I started this thread: https://community.ptc.com/t5/Creo-Parametric-Ideas/Linux-area-in-this-community/idc-p/747238
But it was likely at the wrong subforum, so I continue here instead, although it is not about system administration per se.
I understand that software developers do not want to support more than Windows. I do not expect a Linux version of Creo to plop out anytime soon just because I want one.
I only suggest that PTC arranges a place where we could at least communicate our wants and needs for Linux versions of software. Worst case it gets no attention, middle case PTC can quantify the desire for Linux software, best case we get it.
Me personally, there is nothing that ties me to Windows except Creo Parametric *). I use a MacBook for everything networked, and airgapped workstations for CAD and R&D. In my world, Windows is a liability in that it costs extra money, and the very least bugs me about on-line stuff that I have no interest in and have to turn off manually.
* - Not true, I need to bug ANSYS to release Workbench for Linux as well.
As PTC has had a Linux version of Pro/Engineer/Wildfire/Creo in the past and dropped due to lack of customer demand, I do not see them resurrecting any of the Unix-based operating systems in the future unless something drastically changes in the marketplace. Supporting an OS for 1-5% of your customer base is a very expensive option. If it could be shown that maybe 25% of PTC customers were willing to use a Unix-based workstation for their CAD work, that may change the direction of software OS support.
As was stated in the other thread, the end-user with the workstation using Creo does NOT get a choice of the OS on their workstation, it is dictated by IT. I have been an admin of CAD and Windchill on VAX/VMS, Unix and Windows. With a GUI interface, the OS does not matter as much as it did when everything was command line entered.
The last Linux version came - what? 2004? That was 17 years ago, and we used to struggle to get basic hardware functionality from it back then. Today it is as plug-n-play as it gets.
I do understand that there is a resistance to transition from Windows to Linux, and weak reasons for doing so - but that is based on assumptions and gut feelings. There is no mechanism to actually quantify this - which is what I ask for here...
With your extensive experience as an administrator of CAD environments, you are a great example of the type of input I'd like to see. What would be the easiest and least problematic to run - a network of pure Windows, pure Linux (or *NIX), or a mixed machine park?
Throw in concerns such as "we do not really want to put our R&D in a cloud, because of rampant hacking and ransomware, especially since our military contract explicitly excludes any condition where our data is a) online, and b) hosted on foreign servers. Which goes counter to the increased "connectivity" and telemetry that comes with Windows.
My limited experience as IT monkey comes from many situations where the company does not realize the difficulty level of deploying software and wants to play as little as possible. "Can you install this? It has to run tomorrow. Our budget is one dollar and a pack of cigarettes". And I find that the world where I can smash in Linux and LibreOffice and whatever is an order of magnitude less encumbered with snafus than the cobweb of licensing and registration, with scores of usernames and passwords required, that I find myself in when it comes to Windows.
Also, I find it weird that so many companies choose MacOS as the secondary redhead stepchild. Apple has a marketshare in the single digit percent. Linux powers a billion devices and the only reason it does not take over the desktop is because of the very lack of mission-critical user software. It could replace both Windows and MacOS in a split second if the applications were there.
A single unified network of computers and servers is always the easiest to manage and configure.
As to the cloud, we run Oracle apps for the business on the cloud, but we would never run our engineering apps and data storage on the cloud. I run 3 Windchill systems, 2 are on the open network, only computers within the domain can log into them, and a third Windchill on an air gapped system that handles classified data. Until 2 years ago, we did all of our work on the classified side, but have expanded our working envelope where we had to add the unclassified Windchill systems and Creo.