the download link for each one. But I guess we can discuss details like that later.

3 Likes
6

I’ve had the thought before that the Fedora Project is so big that could use a “map” akin to the ones you see when you walk in the front door of a large shopping mall. I wonder if a Image Map would be appropreate for the Fedora Project front page?

1 Like
7

In openSUSE, @hellcp is working on consolidating to a single network presence starting from opensuse.org and giving an easy path to navigate to everything.

We should redirect getfedora.org to get.fedoraproject.org and consolidate our web presence under the main fedoraproject.org site banner.

In fact, that’s how openSUSE is doing it with get.opensuse.org.

5 Likes
8

You know, I hadn’t even thought of it that way. But we do have space for more Objectives, so… maybe!

1 Like
9

It’s great to hear about this initiative, and I’m definitely interested in it. Also as someone who has raised some of the issues around project navigation, it’s great to hear that I’m not alone in trying to figure this out!

I do have some uncertainty around combining the end user and project participation functions of the website though, for a few reasons:

First, these are different functions with different audiences, and there’s a danger of creating confusion on all sides if it isn’t clear who the site is for and what it’s doing.

Second, given the challenges around keeping the site up to date, and maintaining an active team around it, I’d be tempted to keep it simple and focus on what an MVP would look like, rather than trying to integrate multiple functions into a relatively complex site.

Third, following on, we shouldn’t assume that the new version of the site will ever be “done”. We’re going to have to go through multiple versions before it’s good, and we should aim to make sure that the design is regularly updated. The more complex the site is, the harder it will be to do that.

Fourth, I actually think that the current focus on directing visitors to the editions is the correct approach, and ideally we’d be focusing our own efforts on funnelling visitors and converting them into Fedora users.

That’s not to say that we shouldn’t work to improve project navigation - we absolutely should, and that’s something I myself want to work on. But maybe that would be better as a separate project?

4 Likes
10

@ngompa @mattdm
I can agree that the main Fedora website needs to be updated.

The website of opensuse.org is really great and easy to handle IMO. Maybe we can look and get some inspirations how this layout and design might be a way for getfedora.org

v/r
Andi

1 Like
11

This also helps with SEO if one cares about such things… that way your domains arent working against each other for ranking.

6 Likes
12

I don’t know, but I wonder if using paths like fedoraproject.org/spins, fedoraproject.org/labs, fedoraproject.org/workstation, fedoraproject.org/i3, etc. would help the search engines identify fedoraproject.org as the main landing page for all the sub projects?

1 Like
13

Yes – I think that’s at least a lesson we learned with Hubs. And we don’t necessarily need to have a unified view for end-users and potential (and active) contributors – but I would like a design which considers it all.

4 Likes
14

And see: Website refresh next steps!

5 Likes
15

That’s why someone needs to tell the Website team that rebuilding everything in JS/Webpack/React[or]Angular is not going to be sustainable.

Having one system in place (Styleguide for web components kind of thing) can help to decide on how to build all of the sites consistent with one look and feel.
Should I open the issue on this, maybe there allready is one?
The one that I know works with html is https://fractal.build/ but if Websites are going to be build with html-in-js there are countless other.

16

Hi @ojn,

Rolling back into the discussion that we had on our #fedora-websites IRC channel, as much as it would be great for you to open up a ticket under our Pagure namespace, I would suggest you join us in our weekly meet, which happens every Tuesday to propose this, first-hand.

We have had a conversation on the probable frontend JS frameworks to make use of in our last meeting, but that mostly depends on the implications that we get from the design team (for the most parts). Simply put, if the application is simple, we might just get away with not using one at all but one of the key challenges that we face is with doing away the non-uniform look and feel of all the apps that we maintain.

2 Likes
17

Definitely, for a current example look at ask.fp.o and discussion.fp.o

1 Like
18

It isn’t necessarily a bad thing to have contributors and the rest of the community starting off at the same point. The current page offers the Magazine as fore material, with periphery links, and it is showing it’s age when you look around at what’s new. But it starts us all in the same place irrespective of personal ID. I would like that start page, which I usually use as my browser’s home, to be even simpler in appearance than it is now.
Related, but on a detail, where does the Magazine fall in as stakeholder, I assume it is.

1 Like
19

As someone that was involved with the 2019 redo of getfedora.org, it is a little disappointing that this effort is labelled a “minor refresh”.

While not a major rethink of all the Fedora Websites, the amount of work and effort in 2019 that went into implementing @duffy 's new design, and work on the generation backend to make it easier to update and maintain – being labelled as a minor refresh a year or so later is a little bit sad for the soul TBH.

6 Likes
20

Sorry Ryan, I didn’t mean to diminish the effort that went into that. I know under the hood it’s very different and I appreciate all of the work that people took to get it there. I really just meant that it was following the same concepts, and not that major rethink.

3 Likes