Sunday, August 4, 2019

A Parachute Option... for Cloud-Data?

When aviation was in its infancy, it was enthralling, and aviation still is for many who are captivated by the thought of flying in the clouds. However in those early years of flight, when things didn't go as planned, mortality was high because parachutes were an afterthought. Today, ICON Aircraft, Inc. has developed a parachute system directly into their A5 model. In business, CEO's and others in the upper echelons of commerce often seal their hiring with a parachute clause. In these situations, the intent of the expression 'parachute' is that when things don't work out as planned, you can come out of the situation relatively unscathed.

ICON Parachute System (IPS); courtesy of ICON Aircraft, Inc.
In a similar sense, cloud-based platforms are fine when things are going business-as-usual, even exciting when everyone is in wow-mode. But what happens when all of that ends? What happens when the visionaries who created the business model decide to leave? What happens when a cloud-based provider no longer wants to be in business, or the company strategy becomes to sell-off to another owner? Is there a data parachute option for users who voluntarily want out, or those who are forced out?

Some might recall that Solidworks was sold to Dassault Systemes in 1997, only two years after its first release in 1995. Could something similar happen to Onshape as it continues its trajectory of success? Also, there have been closings of cloud-based programs before. Remember TeamPlatform for cloud-based design collaboration? It was a noble venture that after being purchased by 3D Systems was thereafter closed. With that in mind, the end of Microsoft eBooks has provided a fresh view of what a very prominent cloud-based platform closing looks like. A fresh reminder that it doesn't matter how many backup servers are dedicated, if it's in the cloud, you really don't own it. Granted there are distinct differences between eBooks and cloud-based CAD programs currently on the market, and even how data rights management (DRM) are handled. However, there are similar principles that demonstrate the real vulnerabilities and risks involved when relying on a program owned / controlled by others.

It's understandable how enticing a cloud-based platform can be for owners and developers. It provides them with 100% total control and real-time analytics on how the program is being used, by whom, when, etc. What company wouldn't want that depth of consumer insight on the use of their products? For example, Onshape is a reputable and well-received platform for CAD and they are no doubt benefiting from this insight as they continue to  develop and mature. Conglomerate Dassault Systemes is trying vigorously to stay relevant with their burgeoning 3DExperience cloud-based platform, evident by how frequently the marketing names for their new products are being revised that their domain admins are struggling to keep up.

It's also understandable that many startups and small lean-mean-business-machines have embraced cloud-based platforms. Why shouldn't they when every day and everybody you work with has a go-for-it, throw caution to the wind and take it on, kind'a day? For startups and small businesses, it's all about risk and seeing what happens!

Onshape is touted as being like a Google Docs for CAD. That analogy seems a very appropriate way to help understand how it works, how data is edited, managed and made accessible to others for collaboration. The striking exception in this analogy however, is there is no off-line mode with Onshape. There is no way to work with Onshape data untethered from the provider. And that seems to be at least one of the aspects of the trepidation felt by so many others regarding cloud-based CAD platforms. Fusion 360 by Autodesk has addressed this need with a hybrid approach, but not everyone wants a full program download nor maintain the hardware needed to operate the program. Add the fact made evident with Microsoft recently closing shop on their eBooks, that local files you thought you owned can and will vanish if a provider implements DRM tactics. Hence the reluctance by many to embrace cloud-based CAD. Users simply don't have control over their cloud-data in a tangible way that feels akin to ownership.

In our complex and dynamic world, it's no surprise that technology, companies, and programs are ever changing, being improved, or phased out. The surprise is when it affects us personally. So it's easy to understand that dependency on a cloud-based CAD platform or provider should warrant a user to question how to recover should the service abruptly end or fail to meet expectations.

Some might argue that there is dependency in every aspect of computing and that is true when talking about computer or server operating systems, programs, devices and the like. The difference is there is still a measure of autonomy with such systems. When product support ends, the user still has general control of their data, of when they want to phase out and trade up or convert. When a cloud platform ends, there is little to no control for the user. In such cases, any action plan and timing is inextricably defined by the owner of the platform.

To borrow the cliche, "it's not if but when" your CAD provider changes their business-as-usual approach. It could be tomorrow, it could be next year, or in five years. Honorable intentions aside, when change comes from your CAD platform, and it will, what data parachute options will you have? While it remains to be seen, based upon current technologies, the better cloud-based CAD provider will include the self-sufficiency of off-line mode for assurance and access to data, AND they will be prudent enough to include a data parachute. In other words, providing users upfront with the means to access their data using a simple export engine, and / or the means to translate data into another format if you want, when you want, independent of the platforms own existence. Perhaps through a consortium approach that can rightfully address the current risks to cloud-data. [Update: At present, Onshape is setting a leading example in addressing the concern of cloud-data accessibility. As noted in their privacy overview, they provide export of design data into any of several industry standard or proprietary formats. Notably, they also provide a full, free API that let's you access all of your data programmatically. Thus enabling you to get your data into, and out of, Onshape at any time. Adding, their philosophy is that by making it possible for you to leave at any time, they're forever motivated to improve their service so that you'll want to stay.] 

As beautiful as the clouds may be on a bright sunny day, how confident would you really be to fly with a provider that sells itself as reputable and reliable, yet purposefully neglects to provide you with a pre-flight safety plan in the event you encounter the unexpected? So, with a cloud-based platform entrusted with your data, wouldn't you expect to have a parachute option?