Rewriting Pages CMS with React/Next.js?

[hunvreus]

I want Pages CMS to be an obvious choice for developers when they're building modern websites or apps.

Whether you're building a website with 11ty or Hugo, or an app with Next.js or Astro, Pages CMS should be a no-brainer if your project is hosted on GitHub (and maybe even GitLab, Gitea...).

As a developer, it should be easy to get started by dropping a single file in your repo.

As you grow, you should be able to easily bring in non-technical contributors (e.g. invite by email, commenting, real-time collaboration) and leverage more advanced features (e.g. search API, GraphQL API, recommendation engine...)

I think I got to a reasonable place, validating there is at least some demand, but the backlog is pretty long. Lots to build.

And part of what kept me busy the past few months was that I had to learn coding again after 7+ years or not really building anything serious on my own (that hasn't been part of my day job).

I picked what I remembered was easy to get things done: Vue.js.

Here is the issue: rewrites are mostly a waste of time, BUT, in this case, the current stack may hurt adoption and impact:

• My early adopters AND contributors, front-end developers, have overwhelmingly picked React and Next.js as the standard.
• shadcn/ui is very close (in style) to what I've built myself. And UI is a freaking rabbit hole, especially if you want decent accessibility.
• A lot of what I used to like about Vue (batteries included, simplicity of the API, "get things done" approach, ...) is either addressed by Next.js or simply not holding true anymore (Vue 3 + Composition API).
• I now need more than a few serverless functions to handle advanced features (e.g. invite by email). I've started prototyping with Hono, but wouldn't it make more sense to pick the current favorite for front-end devs (i.e. Next.js)?

I've been muling over this for a few weeks now and really didn't want to do it. But I think I'm going to take the next few weeks and see if I can get Pages CMS ported to React/Next.js.

To those of you out there who tried or use Pages CMS: do you care? Do you disagree with this analysis?

[leejoramo-d51]

First, I need to say that I don't deeply know your project. I have been watching it for the past month or two when it came up on my regular searches for this type of CMS. I had been using Forestry, which stopped service last year (I didn't like Tina, the follow up project). I have experience going back over 20 years with various Static Renders and full CMS's such as Drupal and Phone. More recently I have used Hugo on the back end.

I have been very pleased with the development of Pages CMS, and I am VERY likely to use it for my next site or migration. So far I have only played with it in my local development environment.

I my experience, rewrites are almost always no-win situations. At best, you fix some technical debt that was causing significant problems, or allow some critical new features that could not reasonable done.

In the case of Open Source projects, you risk burning all of the time and good will of your community. Your current users have gotten you to this point, and they will be forced to start over too. This is especially true for your most advanced users.

I would ask some of these questions:

• Are your current users asking for this change? Either directly or via the features they need.
• Will the rewrite pay for itself. Do you have paying clients that want this and are willing to fund the change?
• Do you consider the current state of the project to be an early Alpha? You have made not promises, and features and API's are not guaranteed.
• Will this rewrite cause the project to loose momentum, or is this a change that will in a short amount of time speed up development.

Personally, as a developer, my tools of choice these days are SvelteKit with both NodeJS/Express and Python on the backend. I would be willing to work on Vue, but I really don't like React.

At the same time, I am looking to Pages CMS to be mostly a drop in Application for my users with mostly configuration on my part and little in the way of development programming. So I may not really care about your tool stack.

I will end with one more question:

• Does the idea of this change, make you as the owner of an Open Source project feel energized and excited?

[rohannelson]

I'm only an entry level developer so I don't have heaps of experience to speak from, but...

I do think the popularity of React would mean more people would be able to contribute - I don't imagine heaps of people will be going out of their way to learn Vue these days, although I could be wrong. I know I have thought that it would be cool to contribute to this project but I don't know Vue so I haven't.

My main hesitation with Next.js is what I've heard about a level of vendor lock-in with certain features only really being supported by Vercel. Remix could be an alternative React framework, although I haven't worked with it much.

Another alternative could be using Astro as SSR for a React app? I've used Astro a bunch and it's great (including with React and haven't had any problems there).

I like the idea of Svelte and Sveltekit but I haven't worked with them so can't really speak to that.

As for rebuild or not - I guess in some ways if you're going to do it, better to do it now than later. The longer you leave it the greater the cost. I like the idea of being able to contribute so I'd vote in favour of a rebuild if it was in React.

Anyway, that's my two cents. Love the app so far.

[rohannelson]

@hunvreus - any further thoughts on this one?

[hunvreus]

I've been traveling a lot in the past few weeks, but I did find the time to learn Next.js.

I've just started working on the Next.js version of Pages CMS. Lots of changes obviously, but I'm able to recycle a lot of my code ,although I still need a fair amount of adjustments and changes (e.g. move to shadcn/ui, convert to Typescript etc).

I think we'll be better off in the long run: easier to get contributors, better security, etc. This will be very helpful once I release the Pro features.

[olegfedak]

Have you considered Vue/Nuxt as an analog of React/Next? Community of Nuxt is pretty active.

[hunvreus]

The current version of Pages CMS was built with Vue and I considered Nuxt. But React/Next.js just has more market share and I'd love to get as many contributors as possible.

