I like this idea to some extent. Kind of in the half way of the Drupal's way of thinking, although in Drupal they only ship with the very essentials. I don't like that either but somewhere in the middle would sound quite compelling, like what you're suggesting here.

If you go to a restaurant, do you want them to bring you the whole menu cooked in front of you where you only eat few dishes or do you want to pick yourself what you want? :)

Some themes do exactly that already.

For example, I purchased the Formigo F5 theme, and it came with sixteen blocks which would easily count as a dozen or so add-ons.

Not to sound like a salesman for Formigo, but it came with something like six different navigational elements, multiple page types, image slider, font tools, and allows buttons for color customization. And it's responsive....

Again, I am not affiliated with Formigo, I just bought their theme.



http://themes.formigo.co.uk/?theme=Foundation%205...

He's talking about the core.

I would expect it's a matter of time/effort and has to do with relationships with developers vs core team, but I don't know.

It's like when Microsoft includes a web browser with their OS, there are then questions about what impact that has on the development of other competing web browsers as well as how the quality of the web browser reflects, negatively or positively, on the functionality of the core product.

So if C5 included a core set of add-ons, then those become the de-facto standards, and since C5 is free, that takes away the profit motive for developers to create/sell great add-ons. If those add-ons are no good, then C5 is viewed more poorly, yet if those included add-ons are too good, then why buy any other add-ons?

Perhaps the solution would be to make the 'free C5 community version' the 'basic starter edition for $' and the 'enterprise CMS version for $$$'.

I don't think anyone has suggested adding more to the core than there already is.

How I understand Hissy's original suggestion is that some parts of the current core (5.7.3) would be taken away from the core libraries to avoid adding more complexity for developers. Then, the user could select during the installation (or later) which they want to install.

Referring to the point about the Internationalization support Hissy originally pointed out.

Yes, the thing I suggested is just hide some dashboard pages and block types from beginners like advanced permission mode.

I think this idea goes even beoynd that. E.g. the multilingual comes with its own classes, so those would be separated to the "bundled package" instead of the core's /src directory to avoid adding complexity there for developers.

Kind of the same idea why there used to be the "packages" directory in the core directory in 5.6 and earlier, although I don't know whether they actually used it somewhere.

I'd also like to see this kind of bundling happening through composer packages.