My mom is neither an educator, nor a technologist, but she has taken to reading my blogs anyway (group, "Ahhhh", please). She actually runs a gourmet tea and spice company based just a few miles from Microsoft corporate headquarters (weird, huh?). She'll occasionally send me an email asking for clarification on some point or another or to say, "It's a good thing I read your blogs or I'd never have known you were on vacation" (oops). At any rate, she sent me an email the other night after Marc Wagner posted his entry, "Evaluating thin clients from a position of strength," and asked, "What the heck is a thin client?"
On the same note, a few of the staff we had in training last week walked into our computer lab (all thin clients) and asked about the little tiny computers. So I figured it was time for a quick thin client primer, as I firmly believe that there is a lot of value in server-centric computing if it is used in the correct context and delivers the correct applications to users.
In their simplest form, thin clients are very much like the old green-screened terminals that you or your users may have seen years ago in a variety of settings. Banks, universities, and other institutions used them to access big, expensive computers running specialized applications. They are just very simple machines, often with little power or use on their own that access much bigger more powerful computers. Nowadays, they often look and feel just like Windows or Linux computers, however, since they have a much prettier interface. They are still accessing big, powerful computers (aka servers), but these computers send images (or "screen refreshes") to the thin clients to make the users feel like they are sitting at a full-blown Windows/Linux machine. In fact, the thin clients allow users to access "sessions" on the server as if the server had several keyboards, mice, and monitors attached directly to it, with every user looking only at their own data.
They have several advantages:
- They are cheap (in and of themselves - there are other associated costs, though)
- They let people like me administer just 1 computer (the server) instead of all of the computers that users access (the thin
clients); software installation, virus protection, etc., only happen in one place
- They have little value on their own, so people don't tend to steal them
- They are easy to roll out and, if done right, operate very much like a "real computer" for users
- If you have any interest in Linux, free software like Edubuntu and LTSP make implementing thin clients fairly easy and free (from a software perspective, at least).
They do have several disadvantages that must be evaluated:
- Graphics/processor-intensive applications, like CAD, Photoshop, or Maple tend to run very poorly in a thin client setting; they are best limited to basic productivity applications or web surfing (or the specialty applications that Marc Wagner mentioned). As it turns out, my mom actually had thin clients in her office accessing accounting and order-processing software.
- They require a high-end server (or servers) to run effectively (very small implementations of two or three machines in a classroom can use a fast desktop computer, but a full lab requires some serious backend hardware)
- They tend to eat up bandwidth on your network; if you don't already have a solid infrastructure in place or aren't planning to install a well-planned network, thin clients are not for you (this doesn't apply for the small classroom setting, where users essentially share one speedy machine with a couple extra terminals).
So there are thin clients in a nutshell. They can have a great deal of utility and value in a K-12 setting and new tools are maturing rapidly to reduce costs and simplify rollout. The Edubuntu website has some great introductory resources as well.