A lot of people are finding themselves in the position of looking for a new CAD package these days. The market is stirring itself up, there are new offerings, and the old offerings are just that – old. Not that there’s anything wrong with old. I’m creeping up on it myself.
The process of buying a new CAD system is going to be different for organizations of different size. Smaller companies may or may not have a separate IT group to deal with. And even if they do have one, IT doesn’t always get a say in engineering purchases, even engineering software. You may need to have IT have a look and make sure there is nothing out of range, but they are not going to be the primary signature required to complete your CAD software purchase.
Larger organizations are going to have increasing participation from departments other than your engineering design and manufacturing departments. In particular I want to look at the involvement of the IT department. IT is a service that your entire organization makes use of. For companies involved in creating engineered and manufactured physical products, IT is not generally driving the show, although you couldn’t accomplish what you need to do without them. IT provides computer hardware, internal network, data storage, backup, connection to the outside world, security concerns, commodity software, training, maintenance, upgrades, etc. Every company does things differently, but lets assume for the purposes of this article that this is the general role of IT in a company that manufactures or designs physical products.
The point is that while IT is certainly necessary to do your work, you want mechanical engineers to be the primary decision makers when it comes to your design software.
Let’s take a broad look at the new offerings in CAD systems. Maybe we can oversimplify what’s new by only calling out one distinguishing feature of a “new” CAD system, and let’s call that feature “cloud delivery”. Let’s define that as code executed primarily on a cloud (remote) server, and data also on a remote server. Although you can (and some surely will) argue this, the main innovation for these new systems is the cloud. Now whether the cloud is indeed an innovation or more simply and update to a decades old mainframe paradigm is again up for debate, but let’s ignore that for now.
I’ve had two demos of CAD in the cloud, and they both ended with a lot of hand waving at the screen because something in the Web connection got fouled up. It just adds too many critical failure points for a system that your company can’t live without. A lot of pieces have to be in place for this to work, and you don’t control several of them. So your business is at someone else’s command. Are you really willing to take that kind of risk? Is the reward worth your loss of control?
There are a lot of advantages of this way of working that I’m sure you’ve heard by now, but let’s list some of them here to remind ourselves.
- You don’t have to worry about installing software because its installed for you. You don’t have to mess with upgrades for the same reason. There may or may not be some local component that you have to install, but let’s just say that’s automated.
- You don’t have to worry about PDM because that’s included, and you don’t have to worry about local file management.
- You don’t have to worry about local workstation class hardware. You don’t have to worry about working from the office – anywhere with a net connection is fine.
Those are some compelling arguments. Less cost, less worry, less infrastructure, more flexibility.
One problem with this is that the main innovation here is in IT. It’s not innovating design tools, it’s innovating how those tools are delivered. Many of the new requirements are going to come from IT. So this new CAD system in some ways stops being an engineering tool and starts being an IT concern. So whether IT was involved in your engineering software decisions before or not, it certainly will be now, and they won’t just have an FYI role, they will be a major signatory.
While the compelling arguments above are still compelling, they aren’t the entire story. Imagine the load on your external network connection when your entire CAD production is now going out over the internet. You already know what kind of a load CAD places on an internal network, but external network bandwidth is decidedly more expensive, and worse than that, it’s beyond the control of your company.
Next, your data is all off-site. Backups off-site are a recommendation, but your primary data being off-site? Every week the news reminds us that regardless of how big and powerful your company or government organization is, it is still vulnerable to attacks from (X) for industrial or governmental espionage. They may not only steal your remote data, but they could also erase it, or otherwise deny you access to it.
Granted, an employee with a thumb drive can always walk out of the building, but the only way to guarantee your network is not the source of the hack is to make sure your internal network is not connected to the internet. Sounds extreme? So does giving your design data to whoever might want it. Security is an issue regardless of where your data is, but putting it in a big juicy data center wraps it up in a nice bow, and paints a big target on it.
But is innovation in IT really what you need from your CAD system? Don’t you need CAD innovation from your CAD system? Better, faster ways to manipulate geometry? More reliable ways to make changes? Making sure “design intent” works for you instead of against you?
So if you’re in the market for CAD, make sure that the product you select is making innovations that matter to you. Make sure they are focused on delivering mechanical engineering design tools rather than spending your money developing IT services. Are the benefits that your vendor of choice claims really the benefits that you need?