# Cells vs ViewComponents

As you know we're in the process of [RedesignDecidim](https://meta.decidim.org/processes/RedesignDecidim), and as such is a perfect time to rethink about the current dependencies that we have.

Initially, we were thinking about migrating from our current view component gem ([Cells](https://github.com/trailblazer/cells)) to the shiny new gem that's getting lots of traction ([ViewComponent](https://github.com/github/view_component)).

There are some good things about both solutions, and before making a final decision if we should make the change, we wanted to hear more about the developers community about what they think about these solutions.

## Cells

* We already have lots of cells implemented - 238 only in the main repository
* First release in 2010 - so it's mature
* It's still maintained (last commit is from 10 days ago), but it seems like the current development could has stagnated by some metrics (latest release in 2017). This could be a good thing (stability) or a bad thing (no more new features nor innovation).

See more metadata at [RubyToolbox](https://www.ruby-toolbox.com/projects/cells)

## ViewComponent

* Is the new solution, with a strong organization behind (GitHub)
* It has more marketing (like it even has a their own website for the gem - https://viewcomponent.org/).
* It's more present in the Ruby community hivemind, for instance, it appears more in things like [Podcasts](https://rubyrogues.com/search?utf8=%E2%9C%93&term=view+components)
* As an example of traction/momentum, we see things like Storybook integration (see #7253). There are already [projects with traction](https://github.com/jonspalmer/view_component_storybook) that without too much effort would allow us to use Storybook and its ecosystem. In the case of Cells, I guess we could start something based on [this other project](https://github.com/danieldpence/storybook_rails) but that would be more effort. This is just an example of what will probably happen more with time (a bigger ecosystem).

See more metadata at [RubyToolbox](https://www.ruby-toolbox.com/projects/view_component)

---

As background information, we've already started a discussion in other places (see #7234). The idea of this new discussion is to have a place to only talk about this subject.

A good summary is what @mrcasals mentioned in that thread:

> Cells was the only solution back at the time. Cells is not discontinued, but it's true that GitHub's ViewComponent will have more traction, more community and probably better support than Cells has. So +1 to this! I suggest we first move all cells to ViewComponents, and then keep improving the layouts by adding more ViewComponents directly.

The "more traction, more community" is visible in this graph of a popularity metric (GitHub stars), but I think it'd be extrapolated to others too:


https://star-history.com/#trailblazer/cells&github/view_component&Date

Even though it's not at the same level, it's growing faster.

So, the questions would be: Is ViewComponent ready and mature? Should we migrate? The benefits outweigh the costs of doing so?