Is a Dead End Product Necessarily a Problem?

One of the predecessors of mid-range CAD was a mid-90s Autodesk product called Mechanical Desktop. Mechanical Desktop started out as an extension to AutoCAD R13. Not 2013, but version 13. This was about 1994. AutoCAD could do some basic 3D wireframe and even 3D solids. It turned out that AutoCAD R13 was one of those nightmare releases that you avoid like the plague, but we had to live through it before the legend was created.

I was using a bootleg version of R13 at a job in Silicon Valley to design plastic housings for pressure sensors and tooling on which they were tested. Visualizing 3D assemblies in your head is excruciating work. I can’t tell you how many sleepless nights I lay there in my bed with my fingers out in front of my face in the dark. Lots of reference sections, colors, different hatching styles… We made it to the moon in 2D, it would have been so much easier with better technology.

At a trade show (in San Jose, you can just drive across town to the convention center for all the best trade shows) I heard about a product that would extend these 3D capabilities in R13 and help you create 2D drawing views from the 3D, rather than the other way around.

Mechanical Desktop on top of AutoCAD R13 turned out to be the most miserable pile of junk I have ever used. When I started using it, it was a 1.0 product, meaning you’re a paying beta tester. MDT taught me how to better control memory, how to build a computer, reinstall and repair the OS, fix hacked installations, and many things completely unrelated to CAD or mechanical engineering. It crashed constantly. But it enabled me to do a lot of things with 3D visualization I hadn’t been able to do before. The owner of the company walked by my office one night when I was working late and saw the 3D fixture assembly design for a new product line on my screen, and stopped in to offer me more money for more capability like this. He saw the engineering and marketing value of product and process visualization. How often does that happen?

Not many years passed, and I heard that there was a raft of new 3D modelers. Autodesk was putting out something called Inventor, and then there were two other tools from upstart nobodys in the CAD world – Solidworks and Solid Edge.

So what was going to happen to MDT now that Inventor was on the way? For whatever reason, business people think it is disingenuous to tell the truth. There must be a Soviet School of Disinformation that sales people have kept alive. If management, sales, marketing or PR people are talking, you immediately assume it’s not true. Conditioned response. Inventor turned out to be a lot better than MDT, but it took them a while to copy a lot of functionality from early versions of Solidworks.

My initial reaction of course was to think about all of the time and resources I had invested in MDT, but also think about all the time I waste doing un-productive work. I eventually wound up on SW96, but I kept an eye on the progress of my old nemesis, MDT for years to come. (Read more about the history of Mechanical Desktop here: https://gfxspeak.com/2011/07/21/2008-rip-autodesk-mechanical-desktop/)

What happened with MDT and Inventor may be relevant to more recent events. You might compare the situation to Fusion 360 and Inventor or 3DExperience and Solidworks Desktop. Let’s chart out the time line and see how many parallels you can find:

• Rumors of a bright new promise for the future from a software manufacturer.
• At some point it becomes obvious that they are talking about a different and new product, not the existing one.
• Further along the timeline it becomes obvious that there is no compatibility or easy transition plan.
• Eventually the new product is released with great fanfare, and magical sounding case studies, but nobody you know has used it.
• You start hearing real stories of software that’s about half completed, and extremely buggy, certainly not ready for actual use.
• Development of the existing product continues, but at a slower rate since resources are going towards the not-yet-viable new product.
• Sales of the old product start to drop as there’s uncertainty about the future of the product, but also there are little or no sales of the new product yet.
• They start giving away the new product with sales of the old product, being able to claim that these are sales of the new product.
• Development of the old product slows down even more.
• Giving away the new software to educational institutions is a tax write off and a seat count marketing boon. Plus, in X years when the software is ready, these kids will be out of school and telling employers they have experience in this product.
• Eventually the situation flips, and the give the old product away with the new product, as sales and development have finally caught up to promises.
• The old product is no longer developed, it’s now a liability to the company, cannibalizing sales of the new product and adding marketing confusion.
• The old product is phased out and discontinued.

The fact is that MDT lived through a long drawn out death for two reasons: 1) customers wanted to keep using it 2) the replacement was promised long before it was ready.

Progress is great. But progress in the manner in which we’ve become accustomed to with the engineering software industry involves a lot of misdirection and outright falsehoods from the sales and management teams. Engineers often get the bad rap of being resistant to change. However, change – when it involves sales processes – has very often come to be associated with being played for the fool – misdirection and misinformation – or just lies, if you prefer. So engineers resist change because they associate it with being lied to, or at minimum not being trusted with the truth.

If you want engineers to embrace the new stuff you have for them, just be straight up about it.

If you are trying to sell one product of yours over another product of yours, you should really try to understand why the customer seems to cling to the old product. With MDT, it was the strength of having real AutoCAD 2D behind the whole thing. Inventor was primarily 3D, and didn’t have all of AutoCAD’s detailing or LISP automation capabilities.

Henry Ford famously said “If I had asked people what they want, they would have said faster horses.” That is often distorted into “The customer doesn’t know what they want”. If you bother to read about the vision for 3DExperience, you might find yourself agreeing that the vision itself is indeed beautiful. But its also true that you can’t just swap one vision for another – not all visions are equal. People who buy into one vision aren’t necessarily going to buy into the complete opposite vision just because you tell them this is their path of migration.

I’m sure 3DExperience is good for someone. But I think – even once it is fully fleshed out in a complete suite of functioning software – it’s going to take more than misdirection and misinformation to push people who bought into the Solidworks vision into making that change.

In 3DExperience they say everything is connected. But that has a limit. Everything within the platform is connected, but the platform is not connected to anything outside. Your data, for example, resides on the platform, and there it remains. Oh, and you essentially have to rent the space to save your data, and there is no way to remove it, because you can’t run the platform outside of the platform. The software, your data, it’s all connected. Some people will like that. Some won’t. I would guess that the people who liked and bought into the SW way of doing things are not going to like being held captive by a pay-to-stay-alive scheme.

I think 3DExperience is perfectly valid as a product. I keep saying it’s a beautiful idea. I just don’t see much of it being compatible with the idea that a company or an individual can control their own destiny. A distinctly different philosophy, but one is not a drop in replacement for the other.

Back to the dead end product, in the end, MDT users eventually had to make a switch. Some of them might have reverted to straight 2D, with some other product taking up the 3D slack. Some might have followed the march forward into Inventor land, and maybe supplemented with dedicated 2D tools. Some just did an evaluation and moved forward to a company they thought had a better forward-looking transition plan in mind. It’s costly to change CAD products. Whether you change from MDT to Inventor or MDT to, say, IronCAD, change is expensive.

You need to look at the long term history of a company to make products that don’t disrupt your company’s design process. MDT was a dead end. Dead ends are upsetting and costly to businesses. Can you think of another company with a pattern of dead end products?