PLOG Friday morning talks

published Apr 10, 2015

Report-out and talks at Plone Open Garden Sorrento.

On the Trello board there is now a column for teams. Feel free to edit it. And add yourself to a team if you feel lonely and want to join the fun. :-)

Report-outs from yesterday:

  • Javascript / RESTful api. First needed is the RESTful api. Afterwards Javascript front-end. Talked about technical challenges for REST, like supporting http verbs in the ZPublisher without defaulting to webdav, Ramon actually has a working prototype for this, too good to be true. Roadmap can be found on https://github.com/plone/plone.restapi/milestones. It will only work for dexterity. How can we make it happen. We can discuss that here and online. Second session was on Javascript front-end. Brainstorming about what each Javascript framework actually does. If you compare things, they should be similar, which they really are not. We do not have to make a future decision now. We currently have patternslib, and mockup on top of that. Patternslib may help you to not have to write Javascript, but you should really learn Javascript anyway if you want to be a web developer. In half a year there will be even more Javascript frameworks, we can decide later. Sorry, there was no pillow fight. We have smart people who can implement a solution. We still have a solid, integrated stack. Outsiders who may want to build on top of Plone will not look deeply into out stack, at how difficult it may be: when is the last time you really dived into the SOLR code? You simply use it, without worrying too much, so we should not worry to much.
  • Plone Intranet. Reviewing where we are and how to move forward to ship with a sane architecture. We decided to organize four one-week sprints in the next ten weeks. One online, others in Bristol, Berlin, Arnhem.
  • Hackable Plone, Through The Web Plone (TTW). The need for hackability (tinkering with it as power user) was widely supported. Finishing up Mosaic is an important, maybe vital task. Through the web editable .po files, to change jargon. [Look at prototype in collective.storedtranslations, Maurits.] Editable labels per content type, even in the same dexterity behavior; would be really cool, but we have no idea if that is easy or difficult. Repeatable export and import of content and configuration, in the Plone UI please; per add-on exportable config would be great, but might be difficult. Making view, viewlets, theme snippets through the web; make that be seen as a normal supported practice, instead of being frowned upon. Through the web installable add-ons? collective.jbot allows through the web jbot. Some few but loud people are against jbot, but most seem to think it is a fine method for overriding. portal_view_customizations can be tricky, in some cases breaking things simply by customizing it without changes. Being able to do a similar and less breakable thing through the Plone UI would be good. Undo or better versioning of such changes is helpful.
  • Branding, positioning. Micro managing has not worked. Various regional and cultural differences that do not work. See Fred's talk of yesterday. We could work towards a common brand view. Get help from community and also from outside. We know someone who uses Plone and does Jungian Archetypes for a living, he was actually here at PLOG last year, so maybe we can ask him to help. plone.com discussions, a bit too much American, enterprise has different connotations. Whatever brand identity we have, does not necessarily influence how individual companies market themselves. Press releases on time can be hard with volunteers, we may explore other ways. Agreement among US companies to hire marketing people. Not a finished discussion. Board and current marketing team will pick this up, and be sure to ask others for their input.

Gil: Berlin Sprint

With Stefania we plan to organize a sprint in Berlin (not the one Guido mentioned above). Working on Mosaic. The coming weeks we will send a formal announcement. Middle of September this year.

Fred: Six Thinking Hats

See presentation at Trello.

This is about Six Thinking Hats, by Edward de Bono. Edward has thought a lot about thinking. Gotta love the meta approach. Creative, lateral, structured thinking, brainstorming, etcetera. Some say it is pseudo science. See for yourself.

Meetings can be discussions, about arguments, ending in 'yes, but...'. A suggestion, followed by risk, idea, emotion, fact, cause, effect, all through each other. Familiar? We have seen it. You cannot compare a fact with an emotion. Six Thinking Hats is about parallel thinking. One thing at the same time. First all think about ideas, without facts or emotions. Do not feed the egos. No showing off.

So what are those hats?

  • White: facts, numbers, what do or don't we know, which questions need asking, pretend you are a computer. Examples of data, research. No argumentation, no judgement. Need research, then store it for later. Somebody's facts can be another person's opinion; can you explain, based on your experience why something is 'for a fact' wrong?
  • Red: emotion, but feeling. Fire, warmth, intuition. We don't know why we don't agree, but we just don't. You don't have to prove what you are feeling, argument about it. Emotions are there, the red hat just let's them surface.
  • Black: risk, disadvantage. Critical thinking. Cautions. Being critical is easy, basic survival instinct to avoid getting eaten by a lion, you will not argue with it. It is important. Sit down with six optimists and pick a new framework... not so good.
  • Yellow: optimism, benefits. Look at the bright side. How can we do this. How will the future then look in one year. Proposals, suggestions, constructive thinking. No arguments, but do ground it with facts. Best case scenario.
  • Green: creative solutions. Edward de Bono has written twenty other books about this. Get new ideas. Brainstorming, no judgement. Thought experiments. Postpone judgement. Come up with at least five different alternatives, like five Javascript frameworks. Reverse the proposal to come up with a new proposal. Provocation: choose a random word (banana) and associate with the current proposal in mind.
  • Blue: control, meta. Blue sky. Overview. Helicopter view. Think about thinking. Meta. Organise the meeting. Which hat order do you start with in this meeting? Role of the chairman probably, or some other dedicated blue person. Observation.

Deal or no deal: if you wear the hat, stick to that thinking direction. Everybody wears the same hat at the same time. Do not say "you are too emotional", but say "you are wearing the red hat". It comes across as less hostile.

How do you use it? You can use it alone or in a group. Start without hats, but then separate the hats when you are stuck: you do not have to use it all the time. Limit the time per hat.

Why use it? It makes things easier. No mix-up of emotions. Think outside your own comfort zone: you may naturally be more black or more yellow. And of course shorter meetings.

The group leader watches the process and decides that people should now put on a specific color.

White hat: what about facts presented as opinions. Use "E-prime": English without the verb "to be". So not "no one is reading news letters", but "in my experience, this does not happen." Start with yourself.

Let's try it! Now!

First try: Let's say Plone wants a new logo. There is a proposal. Discuss it now. Proposal is four squares of blue, yellow, red, green...

Second try: we support Python 3 at the end of 2016.

[Can't summarize these tries, but it was fun and interesting to do.]

Alexander Loechel: Patterns for working Project Teams

It project management, team motivation. Novel from Tom DeMarco: Deadline. In a fictional way he describes software projects and what can go wrong. Other books: Peopleware - Productive Projects and Teams. And Adrenaline Junkies - Template Zombies.

Conclusion on why so many IT projects fail: the major problems of our work are not so much technological as sociological in nature.

He makes lots of points, with patterns and anti-patterns.

My personal conclusion: Plone community intuitively does most of his points right. Keep calm and carry on.

Eric Steele: Hey, when is Plone version X going to be released?

I get this question all the time. It mostly takes so long because I am busy releasing Plone...

Check Jenkins, auto-checkouts, check changelog, etc. By the time I am through the long list of checks for the long list of to-be-released packages, the list has grown by at least ten...

By 2020, Plone will dominate PyPI with over 99 percent of the packages being for Plone, and our cyborgs will take over the world.

Nathan: Security Team

Some of the core people are on it. There is some fatigue on the team, because it is a lot of work when there really is a problem. If your company can help, that would be cool and smart. We need someone who knows Plone really well.