Nuxt looks solid though.

[olegfedak]

I found Pages CMS because I was looking for a git-based CMS written with Vue :D

Well, have a great development!

[jppope]

Reading through this discussion and wanted to add in some considerations. I know I'm late to the party.

First to your comment:

"My early adopters AND contributors, front-end developers, have overwhelmingly picked React and Next.js as the standard."

The software industry has more or less picked react/next as a "default" but antidotally I find that is more reflective of naivety than it is preference (from personal observations with hundreds of devs in a large meetup, and numerous larger teams at work) . Most of the people advocating for React/Next over other frontend frameworks have little to no experience with the other frameworks and focus on React/Next because its what they know.

When you do dig in to the preferences of experienced developers/engineers who have informed preferences with frontend frameworks, they will pick React mostly for other reasons such as resource accessibility (react querybuilder, prebuilt components, etc) or hiring practices (which makes a strong case for business reasons... if that is part of your goal it should be a consideration). However, in absence of these very surface level arguments, I think its more common to find experienced developers working with different frameworks such as Vue, Svelte, and others. Personally, I use React at work only, and find that a lot of novices end up with tons of poorly factored code because they don't understand the framework.

So the trade off you are potentially opening yourself up to is really that you can increase adoption but likely lose out on quality contributors, or use a framework with lower mass adoption as a filtering mechanism for who you want to contribute in your community. I'm not advocating for a choice, just making sure you don't want to be dealing with novice pull requests for the rest of this awesome project. I would hate to see it disappear from either lack of adoption OR equally adoption from the wrong people.

Now... I think another consideration that probably needs to be said... the React ecosystem has been veering pretty heavily towards companies with a vested interest in locking users in and keeping them there ("Framework lock-in"). Nextjs specifically has been trying to move everyone to their platform through feature development... Theres a bunch of jokes about this on the internet... including my favorite Hitler Meme: https://www.reddit.com/r/ThePrimeagenReact/comments/1abe5qv/nextjs_14_hitler_edition/ ... But the overarching issue is that you will likely have to make decisions about YOUR project that you don't want to make because Nextjs is trying to push you a certain direction.

Thus far my experience with pages-cms has been awesome, and I don't think the choice of framework really matters at all... the model is simple and very configurable without having to mess around with code... which is AMAZING BTW

In my opinion, the fact that I generally don't have to worry about frameworks, code, or other stuff is what makes pages-cms the best... I CAN do things if I want to, but I don't have to. I don't think framework choice should really impact a project if it is doing things that are meaningful to the community, but I do really get worried about this project ending up as another react plugin where the maintainer gets fed up with the community and drops it.

Also I am happy to throw some money at this project, we are already using it modestly for client work and it has been a dream.

Looking forward to seeing where this heads. cheers!

[meduzen]

Also late to the party since I discovered Pages CMS very recently and I have not started using yet, and I could have written mostly the exact same lines as @jppope did (but not with as much clarity).

It’s your project, so any possible choice is up to you. I wish you to not regret it, and I also wish you in the future to enjoy whatever stack you will discover. Also congrats for being mostly done with the rewrite.

For anyone getting there, I can only encourage you to discover the web platform through tech stacks actually trying to embrace it (like Vue, Svelte, or by using plain JavaScript). You'll find some interesting articles such as A History of React Criticism, The deskilling of web dev is harming the product…, The self-fulfilling prophecy of React, Things you forgot (or never knew) because of React or The Market for Lemons.

[hunvreus]

@jppope Thanks for the thoughtful comment.

I'm almost done with the rewrite.

I wasn't excited with it at first, but I can now say I think it was the right thing to do.

The next version will have much better security, more robust UI and a better foundation for advanced features (e.g. invite by email, granular permissions, 3rd party storage, ...).

This also was (is) the opportunity for me to clean up a lot of my code; I've been back into programming for less than 6 months and the learning curve was real...

For end users, nothing will change much with this upcoming version:

• Configuration remains the same, no need for you to change anything.
• It will still work for ANY framework that uses flat files: Astro, Jekyll, Next, Nuxt, Vue, React, Svelte, Hugo, 11ty, ...
• The user interface is pretty similar with only a few adjustments.
• You can still use the online version for free.
• You can still deploy on your own using Cloudflare, although you'll need to create a D1 database (https://developers.cloudflare.com/d1/). I will likely add more options later on (Vercel, Netlify, Supabase, Turso, ...).

The only annoyance is that when I upgrade, you will need to log in again. That's it.

If any one of you is interested in testing this new version, reach out to me and I'll send you an email to beta test in the next coupe of weeks: hey@ronanberder.com

[jppope]

sure happy to help out where I can, really impressed with this project and I would like to support it

[hunvreus]

Alright, the Next.js version is up and running. I still have a list of small bugs and improvements I'd like to make, but here you go:

• Demo: https://next.pagescms.org
• Code: https://github.com/pages-cms/pages-cms/tree/next

If some of you have a look, let me know if everything works with your setup.

[jppope]

hey there @hunvreus ... I was trying to look at the code and its just trying to redirect me to the demo... is that expected?

[hunvreus]

Yes, because I'm an idiot. I've corrected the link: https://github.com/pages-cms/pages-cms/tree/next