I would consider Accelerando by Charlie Stross, whcih is, I think, a good complement to The diamond age. And maybe have a look at the video's of Robin Hanson for the more "over the top" futurism, very suitable for a critical review and to get to know the more crazy side of futurism (https://www.youtube.com/watch?v=vvjCJE-N34k).
Seems to me a dangerous tactic to base this kind of prediction purely on historical data. There is no way to prove that something that was the case will always be the case in the future. The problem in essentially, I think, the gap between what humans can do and what machines can do. What humans can do and machines not defines the "skillspace" left for humans. Now that space is contracting very fast, for the first time in history. That may well break the "law of always more new jobs". Another related issue is the character of the revolution. It is general purpose machines that replacing humans. Not something like hammers and printing presses. It seems to me that not only will the gap between humans and machines get smaller, it will shrink at an increasing speed. Think about it. A printing press replaces printers, but a general purpose cognitive machine added to a human like body replaces maybe 95% of everything that a human can do. Big difference, IMHO. So every imrovement of the general purpose cognitive machine will replace jobs over a broad spectrum of industries and skillsets.
Probably a lot more detail is needed to give a useful answer to your question. However, there are some issues not mentioned yet. First, what is the budget for system administration and maintenance? Is there a budget for that at all? I do (volunteer) system administration for a couple of small human rights organisation (about the same size as yours). They are cash strapped and don't have the money to pay for a system administrator, or to contract for the work as needed. The rely on volunteers, and these are really hard to find. So, ask yourself what kind of expertise is available before you decide on a system. makes no sense to design a superb system when you have no one to keep it running. Hardware is generally kind of uninteresting. I would go for wireless (RADIUS) for as many clients as possible, and don't buy unnecessary powerful PC's. Waste of money. One system I build was based on Google Apps (Education license available for non-profits) for mail and remote access and a local NAS with LDAP that synchronises with Google. Create an account locally, a Google Apps account will be created automatically. Clients Windows XP / Windows 7. What makes this a good system is very low maintenance, easy deployment (everybody knows Gmail, etc) and good support for remote users. Office staff can deal with almost anything needed to keep the system running. For the NAS I used a Intel SS4200 NAS with 4Tb raw storage and installed a core version of Ebox (zentyal) on it for filesharing. LDAP and RADIUS. Web interface, office staff can deal with that. The second system is a MS Small Business Server 2003 with about 12 clients. That works well, problem is you need someone who knows SBS, and can handle sysadmin tasks. (And no, in my experience most people working for non-profits can't handle that). Licenses for SBS (and Windows) can be purchased through the Microsoft program for non-profits. it's cheap, and the money should be no problem. Mail runs on the SBS server, remote acces to the office PC's too. be ware that security is a bitch in this setup. Much harder to keep it safe that the first system. You say you want to run the website from the office. I have no idea why you would want to do that. It's a headache. If you go the Googel Apps way, use Blogger for a website (if simple is good enough) or create your own website with Joomla (host it somewhere) and handle authentication for your website through Google Apps.
An issue that is completely ignored in the discussion of the 50-button remote control is the fact that a lot of those buttons are really an answer to the horrible complexities of modern technology. I own a DVR (a pioneer) it supports something like 15 different (video) file formats (not counting the numerous varieties of these formats) the manual is full of exceptions and warning of things that work (or not) for a particular file format. the question from an ignorant consumer would no doubt be why we need so many file formats just to play some video at home. next is the programming functionality,which comes with an huge numberer of options. included is an electronic programming guide, which looks very nice and actually works, apart from the fact that the guide only knows about the published program times,not the actual times,This makes the thing completely useless. then there is the fact that the machine supports about 10 different recording media, all possible varieties of CD-R and DVD-R. Same story of warnings and exceptions. then there is support for a number of picture aspect ratios, which may,or may not be automatically recognized based on the transmitted signal. And then there is a complex guide about how to copy a file from one medium to another. Same warnings and exceptions. And copy protection. And you think it will just recognize a DVD formatted on a PC? No of course it will not. So what is the conclusion? What we see is a lot of useful things implemented by the engineers that designed this particular PVR. The problem is that many of the features that they try to get working depend on other infrastructure being there, like a reliable electronic program guide that also works as aprogram is late, like a simple video file format standard, good standardized media etc. And the simple fact is that these things are not there. And as long as these standards are not set and widely implemented, the mess will stay, and people will get confused. Compare for a moment with the situation round the time that electricity was new, and voltages, connectors and system (AC/DC) changed from one city block to the next. i bet people were confused then,like we are now with allthis new technology that surrounds us. And the it makes no sense to perster the engineers with this problem because setting standards is most of the time more of a political than a technical process.