Networks are negligible latency, your company is still run by idiots and trying to save money on hardware by paying more for the same performance, but it's at least technically feasible.  Internet based computing will never work, no one wants to do their word processing with 200 ping.  By the time it's reasonable, you'll pay a couple hundred bucks for a computer that fits in your hand and still runs everything in existence ten years later without any problems.

1-- When was the last time you rented (or bought) a car twice to drive it straight to the scrapyard eventually anyway?

2-- I've spent 12ths of a cost 12 times in a year for worst values.

3-- Pay per view on cable TeeVee wasn't enough of a punch bag for equalizing value with privilege?

4-- No, and i'll sue the Industry for taking away my legitimate control BY choice.

5-- Both.

6-- You get what you can afford wherever & whenever already. Fiberwires connect, accessibility is a potential, infrastructures deploy where opportunity justify the means.

7-- Half a second, maybe less.

Here's what really has me worried: We are not even prepared to handle the server-side networking. Most of my fellow "end users" that I have discussed this with are convinced that the network will be so slow that our productivity will suffer while we sit at our shiny new terminals and stare at little rotating hourglasses on our screens. Additionally, when a server goes down we will not be able to do anything at all.

There seems to be a serious breakdown in decision-making in large organizations these days. It does not seem to matter whether these are governments, businesses, political parties, or "other". Therefore, despite the fact that server-side anything on a massive scale is not feasible right now, I have little faith that the people in charge are not going to dive into it anyway.

Your company is just missing a couple of logical steps.

Some airhead with insufficient tech knowledge is looking at those desktops he's paying for, thinking about how you're only using 10% of it most of the time, and he'd save a fucking fortune if you were just sharing processing time with the rest of the users.

He's missing that the processing power is more expensive at larger scales, takes more knowledge to maintain, breaks more often, breaks everyone together, and requires a massive air conditioning system with a quad backup system to maintain reliability even without hardware or software failure.  The costs in power consumption for large mainframes aren't pretty, especially since you can't just turn them off when the shift ends.

He's also missing that he keeps buying those terribly expensive pieces of equipment because the minimal performance gains over the old ones were enough to justify their existence, gains more than lost by network latency alone.

What your company has done is regress to the model of the 70s and 80s by having semi-dumb terminals hooked to a central server. This will always work better in conditions where the majority of the work is done on the terminal using terminal resources with little or no interaction with the server. Unfortunately, most corporate database systems don't work that way and you will indeed be limited by network speeds.

Frankly, computing power and memory are cheap enough I'm surprised that anyone would look at the old systems model, but then again companies often to stupid and silly things. I wish you the best of luck.

All online games use a variation on this system model, the only difference is the degree of centralization and dependence on network speed.