Plone Lightning talks Friday
Lightning talks at the Plone Conference 2014 in Bristol.
Add-on designed to offload large data to cloud storage. Transcoding of videos to web-compatible format. Control panel to configure things. You can also stream it.
We will make a release of that soon. For now, see https://github.com/netsight/netsight.cloudstorage
Oud customer requested highest available system, lots of requests, large requests, more than 384 MBps. CDN: Content Delivery Network. Using CloudFront. nginx, varnish, Plone 1-3. With some configuration nginx will first get the content normally from Plone, later requests will be handled by the CDN. Using this for various universities, including single sign-on for students, some for local government.
Please join Plone Symposium Tokyo group.
We check Jenkins, making sure all tests pass, check for new packages, make sure changelog are updated, check crazy commits, update versions.cfg, check p.a.upgrade, ask for locales, update metadata.xml, make pending release, unified change log, notify installers team, probably forgetting a few.
You make some progress releasing, then someone does more changes, hard to keep up and keep it stable. Ticket 13283: merge several packages into one, may help.
Thank you to Zest for creating zest.releaser, keeping me sane.
Studies on security. Take at least fifteen minutes per day per system to check for updates. Drupal had some problems recently, unless you updated your system within seven hours of releasing the hotfix.
Check these packages:
- plone.hud, plone.app.hud
- Check for vulnerability checks: plone.vulnerabilitychecks
Keeping a green build since 2011. I am the guy who shouts at you when you break the plone testing build, that is why I now sound like I sound.
Continuous integration rules: http://buildoutcoredev.readthedocs.org/en/latest/continous-integration.html
We did some sprinting this year. I went to the Jenkins User Conference Europe in June 2014, presenting how Plone uses it.
CI infrastructure. Sven did a really great job of creating an infrastructure with nodes, multiple builds and test jobs. Ramon created mr.roboto to report results back. Asko improved acceptance testing, with robotframework. Testing cleanup, moving from PloneTestCase to plone.app.testing.
Future plans: maybe Docker setup.
Team page: http://plone.org/community/teams/cit
Have you ever not finished what you are sprinting on?
Add info on your sprint topic: titel, what, who, code status. Update it at the end of the day.
Another look at how you structure information on your site.
Too much hierarchy is deadly for customer centric. Plone has a hierarchical folder structure, that always shines through, even when you add relational items or tags. Conflict: chaos or relational.
We want to use our hierarchical Plone hammer and see nails everywhere. Designer had other ideas, maybe because of Drupal background. They wanted 25 different content types, just so they could have relations between themes, projects, etcetera. Oh, there was no Folderish type.
But couldn't we use mostly simple Folders and Pages? The functionality is not actually different, just the label.
Landing pages around themes. collective.cover. Theme page has links to reports and other stuff.
Two part labeling: categories and themes. collective.classifiers dexterity behavior.
You can search in collections on these classifications: give me all 'themes' on category 'water'.
Similar solution may be collective.taxonomy.
This winter we will at last again have a sprint in Austria, in Innsbruck, capitol of Tirol. Lots of (olympic) ski areas. Very nice view from high restaurant that we will be visiting. Your family is welcome too. Guided tour in quantum physics institute.
Sprinting on probably Plone 5. Stabilize what we have, for doing daily business with Plone without headaches but with fun. plone.app.multilingual needs some love. You are invited to work and have some fun and maybe stay a few days longer and enjoy Innsbruck.
Find us on Coactivate.
ADFS: Active Directory. Single Sign-on.
Using dm.zope.saml2, extending it with some functionality.
Purpose: link data from multiple sources. Fast search for semantic data (rdf). Index data from an endpoint (Virtuoso) into Elasticsearch.
You need an RDF marshaller for the Plone website. Via Virtuoso. RDF River plugin for Elasticsearch.
But you can use the plugin to get the rdf from Plone directly into Elasticsearch, without Virtuoso in the middle.
Mosaic is a rendering engine for Plone. Flexible custom layout of the page. You can make prototype layouts that editors can use.
We connect teachers with entrepreneurs. Get them to speak to school children. See http://Founders4Schools.org.uk
We have an Appathon:
- Think of an app. Work in teams to think of an idea for an app.
- Create an app prototype. Go into universities to get students with code skills.
- Winners announced.
Guest house is in Mountain View. Companies like Google, Facebook, Tesla. Consolation prizes of Raspberry Pis.
Please host an Appathon yourself, very nice. Contact us for ideas and support.
We have it running. You need a credit card currently, but we will work on that.
See code here:
In your own buildout you need to change a few things, enable RelStorage, all documented on the above repository. Our company website is running for a couple of months on Heroku.