Cadalus
Mission
Cadalus is a project to free the architecture, engineering and construction (AEC) industry from the constraints of proprietary software.
Over the past 40 years, the AEC industry has been dominated by a few proprietary software vendors, which has led to a lack of innovation and user experience stagnation. The main developments in the industry has been new and exciting monetization strategies, from overpriced updates to overpriced subscriptions, to finally overpriced and hard-to-calculate token-based pricing.
Cadalus is to provide a European, free and independent alternative to proprietary CAD, BIM and Archiviz software which nevertheless remains interoperable with existing workflows.
Goals
Replacing an entire tech stack is best done incrementally, step by step. Each step individually is designed to provide value to the AEC community.
Stage 0: Preliminary work
Before any development can start, a lot of preliminary work needs to be done. This includes research into the problem space, interviews with architects, engineers and builders, and a lot of planning. It also includes incorporation, and sorting of financing. This stage is already underway and is yielding documents like a finance plan for this and the following stages, SWOT analyses, a draft for the articles of association and so on. This work is currently being done by Leo Wattenberg, the initiator of the project.
Stage 1: Quick sketches and floorplans
Being able to quickly generate sketches and floorplans is the core feature.
Plans should be easy and fast to create, with the program intelligently layouting the document to create a high-quality output which can be used by builders without further tweaking. Additionally, the output of this program should be interoperable with existing software, so that using Cadalus at this stage remains useful in production.
This stage would require two people to work on Cadalus.
Stage 2: Sketch to 3D geometry
CAD sketches by definition contain all measurements necessary to construct a 3D environment. This process should happen automatically (perhaps: in the background), to create a fast check with a three-dimensional reality and quickly spot impossibilities.
Given that this work is now "real CAD" work, a full team is required here: Product owner, designer, 3x developers and a QA.
Stage 3: Geometry to visualizations
Once 3D geometry exists, the European open source program Blender can take over and do PBR materials, lighting, animation and rendering. Blender has been doing massive strides recently, and outperforms many commercial applications already. Cadalus would provide a method to integrate with Blender to create high-quality visualizations of the 3D geometry.
The work on Cadalus at this stage would include improving the CAD workflows, and providing meaningful base assets for visualizations. The aforementioned 6-people team would be sufficient to get this stage done, though a given the parallelization of some of this work, a larger team of up to 15 people may be appropriate (depending on funding).
Stage 4: A fully-fledged BIM
Going from a CAD to a BIM certainly is a longer process, but ultimately only a problem of data input and features which perform calculations based on that data. That is to say, it's all in the realm of possibility.
What features would need to be implemented with what priority would be up to the AEC community. The team size here would grow appropriately with the people, though likely will remain comparatively lean as building new features tends to be easier than maintaining ancient, crusty code found in other CAD software.
Stage 5: Continuous development and maintenance
Once the initial stages are done, the project enters the fairly unexciting stage of just maintaining a good tool. This means supporting additional use cases, fixing bugs, ensuring compatibility with new hardware and software, and ensuring that the software remains usable and useful for the AEC community.
Realization
Cadalus requires comparatively few resources to get started. The equivalent of 8 seat licenses of the Autodesk AEC Collection would be enough to hire a full-time employee, so to complete stage two with with six employees, an equivalent of 50 seat licenses might well be enough. This means if only 1% of the architecture firms in Denmark alone would be willing to support this initiative, this project would be sustainable.
Technical realization
On a technical level, Cadalus will most likely be a Blender extension. Doing so has various advantages:
- Blender will be part of the chain anyway – at latest in the visualization stage.
- It's widely known that Blender is a powerful 3D editor, though maybe lesser known that it's a powerful 2D drawing and vector graphics editor as well, with its grease pencil and curve editor.
- Being a Blender extension means that a lot of the difficult technical work – UI system, data model, data management, etc. – is already done, and can be used by Cadalus. This massively cuts back on development time.
Organizational realization
A European Co-op (SCE) likely would be the best approach here. Individuals and architecture firms would join the coop, and both finance the operations of the development, and oversee the direction of the development. Details may be found in the articles of association.
For example, a full membership might be 500€/month (the price of a single Autodesk AEC Collection license), with supporting memberships below that level being available. Members with a full membership get to vote on what features get implemented next, to ensure that the development is in line with their very real needs. Supporting members may leave suggestions, but won't have the ability to directly influence development.
With this model, the co-op would be sustainable extremely quickly. If the members feel this price is too high, they might vote to lower the membership fee. Likewise, if they feel more development is needed, they might vote to increase membership fees, or figure out other means to invest into the co-op – or charge non-members for access to the software.
Unlike a normal corporation, this model also avoids a misalignment of interests: A VC-funded corporation would enjoy a period of hypergrowth where they de-throne the encumbents, but very quickly pivot to being comparable to the encumbents, with incredible pricing and draconian terms, to make up the 1000x ROI the investors so desperately seek. A co-op, funded by members which are also its customers, cannot experience this misalignment.
FAQ
or rather, miscellanious problems and solutions.
How to ensure adoption of Cadalus? The industry is caught in a chicken-and-egg problem at the moment: Everyone uses AutoCAD and Revit, therefore universities teach AutoCAD and Revit, therefore everyone learns AutoCAD and Revit, therefore everyone uses AutoCAD and Revit. To break this cycle, Cadalus needs to have explicit support from the industry, so universities can confidently start taching Cadalus. By using a co-op model, the industry is involved from the start.
Should Cadalus be free? Ideally, yes. Charging for the software creates an unnecessary barrier in an already fairly gatekept industry. Given the strong professionalization of the user base, there's a strong sales argument to pay for high-priority bugfixes and support (becoming a member), to avoid a day of lost work or downtime. If Hans-Peter can now lay out his living room slightly better, or Dmitry from Unregulatistan now is able to design a building which won't fall over – that's great, but also not the sort of audience which would have paid for such a software to begin with.
Should Cadalus be open source? As a Blender extension, it is required to be. However, even if it wasn't, there is a good reason to go open source: To avoid the risk of vendor lock-in, to encourage interoperability, and to avoid the risk of mismanagement ruining the software forever – you can always fork.
Are there regulations, Norms, Lobbying, which requires proprietary solutions? Maybe. But for once, the lobby is on our side: There are very few creators of proprietary CADs and BIMs, but a lot of AEC businesses responsible for unimportant things like our entire built environment. Whatever regulations in place surely can be adapted to allow for something like Cadalus to exist.
Is there a financial reason to become a Cadalus member? At its current rate of 460€/month, across a 40 year carreer, the average architect can expect to pay 220k € to Autodesk. Given there no doubt will be price hikes throughout, this number can be expected to grow well beyond a quarter million euros, per architect. Cadalus meanwhile would be more like an investment: Once the software is mature, you may choose to drop out of the co-op, and keep using the software – for free. In the long run, this will save you a lot more money.
About
The Cadalus project has been initiated by Leo Wattenberg, who previously worked on the Audacity audio editor for 4 years. The task of his there was similar to what he envisions for Cadalus: creating a free and open source alternative to proprietary software which is easy to use, reliable and eventually to surpass proprietary software in terms of features and user experience.
If created, Cadalus would likely be licensed GPLv3.
Join Cadalus
At the moment, Cadalus is still looking for partners to found the company. If you'd like to be one of the co-founders / founding members, please contact me.
Contact
If you want to get in touch, you can do so via email.