Boldport

View Original

A farewell to EDA

Credit: ITV

EDA is dead, and we're to blame.

EDA is dead. It's been dying for about 20 years now. Big EDA have given up on innovation and the 99% of their users that cannot buy "platinum" support. Their software is an out-dated patchwork that's occasionally given a botox injection in the form of a new set of Windows icons. Their interfaces look like a Dreamliner cockpit bolted onto a DC-10 frame. New features are underwhelming, few, and do not address real problems, like source-control, inefficient design processes, and usability.

The thing is that despite the talented engineers working on these tools, culturally and structurally, Big EDA are not capable of innovation, and we should stop expecting them to.

'Cloud' EDA tools such as Upverter, circuits.io and others are disappointing in that they have resisted the urge to innovate. Interfaces are largely the same as their desktop counterparts and the design process is the same. But it's in the 'cloud' and 'social', neither of which are exactly what we're craving. Sadly, these tools are engineered to get aquihired by Big EDA, not to get us to perform our work more effectively, or for disruptting our outmoded industry.

Open source EDA is nearly non-existent due to the tight grip of Big EDA. The KiCAD project, for example, admirable as it is, suffers from typical community-based development issues and its 'old' core. I wouldn't be surprised -- I don't keep track --  if most of the current work is dealing with historical artefacts and personalities, instead of accepting the fact that the software needs a reboot, a natural occurrence in any software development, and one that should be much more manageable in a project such as KiCAD compared to Big EDA tools.

EDA is dead because it is no longer fit for purpose. As unholy as it is, we engineers need to accept that software development has outdone us and we have to catch up. No more exchanging zip files, no more broken tools held together by custom scripts no-one can maintain, no more Digital Stockholm Syndrome, no more status quo. We're partly to blame for letting the situation deteriorate here.

The future is not EDA. People may still call it 'EDA', but it will be something else. It won't have any of that old, mouldy, ancient feel to it -- it will be fresh, fun, and different. I'm sure of it, because the path we're currently headed leads to a low point from which new innovative tools will emerge. If you're currently developing tools for engineers, I beg you to not accept the old ways and innovate in usability and the design process, not only in crunching numbers, and the obvious 'social', 'mobile', and 'cloud' buzz. I'm trying to do that with PCBmodE, and I can assure you that it's hard, but it may be worth it!

* I mostly talk about circuit design software here, but I know that it applies to all flavours of EDA. Prove me wrong.