I was just over at Scott Hanselman’s Blog and browsing around some recent posts when I ran into this note about the demise of NDoc. NDoc of course is free tool to document .NET assemblies. Go ahead and take a look at the note…

This is pretty disturbing to hear. NDoc is probably one of the more widely used free tools on the .NET platform and apparently according to one of the developer leads there was grand total of 11 donations since the last version was released which was quite a while ago. Ouch…

I’m really sad to hear NDoc’s demise even though NDoc is one of the biggest competitors to Help Builder precisely because it’s a free product. To be honest, for pure .NET doc generation NDoc is a better tool than Help Builder (which also generates documentation from .NET assemblies and applications) primarily because NDoc is focused specifically on class documentation generation while Help Builder is a more generic tool that has many other features and allows modification and adding to the generated content.

But aside from that I know a lot of developers who use NDoc on a regular basis – less so now because .NET 2.0 support is not there and now apparently will never come. I’ve never used NDoc for anything more than testing, so I don’t feel guilty here, but I’ve worked on projects where NDoc was used and I can tell you that it’s an integral part of the development process when it is used.

All of this raises an important point I think. First and maybe foremost is that if you use ‘free’ products – especially if you rely on them – you should really support those products even with a small donation if the developers are asking for it. I know it seems easy enough to just ‘forget’ – after all it’s FREE, right?

I am guilty of this as much as the next guy. I use a few open source products where I could have donated and haven’t. I guess I too have a donation threshold folder of criteria that determines what gets my donation. <s>

But there’s the rub – when does something become worthwhile and how do you assess the value of a tool without a ‘Suggested Donation’ price? Ultimately when you’re given a choice you grossly underestimate the value of what the tool is actually worth and pay a fraction of what you would pay for a comparative product.

To make donations makes sense as it gives the developers at least a little incentive to move on the project and dedicate more time to it. As stated in the note if even a small number of NDoc’s user base would have put up a small donation the tool would have probably stayed alive.

But this is also something that bugs me about the whole open source ‘movement’. The point here is that ‘free’ which is the main draw that brings people to the products in the first place, rarely means free. Somebody has to work for the functionality and invest at the very least time into the process, so free is certainly not free for everybody. But labeling a product as open source to many has the immediate connotation of free and the release of any requirement for compensation. Compensation can come in many ways – money, but also supporting the product and contributing possibly, although that’s probably more difficult to manage for the core development team than anything else. But even so the smallest amount of users actually contribute in anyway.

And in the end the products are not really free, if they are begging for donations to keep them alive. This whole open source business of begging for people to support it is not automatically a better approach. As a customer I rather work with a product I paid for with a license because it's something you paid for and in most cases have at least some right to expect support for. There’s also a price that you either accept as reasonable or refuse by passing on buying. On a purchased product there’s also a chain of responsibility there (although with some vendors that’s not a certainty) that usually isn’t there in open source.

In the end that’s just another – ineffective – payment mechanism. So overall I can’t really feel so sorry for the developers of NDoc. If you create a free product don’t expect to get paid no matter how much value it provides. If you think the tool has enough value, well, then make it a product. See if it can actually make it in an open market with people paying what you think the value of it is. Of course that’s difficult going from free to for pay, but that’s one of the problems of open source…

The problem I have here is that we’re working of the honor system and the honor system in many cases just doesn’t work well. I don’t think that people necessarily are evil and wouldn’t pay, but the honor system isn’t exactly providing proper guidance. What’s it worth to you? $5 was thrown around as the value for NDoc. $5??? Are you kidding me – that’s what 2 minutes of billable time? Think of the value provided on your project. $20, $30? $150? I bet if you look at it in terms of value added to the project(s) that use NDoc it’s almost certain on the top end of that range.