TYPO3 Conference 2016 - Day 2

The TYPO3 Conference T3CON16 is happening in Munich. This post describes talks of the 2nd day.

The History of Future

Erik Händeler talks about the future. We have do identify the bottleneck in our companies. This has also the same in the history. Examples are the railway or the steam engine. He thinks that health is the next bottleneck in our future.

Working means solving problems. Therefore we will never run out of work. It is the way we work which will change.

Wealth depends on productivity. In the information age, it is hard or impossible to measure the productivity. The biggest problem is how we can improve the work with information and the ability to work together. However this is no technology problem. If people can work together it will solve a lot of problems.

The russian economist Kondratieff (1926) found out that there are long waves and a big invention in between. Those have been the steam engine (1815), the railway masstransportation (1873), electricity/steal/chemistry (1918), automobile individual mobility (1973) and IT digital transformation (2002). One eaxmple for the curves is the usage of the mobile phones in Germany. In 1991, only a few people had a mobile. From 1997 people began to buy a mobile and now the amount of mobile phone users won’t getting higher and now the chart won’t get up anymore and the business won’t grow anymore (that much as previously).

We don’t predict the future right as we think in the present and that things won’t change. We need a better way of struggling. Erik predicts that we need to work with less pressure as we need to work for more years in future.

Where will new work come from? New work is only created in informational work like developing, organizing, planning, deciding. THe segment of service will stay the same and the required work places in industry will getting less and less. Again he says that it is hard to measure the productivity. The price of an article sold to a newspaper is not bound to the amount of time invested in writing. The bottleneck is to get the right information and not to work harder.

It is about dealing with information and dealing with people. Everything else is exchangeable (like buying a machine). The culture makes the difference and not the technology:

  • Truthfulness instead of manipulation
  • Solving problems instead of suppressing tensions or fighting them
  • Mobilising knowledge of whole organiziation
  • To look for another’s benefit instead of only one’s own benefit

Performance Solution Stacks

Bjoern Gottowik begins with a story about a client. A new projects starts and there are a lot of things a client expects and the company promised like features, good search, good performance, connect with other systems and working on mobile devices.

The project starts from scratch with the idea to make it better than before. THe website is finished, the client is happy and the first stack is done. It is a TYPO3 server instance with a LAMP stack and a solr server instance. The client demands a better performance and the next stack is added with a proxy cache by using nginx and a separate instance holding the database.

The clients needs more like A/B testing, better file handling with multiple languages, product management, digital asset management (DAM), e-mail marketing. Those are boundaries which TYPO3 can not handle that well as content.

A possible solution statck would look like using solr, celum for DAM, akeneo for product management and shopware. It is not about leaving the TYPO3 community but to find out how those systems work together. Can we share knowledge in that regard?

The idea of the performance solution stack consists of:

  • Supporters: Solr, Akeneo, celum
  • Provider: TYPO3 & Shopware
  • Sender: Apps, Mailchimp
  • Tracking, Analytics, Monitoring with sistrix, optimizely, datameer, google analytics

I am looking forward to more sharing

How to change the world and survive the journey

Gina Steiner is talking about how to guide the change of a line-staff organization to an agile organic structure, without losing control. The most important topics are:

For all

  • Entropy
  • Trust
  • Leadership

For the team

  • Autonomy
  • Responsibility
  • Common misjudgements

For the “bosses”

  • Alignment: A state of agreement or cooperation ammoung persons or groups
  • Delegation
  • Purpose, visition, strategy, tactics

Alignment” and “Autonomy” interact with each other. The “happy spot” is the one with a lot of autonomy and alignment, which is the target to be reached. Imagine having full autonomy and no alignment, it it hard to get to the happy spot, same as with full alignment and no autonomy.

Entropy is cooonly understood as a measuer of disorder. Such system will proceed always to become more. To follow the path to the happy spot, you need to add alignment to a group as the entropy will lead to more autonomy (“people will start to do things they want”). The questions are:

  • Who regulates?
  • How do you regulate?

This is where responsibility comes it. Parts of it are delegation, judgement and trust. Delegation is like how you let your children learn. It is about increasing responsibility slowly and about the levels of delegation (Jürgen Appelo). There are 7 levels

1) Tell: You make a decision for others and you may explain your motivation. A discussion about it is neither desired nor assumed. 2) Sell: You make a decision for others but try to convince them 3) Consult: You ask for input first, after consideration you make a decision 4) Agree: YOu enter discsusision with everyone involve 5) Advise: You will offer others your opinion and hope they listen to yourwise words but it will be their decision, not yours. 6) Inquire: you first leave it to the others to decide and wadterward you ask them to convince you 7) Delegate: You leave the dicsion to them and you don’t want to know about details

Common misjudgements

  • Liking/Loving: You are not objective about people you like
  • Disliking/Hating: Hatred causes distorted thinking and blindnes to the fact
  • Doubt Avoidance: To remove doubt, people rush to decide
  • Inconsistence avoidance: YOu resist to change and are unlinkely to change your loyalities, beliefs, habits
  • Social proof: Automatic tendency to think/act like those around you
  • Express self-regard: You have inflated opnions not just of yourself, but also of people like you

Trust - golden rule

Each has given his best at all times, with the possibilities, the skills and the knowledge that he had given to the time available.

