Enterprising the gnuniverse, a story about business.

There I was with a friend of mine, we were chatting about people developing software. Our focus turned onto small companies, and their needs, and the edge cases for selling into them, either in philosophy, or service, the "FLOSS" software solution.

We thought about it and working on the basis that backups are interesting, and they are seen as a necessary evil, how could FLOSS get that sale?

This created a conversation around the what it is that a business wants.

They want to spend as little as possible for as much as possible.
We discussed File and Print services, there are several ways to do that, samba if you have a non GNU/Linux network, and CUPS and NFS if you do have one, you can also use samba there.

What is it that the business wants. It wants as little pain as possible, religion is a pain, you are in love with an idea if it is there for it's own sake, that does not sell. What sells is if the people using it benefit from it, mostly by having an easier life as a result of this.

With this in mind if you are going to build a small business server here are a few things to think about.

Plan a basic model, , a mid range box, and then a top of the range box, each offering should build on the one below.
Sample offerings:
A bronze offering, perhaps a basic file and print server, capable of doing samba, with some file sharing for the office network.

Then plan a silver offering which offers the above and might also offer a wiki.

Then perhaps the gold offering would be a single sign on solution, again something in the domain area, using LDAP or yellow pages depending on the business.

Each version should allow you to easily move to it, and revert back to whatever business process that the business had before. This allows for testing, and reversion if something does not work out for the company trying this implementation.

It is vital to the success that the business can pick and choose any or all of the parts of the gold offering.

Consider also off site incremental backups, they cost power and some kind of broad band on the remote site, a directors home perhaps. What, not how it is done is of interest to us here, this is only suggested as one possible way. OpenVPN, being used and the box doing an incremental backup. Those who would decry this might suggest that there is too much data to start out with. Part of the process here could be to backup onsite first before deploying the offsite box.

Now you are offering things that make sense to business.

However the easy thing is to say, we can do this with tools X or Y and Z.

Well the hard thing is to say we do it with X, and allow people use Y or Z themselves. This is why you see "supported methods or software" from some companies. The fact you can do "it" whatever "it" is, twenty ways other than the supported way is the way forward here.