The Plone Conference account tweeted that a State of Plone talk would be awesome and that the Plone community missed me. I miss the Plone community too, so I am here.

I will state it clearly: Pillow would not exist if not for Plone.

In July 2010 I announced Pillow as "friendly" fork of PIL. The mailing thread and future answers are interesting to read.

Some history:

• 1991: Python 0.9.1
• 1995: PIL started
• 1998: Zope
• 1999: Zope2
• 2000: Python 2.0 with distutils
• 2001: Plone
• 2005: Buildout
• 2006: I attended my first Plone Conference, in Washington
• 2006: setuptools was born

PIL had an issue, or Plone had an issue with PIL:

• PIL used distutils.
• Plone 3.2 used Buildout and setuptools
• PIL was not installable in Buildout and setuptools
• Specific problem: import Image could mean the Image module from PIL, or the Image module from Zope.

Various ways of repackaging PIL started, for example PILwoTk. You can still find various PIL derivatives at https://dist.plone.org/thirdparty/

PIL 1.1.6 from 2006 is still the last version on PyPI. I got maintainership of this page this year, actually. Pillow 1.0 is basically the same, except that it uses setuptools. This worked in buildout. I was happy.

Couple years, nothing really interesting happened. But some contributors came along. Pillow 2.0.0 in March 2013 had Python 3 support.

An important milestone in 2015: we added release notes.

Release schedule: in the beginning of every quarter.

We get some money from Tidelift for maintenance.

I want to talk about some categorisation and classification options in Plone, next to the folder structure.

Faceted navigation: drill down on 'facets' when you search for items. It was popularized by online shopping. Facets in Plone for developers is: whatever is in the ZCatalog, and for users: what you can search on in Collections. Gold standard is eea.facetednavigation, developed for the European Environment Agency. Examples: EEA, and on two sites by Zest: Vaquums and Minaraad, where it replaces the standard search.

collective.collectionfilter is a much leaner, meaner, but also more limited version of faceted navigation. Demo with standard Plone News Items with some tags (also known as categories, also known as Subject). Add a Collection that filters on News Items. Now add collection filter portlets.

eea.facetednavigation takes over your complete page. In an action you enable or disable it.

Now a demo of collectionfilter in SGBP, a documentation website for water management planning in Belgium/Flanders. The customer wanted to take some graphs and compare them. We did that with collectionfilter and collective.classifiers. With the last one we added structured categories: one for water basins and one for parameters of the graphs. Now we use collectionfilter to query a parameter and show the graphs for all water basins.

You can adapt several things in the collectionfilter UI, for example change how search options are displayed. This is documented, but took me a while to get right.

Collectionfilter also works with Mosaic, because the portlets are also mapped to tiles.

Here are the slides.

I am a software architect at University of Jyväskylä. I have been using Plone since 2004 and GatsbyJS since 2018. The university wanted one student information management system to rule them all, but... every organisation shall do their own integrations, using granular REST API with deep JSON responses. And there should be branded study guides, which we crafted with GatsbyJS. But this was not enough for the Open University part. They really needed a CMS.

We use Plone 5.2, Volto, GatsbyJS, and have 6000 html pages, times two languages, out of which 760 are Volto pages. With Plone we could extend content types without needing to do any coding, in the content types field editor. In volto we added auto-complete widgets with custom vocabularies. On the GatsbyJs side, we query the connected pages with GraphQL. We render Volto layouts with React components, rendering individual blocks.

Why did we choose GatsbyJs? It is a ReactJS-based site generator. Being static, it is very fast. You can use multiple sources as input, using a plugin architecture. Data lookup is done with GraphQL. It is easy to get started, with comprehensive documentation.

I mentored two Google Summer of Code projects for the gatsby-source-plone plugin. It supports default types and most TTW types, also Volto blocks. You can do incremental updates by modification date, so it is really fast.

Not everything is easy. The full "GatsbyJs experience" requires practice. You want to replace inline images and links with GatsbyJs images and links, replace file links with direct downloads.

Using @plone/volto as dependency to render blocks seemed like a good idea, but it required webpack overrides to be impartable, and could not be used for images and links.

The ugly parts of GatsbyJs:

• The GraphQL source plugin cannot cache.
• The build may take hours, and gigabytes of memory.
• The build result in readonly.
• For me it is hard to follow GatsbyJs development, especially individual plugins, because they use a monorepo.

Editors can work on the site during the day, and then wee rebuild the result during the night

Here are the slides.

With some other people I have created pendect.com. The idea was to build a TL;DR (Too Long, Did not Read) platform. Create short cards, linking to long articles. Cards are submitted by our community (Pentributors). Also: for every card, we plant a tree.

Technical requirements:

• Collaborative workflow
• Permission control, who can do what, where and when
• Metadata, categorization
• We are a startup, so we need SEO.
• Open source

We decided for Plone as Content Management System. It has most features out of the box. I know Plone of course. But I considered building a simple API on top of Pyramid. But with Plone we could go to market immediately. It has a proven record with news portals, and also a friendly community with smart people.

We use DBpedia for Metadata and categorization. This works on a Wikipedia dataset. They have a tool Spotlight to detect entities in texts. We use a Sparql connection to DBpedia to query for more information.

We use a bit of everything else: Cloudflare, nginx, varnish, HAProxy, Ansible, Thumbor, Sentry, Mailhun, IFTTT, Zapier, Gravatar. We already survived twelve different surges in traffic from Reddit. We have tens of thousands of users. We use Thumbor for image scales. Actually easy to integrate in Plone.

Our toolset and add-ons:

• We have development speed thanks to some friends, like Python 3 with F-strings, Type hints (I put them in from the start, helping me to think about my code), data classes (better than returning dictionaries everywhere). Also: black, isort, flake8, PyCharm.
• We do not use many Plone add-ons: collective.z3cform.datagridfield for table-like content, collective.sentry to talk with Sentry, contentrules.slack to send messages to Slack, souper.plone to store many small data records/

Content types:

• Default folders, documents, images, collections
• Category: collection behavior with sub-objects
• Card: similar to news item, but more categories

Adapting Plone:

• Browser views: dashboard became "My Feed", Author page became the profile
• Some content rules

New features:

• Aggregator pages for tags, people, locations, organizations.
• Development with souper.plone: you can follow categories, tags, people, locations, organizations, and Pentributors

Sync to archive.org. Takes too long, so this needed to be asynchronous. Also other external services, like translating.

Lessons learned:

• Always create upgrade steps. Be aware of existing registry configurations: use purge=false.
• Plone training materials are the de factory documentation for Plone.
• I hate the current resource registries.
• Plone lacks a simple and working async/delayed solution.
• Webp images are not for all. Be careful with caching them.

Future:

• Translation and auto summary to all Pentributors.
• Card threads, similar to Twitter threads, using content relations.
• Use ElasticSearch.
• Move to RelStorage.
• Plone as headless CMS, with Volto and mobile applications
• Move away from the default user folder, it is getting slower already.