Purpose, vision, strategy & tactics

It is about “why do we the things?”. A lot of times this is not known.

  • Purpose: Is your passion
  • Vision: is the world you wish to create
  • Strategy: is how to get there
  • Tactics: ensure you survive the journey

Those questions should be solved top to the bottom, so starting with the purpose and solving the tactics last. You can’t delegate everythink. So the bosses have to care about where to go.

Leadership: This is a check list of things a leader should do:

  • Create a strong vision
  • Set a direction
  • Set boundaries
  • Forget carrots and sticks
  • Be a teacher
  • Admit mistakes
  • Lead by example
  • Encourage leadership at all levels
  • Help people deal with fear
  • Adress the elephant in the room
  • Improve the system
  • Be prepared for a long journey

What Gina did in your company:

  • Involvement: Incorporation of everybody
  • Contribution: Team retroperspectives
  • Leadership & responsibility: roles in the team, delegation level
  • Coordination: Retroperspective of these roles
  • Communication & commitment: COPs (community of practice)
  • Controlling
  • Strategy: Company retroperspectives


  • Keep going, because of the entropy
  • Do not overact
  • Trust: The golden rule
  • First purpose

Lesson learned from A/B Testing

Ingo Schmitt got the first contact wiht A/B test during sailing. There are different ways to reach the target. The question is: Which way is the fastest?

Why do we do A/B tests:

  • Find better solutions
  • Verify the best solution

It always starts with a hypothesis which needs to be proven. This can be done by creating a 2nd version of a site and presenting both versions without telling the users. Each visitor and click is counted and after doing some statistics the better version can be found out and used.

But what if the new version is afterwards less successful? After each step a user can abandon his action, so it is important to set the goal which is measured to the last touch point. As an example this is for a shop the sale and not how money visitors go to the checkout.

Every link on a page is a possible way for a user to interact with the apge. Additionally a last option is also to close the browser. Important is to check the calculcation of significance, so if the difference between 2 variants is significant or just a result of randonmness. Do not trust your “feeling” about a result. One option is run a A/A test.

It is important to take normale variance of visits/page impressions. A lot of different things can change the tests like doing tests on gardening products will bring different results while it is raining or sun shine.

Also use the normal status quo as control variant to varify the results. So have the default site and 2 variants instead of just 2 modified variants.

Not all tests have a result and accept failure. Negative results will lead you to different areas of optimization.

A/B Tests must reflect your business case. As an example, improving the usage of paypal or a payment provider can lead to more expenses for the company itself as they have to pay for usage of paypal more than by e.g. invoice.

The future of A/B tests could be to use a recommendation engine to improve the result automatically.

Future of TYPO3

Benni Mack and Mathias Schreiber hold the last talk of the T3CON. Topics are

  • Cloud,
  • Technology
  • Midterm strategy
  • Industry partnerships
  • New Markets


Cloud in a nutshell.

  • Application stack: It’s about the runtime, service, asset storage, persistent storage. If you need to scale it, you need to split those up into multiple instances.
  • Simple setup: Deploy to multiple regions, using orchestrated, consistent, dependable deployment and building site assets before the deployment
  • Scalability: It should be dynamically for up and downscaling. It ensures great customer experience. In the end it is pay what you use, so really cost efficient.
  • Security: Remote code execution is a big problem which is not relevant on a read only filesystem


  • Database abstraction: Using DBAL doctrine since CMS8 enables to use a wide range of supported databases.
  • Multilanguage: It is currently very flexible but can be improved for sure.
  • Frontend editing: Awesome frontend editing will come to the final 8 LTS, released in April 2017
  • Event sourcing: Oliver Hader wrote a master thesis about this modern paradigm and it is checked if and how it could be used in the CMS

Midterm strategy

  • Accessibility & SEO: It is a lot of editorial work. Editors should be supported editing content and reports give additional information.
  • Personalization: Personalized content is about targeting content with the proper content. THe same is about A/B tests
  • Integrations:
  • Time 2 market


Seamless integration of external components: More standardized APIs, e.g. to ElasticSearch, Service Integration

Time 2 market

  • Easier configuration
  • Better defaults
  • Faster integration

New industry partnerships

  • Microsoft
  • platform.sh
  • amazon webservices
  • google cloud platform
  • blackfire

New markets for you

A project as you know it consists of: Pre sales, consulting, design, implementation. Compared to proprietary CMS, TYPO3 is quite cheaper regarding the total cost of ownership is there is no extra costs for additional users or deploying to multiple regions. This should be seen as opportunity.

Expand the market, seize opportunities and pave the way to the future.

The TYPO3 Inc supports agencies here as well with:

  • Enterprise support
  • SLA
  • Sales Rep
  • Security Audits
  • Project reviews
  • Lead Management