What are the impacts of implementing a hot-desking model for grad student workspace?

I have seen hot-desking (or the equivalent) work well with undergraduate researchers. In the particular incarnations that I have seen it, however, it wasn't a large-scale formal hot-desking model like you're talking about, but rather a sort of "pool office" that was the shared space of a bunch of undergrads. With the undergrad researchers, this worked well because the number of hours each student typically spent in the office was quite small, and they tended to work on projects for relatively short periods of time.

Graduate students are a different matter. They tend to be around for a much larger fraction of their time, and you want them to be in the office more and to develop professional networks and commitment to the group. Thus, I don't see the advantages of hot-desking for undergraduates translating well to established graduate students in most STEM environments. Here is a report of a personal experience that seems to corroborate the problems.

That said, there are some environments where is seems like it could be appropriate, in which one does indeed expect an office space to be transient and sparsely used. For example:

• If students spent most of their time in lab, then their "home base" is their lab bench, and a hot-desking space for non-lab tasks could be workable.
• Some graduate programs start with a year of "rotations" in which students work with a number of different groups before committing to one, and since rotation students are inherently transient, hot-desking might work for them as well.

This all seems to point to some potential metrics, along the lines of measuring how much time is spent working in the hot-desk space vs. how much of the tasks done in that space are being done elsewhere. If students are elsewhere because they are doing different types of tasks, then it might be guessed to be succeeding; if they are doing the same sorts of tasks in coffee shops, libraries, etc., then it might instead be driving them away.

I am a graduate student in computational neuroscience and the "hot-desk" model has worked for me personally. Here are the reasons why it works for me:

• I do all of my work on my personal laptop which I can cary from office/home as needed (code, manuscripts, grant applications, PPTs). I use a cloud service for backups.
• Majority of my communication is done via email, TXT, phone, Skype/GH. For in person meetings I usually go to wherever the other party wants to meet.
• All my grading/teaching materials are online (Blackboard, PDFs, PPTs)
• All my papers are digital/online (PDFs, I use school VPN to get library access to papers)
• I have all my textbooks in digital format (again PDFs)
• If someone gives me a "paper" paper, I usually find its PDF online and recycle the original. I use apps like CamScanner to digitize to PDFs most of anything else.

I personally like the model because I am free to work from wherever convenient (lab office, home, library, cafe, etc...) and can control the level of interaction with others more easily (difficult for others to walk-in and interrupt me when I'm not in the office). One downside is that my laptop is not as powerful as a desktop would be at the same cost level. However, remote login to a desktop/cluster at the lab alleviates this somewhat. Another downside is that this model makes it too easy to work from home, resulting in reduced socializing. I personally find it be usually distracting, but have committed to being in the lab during certain times in the week, and take an active role in creating idea-exchange opportunities with others.

I can see the following barriers to wider adoption of the above approach:

• Funding for laptops is problematic. I understand there are restrictions/prohibition of using grant funds to pay for laptops, but not for desktops. Your Dean would have to address this problem somehow, as not all graduate students can be expected to own a laptop, esp. a high-performance one, which might be necessary for research. I urge influential readers to use their positions of power to lobby federal agencies to loosen this restriction. The argument that every researcher needs a laptop to present one's work at conferences or seminars might be persuasive.
• Path dependence on earlier paper-based methods. If you already have stacks of papers on your desk, it would be difficult to digitize them, and remove all sources of them. I started at a time where being "fully-digital" was a possibility. In that respect my situation is unique. However, trends in academia, and outside, point to increasing cultural connectivity and digitization. So this model might work for incoming students, but would likely be difficult for existing ones.
• Requirement for some technical sophistication. I have a software development/computer engineering background, which has made it easier to be fully-digital. So the above approach might work for students in CS, engineering, or other computationally intensive fields. In less computer-dependent fields, students might need additional support for the transition. Your Dean could consider hiring a department level "all things IT person" to provide such support. But again, with cultural trends in mind, it seems that students across many other fields appear to be becoming more technically sophisticated, so this problem might disappear over time.