search

plone / plone.app.mosaic

Plone Mosaic main repository
https://pypi.python.org/pypi/plone.app.mosaic
GNU General Public License v2.0
35 stars 26 forks source link

PLIP plone.app.mosaic for Plone 5.x #212

Closed datakurre closed 8 years ago

datakurre commented 8 years ago

At Boston PloneConf it should be clear if, when and how Mosaic could be bundled with some future Plone 5 feature release. Therefore Mosaic should be PLIP'd ASAP, Mephisto Sprint 2016 being the last possible time and place to do that.

jensens commented 8 years ago

I would separate this in several better consumable parts:

  1. get plone.tiles, plone.app.tiles, plone.app.blocks and plone.app.standardtiles into core - this can happen in the 5.x series. Those packages are stable. Not sure about plone.app.drafts (I think it can go with mosaic).
  2. use plone.app.blocks with tiles and layouts as default mechanism to build pages - this is a Plone 6.0 thing and need to be introduced carefully.
  3. get plone.app.mosaic into core - as an core addon this can happen in the 6.x series
datakurre commented 8 years ago

Good thoughts. Make sense from stability point of view. But part 1. would be confusing alone, because just "installing the tiles framework" would do nothing. Part 2. would be a lot of work, beause it implies moving portlets and viewlets into tiles, and main template into site layout.

jensens commented 8 years ago

To get (1) into a state were it makes sense: I think what is missing are the site-layouts of mosaic. We're using parts of mosaic in a project at the moment, and its a great advantage. What we're not using is the site-composer (aka mosaic editor). But we're using fixed content layouts a lot and it makes building pages so much easier. It works together with behaviors like a charm.

Part two is in fact the third missing part of what was initially in the 3D (Dexterity, Diazo, Deco). Thats huge, but it gives us the chance to redefine how definition of UI in Plone works, adds maximum flexibility for integrators and simplifies the whole ecosystem a lot. We would not need viewlets anymore, because then everything is a tile. Portlets (which are in facts a persistent inherited viewlets) should be replaced too (how in details needs some discussion, not because its difficult, but because there are some options).

Mosaic as "the site composer" (or site-builder) for content - and in the long run for the whole site - can be evolved in parallel and consume all those upstream developments.

datakurre commented 8 years ago

:+1:

On 22. kesäkuuta 2016 klo 11.12 +0300, Jens W. Kleinnotifications@github.com, wrote:

To get (1) into a state were it makes sense: I think what is missing are the site-layouts of mosaic. We're using parts of mosaic in a project at the moment, and its a great advantage. What we're not using is the site-composer (aka mosaic editor). But we're using fixed content layouts a lot and it makes building pages so much easier. It works together with behaviors like a charm.

Part two is in fact the third missing part of what was initially the third D (Dexterity, Diazo, Deco). Thats huge, but it gives us the chance to redefine how definition of UI in Plone works, adds maximum flexibility for integrators and simplifies the whole ecosystem a lot. We would not need viewlets anymore, because then everything is a tile. Portlets (which are in facts a persistent inherited viewlets) should be replaced too (how in details needs some discussion, not because its difficult, but because there are some options).

Mosaic as "the site composer" (or site-builder) for content - and in the long run for the whole site - can be evolved in parallel and consume all those upstream developments.

— You are receiving this because you authored the thread. Reply to this email directly,view it on GitHub(https://github.com/plone/plone.app.mosaic/issues/212#issuecomment-227673437), ormute the thread(https://github.com/notifications/unsubscribe/AAJyv02N1k3WpWoY5DqIYJ71n5vvePkfks5qOO5pgaJpZM4IyU95).

djay commented 8 years ago

In terms of the user experience a user would be electing to use mosaic on a given Page? or it would be the only option for a Page? or would it go further and replace Folders and Collections as well?

jensens commented 8 years ago

@djay I think the details need to be defined in the upcoming discussions.

djay commented 8 years ago

@jensens what upcoming discussions?

djay commented 8 years ago

btw, the concept of merging page, collection and folders using tiles is whats analysed in this ticket. https://github.com/plone/Products.CMFPlone/issues/519

jensens commented 8 years ago

@djay well, since this is a huge thing I expect that there are some ;) and they just started. Memphisto sprint also addresses this and it would be good to collect needs and user stories before this.

What I (personally) expect at the end of the process is a Plone where our current views are replaced by (static) layouts. I would not like to see everything a composeable page (dynamic layouts), because for at least most of our customers this is not what we need.

I think merging page+collection can work if a collection is a dynamic tile of a composed page (mosaic).

Overall folderish types is imo a good idea anyway and there is already a PLIP which needs some energy.

I see that we need an overall concept on the UX and the way we create content. But I'd not try to implement it all in one big gulp. We need to define consumable steps, release them early and then move forward for the bigger goal.

djay commented 8 years ago

ok. Just checking whats in scope by this proposal. It seems as if the above proposal would likely mean it's at most available as a displayview option on the default Page type and not go any further. That would at least make it possible for someone to configure their site to have a single content type. but they would still be missing functionality unless we fix the additional issues highlighted in https://github.com/plone/Products.CMFPlone/issues/519

hvelarde commented 8 years ago

just thinking: how does this fits the split of Plone into two separate components in the future: Plone server and Plone client?

jensens commented 8 years ago

@hvelarde well, thats at the moment a rough idea. At least using tiles everywhere would make that much easier.

vangheem commented 8 years ago

For those interested, we worked on a roadmap: https://github.com/plone/plone.app.mosaic/wiki

I'm going to close this ticket in favor of the roadmap and the assumption this will not be plip'd for 5 but some future version.

hvelarde commented 8 years ago

that was awesome! now I can say we have learned a lot from past mistakes; thank you guys for planning things like that.