Pollination [Development] Roadmap for 2023

mostapha · December 23, 2022, 6:00pm

Hello community!

2022 is coming to an end! It was a critical year for Pollination. After years of development, and almost a year of early access we transitioned our products to paid products. The Rhino plugin was the first to go to market in early March and was followed by the Revit plugin and Pollination Cloud later in March and April. We also introduced Pollination Apps as a beta product.

As it always happens we had so much to learn from you! The feedback from our users has been invaluable and we have done our best to listen, learn and improve. We have already thought several workshops on Pollination, and it has been really satisfying to see Pollination is doing what it is supposed to on real projects! Thank you!

I’m writing this post to share our development plans for the upcoming year and get your feedback on them! Keep in mind, that this list can and will change as we work on them during the year as we keep learning about your needs and receiving more feedback. I will also try to limit the explanation for each item to a few sentences. Let me know if any of the items need additional explanation.

Before Getting Started

There are two topics that we will be working on in 2023 for all the Pollination products. One is improving the documentation and providing more educational materials, and the other one is improving the overall user experience for running the simulations and visualizing the results.

1. Documentation, tutorials, and workshops

We are well aware that we need to do more documentation and presentation on Pollination products in 2023. We have already planned for a series of workshops in collaboration with Performance Network, and we will be planning for more tutorials and workshops.

In case you missed it the recordings of our recent workshops at the AEC Tech 2022 are now available online:

Let us know if there is a particular topic that you want us to focus on first. We currently have the documentation for the Rhino commands and recording tutorials for Pollination Apps on top of the list.

2. Architects vs consultants vs engineers

One pattern that we are really proud of about Ladybug Tools users is the almost equal distribution of architects, consultants, and engineers using the plugins. We have always tried to be the inclusive platform that brings everyone together to share their expertise during the process.

In the case of Ladybug Tools, architects picked it up first, and it was then followed by consultants and engineers. In the case of Pollination, and in particular the CAD plugins, it is consultants and engineers who are leading the adoption.

We think the best explanation for this pattern is that a clean simulatable model mostly benefits consultants and engineers who are responsible for creating valid models. The other explanation is that our current workflows for submitting simulations and visualizing the results for simple analysis are still under development. In particular, for Ladybug Tools users, it is still easier to use the Ladybug Tools scripts.

We are looking forward to seeing how this pattern will change in 2023 after we improve those features. We have already started working with a few architecture offices to help them with their transition to Pollination for running their studies.

If you are an architect and have been thinking about using Pollination but see some gaps between what it offers, and what you need we would love to hear your thoughts.

Pollination Products Breakdown

Rhino Plugin

The Rhino plugin was the first to the market and is our best-selling product! The most-loved features of the Rhino plugin are the features for preparing and fixing analytical models. It provides the best user interface for generating building performance analysis geometry. That in combination with providing reliable export routines to other file formats has made the Rhino plugin the goto plugin for building models. Most of the users are still running the simulations in Ladybug Tools Grasshopper plugins, or other 3rd party tools like IES-VE, and DesignBuilder.

Here is the short list of the topics related to the Rhino plugin for 2023:

Add new repairing routines, and improve the existing ones: we have already developed tens of routines to find the issues in a model and fix them. We are now starting to combine these commands into routines that can be used for repairing models quickly. The workflow for fixing the misalignments in a model is an example of what we will be working on.
Improve interoperability with other simulation tools: we already support exporting to a handful of simulation tools. We are considering the development of a native export to IDA-ICE, and also improving the export routine to eQuest to support a larger range of geometry.
Improve interoperability with Ladybug Tools Grasshopper plugins: we are working to improve the routines for bringing the updated models or simulation results that are generated using the Ladybug Tools plugin to Rhino. You should be able to replace your current model with an updated model in Grasshopper or bake the results of the study as a smart results grid in Rhino.
Streamline the process to run the simulations and visualize the results: we are working to unify the user experience for running the studies and visualizing the results by using the Pollination panel. We have already started the development, and the new workflow should be available for testing early next year.
Improve the documentation: See above!
Provide better solutions to create and share custom standards: we will develop a streamlined way to manage a custom standards library of constructions, schedules, programs, modifiers, etc. We had several requests for easy ways to build and manage their own office standards library with the Rhino plugin and share them using the Pollination Cloud organization.
Provide better solutions for detailed HVAC simulation: This has been discussed here.

Revit Plugin

The Revit plugin is our second best-selling product overall, and the best-selling product among consultants and engineers who are not necessarily Rhino users! We have come a long way from the first release of the Revit plugin. We had to go through the process of testing the Revit plugin with several real-life models and hear your feedback to identify and focus on the most critical features.

Here are a few learning points:

Editing a Revit model directly can be challenging for two other reasons. The first one is model ownership, and the second one is the lack of knowledge of working inside Revit.
We heard over and over that consultants and engineers want to get information like floor plans and apertures out of Revit and continue their work in different software.
Interestingly enough, architects who use Revit have a strong desire to stay inside Revit forever and do everything inside Revit.
Until we have better-repairing methods, the most realistic and efficient option for preparing large energy models is to export a workable model from Revit and prepare it as a simulatable model in Rhino. This workflow is very reliable and is still times faster than any other available workflow in the market. This is one of the reasons behind introducing the bundled license for the Revit and Rhino plugins.

