Maybe you're providing IT for a company dealing with point of sale stations, thus rendering the idea of having a flash drive for each employee problematic at best, impossible at worst.
There are a plenty of reasons why it wouldn't be right to carry out this kind of plan. Yet at the same time, there are also countless companies that might genuinely benefit from taking the "dongle vs PC" approach to employee access to company resources. Allow me to put some emphasis on "might" here, as every workplace situation is going to be different.
User portability and access
As stated above, every company has varied needs when it comes to balancing out resources and security. But some of you might be wondering what the real advantages might be in working off of flash drives over that of actual thin client boxes? After all, I can run over to Google Shopping and pick up a thin client box for a mere $40. Clearly the pricing difference between that and a flash drive is fairly slight, no?
I happen to think allowing greater user portability has merit in the long run. With company policy and overall function allowing, I think it's worthwhile sticking with an easy to configure flash drive friendly Linux distribution.
And why not? If sticking to a portable method of managing workplace PCs translates into less dependency to any one computer, all the better. With that said however, I should point out that these drives would still be acting as thin clients.
Yes, the term "thin client" still applies. Basically the thin client becomes a flash drive concept instead of a workstation-related one. Yet unlike a traditional thin client hardware device, the flash memory used with a USB flash drive is truly portable. So you needn’t be shackled to an office to get your work done.
Flash drive vs Web thin client
The last item I wanted to touch on is the advantages of a Web thin client. Like the USB flash drive, a Web thin client offers plenty of PC independent portability. Another advantage with it is that the fear of losing a device is a moot point entirely. Your company computer is merely an address and a login.
The only real disadvantage to using this approach is that you still must use a storage medium if you wish to take data with you off site. Something like a flash drive, for instance. Because if you should lose network connectivity for some reason, you'll be without a local disk to store your project on. Outside of this, I would otherwise suggest that the differences between a flash drive driven solution vs. a Web thin client are going to be fairly slight.
The future is with the Web
Despite the localized advantages of a flash drive Linux installation, I see the long-term future being with Web thin clients. Thanks in part to the popularity of Web-based operating systems and a push to adopt "cloud computing" as the accepted way of doing things, the idea of using flash drives might seem antiquated. Some among you might even point out that using flash drives running Linux for enterprise use is too little, too late. This might be true to some extent, except that I've found that having the ability to scan documents, print my work or just running an email client (in view only mode) has its advantages when the company network is down. A Web OS, requires a network connection. The last time I checked, using a vanilla Linux distribution on a flash drive, does not.
To further recap my reasoning for giving flash drives a strong second look in the workplace, consider the following: flash drives offer data portability. It would also translate into reuse of existing computer hardware. Using a flash drive leaves out any concern over a network outage. While a company network outage could be an issue for accessing the company server, once the data is pulled into LibreOffice or another FoSS application, network connectivity suddenly doesn't become as important. This is merely my take on this idea, your mileage may vary.