The reason for comparing costs came from the purpose of Kindred Hands. I wanted a direct route to customers without our taking commission, and the initiative was intended as not-for-profit. Continuing costs around the first software route led me to explore alternatives. The purpose and the commitments needed to run the service had to belong in the same explanation.
Moving to open-source tools did not make every future task disappear. WordPress itself is open source, and the initial route involved particular multi-vendor options. I was responsible for bringing the alternative setup together with guidance. That was a different arrangement of software and work. The useful question became what the initiative would have to keep doing after the first version, including the work that does not arrive with an invoice.
The reason for looking at alternatives was practical. This is a not-for-profit effort, and continuing costs matter. I want that change of direction to lead to a manageable setup, rather than a different arrangement that only appears inexpensive because its difficult parts have no amounts beside them yet.
Consider hosting first. The software needs somewhere to run. What is appropriate depends on the actual components and how the service is used. I can’t choose an amount in the abstract, and I should not make a historical price out of a current offer. But the absence of a final figure does not mean the requirement disappears. It belongs in the comparison as something that still needs a reliable answer.
Then there is maintenance. It includes more than looking at a finished website occasionally to make sure it still opens. Somebody has to understand updates, notice problems and know which part to investigate. A system can keep showing its home page while a less visible task has become confusing or stopped reaching the next person.
Here is a choice rather than a result from our setup. In this case, one component saves a licence charge but requires much more attention to keep its connection with another component working. That might still be worthwhile. It might not. I can’t answer by counting the charge on one side and ignoring the time on the other. I have to ask who can do the work and how often the organisation is likely to need that help.
Volunteer time is particularly easy to make invisible. If I spend an evening understanding a tool, no invoice appears just because I used my time. The organisation may not need to pay for that evening. It should still know whether the arrangement depends on repeated work of that kind. An unpaid task does not become a task that takes no effort.
Support for the people using the site belongs here as well. If an account or instruction is confusing, somebody may have to explain the same step repeatedly. Improving that step could be more useful than adding another feature. I can’t put a precise saving against it without records, but I can recognise it as part of what keeps the service usable.
There’s also the question of what happens when I’m not the person looking at the setup. A choice may seem simple because I remember the chain of explanations that led to it. Another person will not have that chain in their head. If the arrangement only makes sense while I’m standing beside them, the missing explanation is still work left to do.
This does not make open source a bad route. In fact, it is a reason to take the route seriously. I’m not just looking for something without a price; I’m choosing work that can be understood, shared and continued. Good documentation and a clear division of jobs can matter as much as whether I can get a first screen running quickly.
I want my mentor's judgment on the assumptions I can’t assess confidently yet. Which parts are ordinary to look after? Which parts am I underestimating because I have only seen the simplest case? There’s no embarrassment in needing help with those questions. The mistake would be leaving them out of the comparison because the licence looked easy to explain.
A blank in a cost comparison should mean unknown, rather than zero. I want every number to have a reason and every uncertainty to remain visible. A less tidy comparison can be more useful if it shows what the decision actually depends on.
I can be honest about a cost I can’t yet calculate without giving up on the comparison. Writing the question down is useful progress if it leads to the right information. Pretending the missing figure is zero would make the page look more finished while making the decision weaker. I would rather leave a visible gap than build a conclusion on one I have hidden.
The useful version of “free software” gives me a starting point and more freedom in how to build. It does not finish the work for me. I’m beginning to like that distinction. The work of making the pieces fit is why I wanted to do the setup in the first place; I just need to remember that the responsibility continues after the first successful-looking page.