Our goal for this year is to minimize the need for additional work outside the Revit plugin, knowing that it will not realistic for all the models but there is still room for improvement.

Here is the short list of the topics related to the Revit plugin for 2023:

Develop more forgiving solutions for exporting models: we are working on a new method to use Revit areas for extracting the model. This option will remove the need for placing every single room and space in the model.
Improve the solve adjacency routines: if we want to recommend the Revit plugin for detailed energy modeling we have to ensure that our solve adjacency routines are reliable. Our current plan is to use our core libraries for solving the adjacency between the faces instead of using Revit routines.
Improve automated repairing routines: we have started introducing a few fixing routines to the Revit plugin but they have a few limitations. One of the most important ones is that they only work in extruded mode. We have plans to expand these features to also be available in the detailed geometry routine.
Streamline the process to run the simulations and visualize the results: The approach here is the same as the one for the Rhino plugin. See above!

Cloud Computing

Cloud computing was a topic that we had so much to learn about balancing the expectations of the market, and the reality of what can be done.

So far, we have found two scenarios where cloud computing is very helpful. The first case is for the offices that are frequently dealing with long-running simulations, and the second one is for the offices that have specific projects that need help with scheduling a large number of simulations for their projects with a dedicated budget. We picked one testimonial for each category to add to the Pollination website.

The expectation that we couldn’t fully fulfill is to provide an affordable service for infrequent use of the service. The reality is that offering a cloud computing service is fairly expensive which limits our options for what we can offer at this point. The recent changes in our pricing is meant to optimize the availability of the cloud computing for those cases.

Here is the short list of the topics related to Pollination cloud computing for 2023:

Optimize the execution of running short-lived steps and small models: the original design of Pollination for submitting several long-running simulations. What we saw instead were several studies with thousands of shoebox models for parametric studies. We have started optimizing our workflows to improve the run time for small models. They will be available in early 2023.
Implement better queueing for large parametric studies: even though the cloud computing resources are almost infinite the resources that are available for each account on Pollination are not! We have seen users submitting thousands of simulations when an account has 50 CPUs and then being unimpressed with the results. We have plans to make the user experience closer to what one would expect by queueing the submission of the simulations so the ones that are submitted first will be finished first.
Provide simple solutions for customizing existing recipes: Pollination recipes are customizable but the process of customizing a recipe is not easy. From what we have seen most of the customization that people ask for is for additional postprocessing steps. We have plans to provide easier ways for users to add those steps to the existing recipes.

Apps

Apps are still in beta which makes them different from all the other Pollination products. Also, unlike the other Pollination products, we put the apps out into public almost from day one to get a sense of what you want to do with the apps.

We spent this year building the infrastructure for building, deploying, and running an app. We also worked on building a robust integration between the apps and the CAD plugins. Finally, we taught two in-person workshops to get a better sense of the level of expertise in the industry, and what should be considered for a successful larger adoption of the industry.

We have made major progress in all directions but we still have work to do for getting all the pieces to work together smoothly. Our goal is to officially release the apps at some point in 2023.

Here is the short list of the topics that we will be working on in 2023:

Improve pollination <> streamlit libraries: we have developed a few libraries to transform a Streamlit app into a Pollination app. We will be working on improving these libraries to make the process as smooth and as easy as possible. The most important libraries are: pollination-streamlit, pollination-streamlit-io, and pollination-viewer.
Improve documentation & tutorials: See above!
Establish development best practices: we need to establish the best practices for developing Pollination apps as we are developing more examples. By using these best practices in our sample apps it will be easy for others to develop their own apps quickly.
Improve the start-up time: we will work on finding a cost-effective solution to improve the start-up time for the apps. There are several available options including the option of using users’ desktop resources for running the apps from inside the CAD plugins.

Before letting you go

If you have made it this far into the post you are here for real! Thank you!

The main reason for writing this post is to share our plans with you and ask for your feedback. We have been always building our products in public and Pollination is no different. Looking forward to hearing your thought. Don’t forget to share them here!

This is also an opportunity for me to thank three groups:

First, I want to thank everyone who trusted us and trusted Pollination and became a Pollination customer from year one! We really appreciate your business! You are the reason that we can continue doing what we do! Thank you!
Second, I want to thank all of the members of the community who tested the products, reported the bugs, and shared their feedback. We really appreciate you. Your help has been invaluable and you are our unfair advantage. Thank you!
Finally, I want to thank our Pollination team and collaborators who made Pollination happen. I have been boasting about having one of the most efficient and capable teams in the industry. It’s a pleasure to work with you. It is a privilege to be part of a team that is dedicated, capable, patient, and focused on fixing some of the hardest problems in the industry. You are incredible! Thank you!

I’m looking forward to hearing your thoughts in the comments, and I’m looking forward to seeing you around the ecosystem!