A Correspondence on 'The Problems of Open Source'

This correspondence came from an FOSS supporter. I publicise it because it was about the only intelligent response I received to 'The Problems of Open Source' from the FOSS camp and it reveals no confidential information. I've not included the email of the sender to respect his privacy. His comments are italicised and my text is not. Replies to replies etc. are indented.

Hi,

I'd like to give some belated feedback on your open source article. I'll focus on your main points here:


1. Good software arises when one or more very good programmers work closely full time together over a period of time developing, maintaining and improving it.

Yes and no. It doesn't require full-time attention or very good programers. Just good programmers and enough time to do it. Depending on the project, it can be done in spare time as a hobby as well.

The problem is that in a commercial environment, there is not much incetive to get good programmers to work together intensively for a long time, because that costs too much. Rather, they get the cheapest possible programmer who can just handle the job, and push that person to do the job as quickly as possible. In my experience, the quality of the software I'm writing together with my colleagues at work is normally lower than the open source software I work on since most employers don't want to invest in good software. They are OK with sofware that "just works" with the emphasis on "just". There's even a perverse incentive not to make the software too good if it's sold to a third party, so they will buy the next version with much needed improvements.

I wouldn't altogether agree here.  It depends on the software but generally there is a steep investment of time in grokking any large piece of software and if you have a good team with experience it is hard to replace.  I've seen software be degraded by people arriving on a project with ideas but lack of knowledge.  Hobbyism doesn't really provide the degree of financial support necessary to underwrite professionally produced, documented and finished software.

2. If you give your software away under GPL (free as in free speech) its very difficult to charge for it and so it ends up being free as in free beer. Ditto for FOSS.

Yes and no. Usually the software itself is free. Usually, there is some indirect way to make money for it involved. Often that is sufficient.

This is the nub of the article; often it is not.  The point is that most OS people start with saying 'OS is great' and then hunt around for some economic model to sustain it.  Funding should not be an afterthought.

3. Hence if you want to maintain yourself and/or your team, you need a viable economic model and FOSS/GPL does not always supply it. Their standard models are support and advertising.

No. There are other models as well, such as corporate cooperation. The reason Linux has taken of as the Unix operating system of our times is that many corporations wanted a Unix, but before you had tons of incompatible Unix-alikes from different vendors, like HP-UX, AIX, Xenix, etc. It makes more sense, even for those commercial entities to invest in a jont venture like Linux than to pump money in competing non-compatible systems. In other words, standardisation cuts costs enough for it to be attractive to invest in. Open source guarantees standardization much better than commercial software, hence investing in open source software is attactive for certain problem domains. Other models for generating income from open source not mentioned by you are ransomware , freemium, and other ideas yet to be discovered.

I think Linux was covered and I pointed out it was a special case of something complex enough to make money through services.  But this does not suit many development projects. Ransomware is thought of as malware.  Freemium doesn't really work well with OS.

4. Since the economic model for FOSS is not generically viable it often relies on corporate and taxpayer money to sustain itself.

False, but even if it were true, why would this be bad?

Well I think if OS software mostly cannot internally generate revenue; if it is dependent on the public purse or takeovers of abandoned software then really this needs to be flagged by OS people.  What they are saying is 'Our model is by itself incapable of financially sustaining development w.o. external funding or giftware'.  That is a major weakness and needs to be flagged.

Universities and corporations do decide on what projects to invest money in and which ones not, and they invest in open source software because it's profitable for them in the long run. It doesn't take much thought to figure out why. For many domains, it's much cheaper to finance a few enthusiasts and get a server OS or a speadheet package out of it than to fork over endless big license money to commercial entities who want to keep you in the upgrade treadmill. This also explains though why there is not much progress for, say, Linux on the desktop, since that's not a domain that the powers that be find very interesting to invest in. Compare this to how Android has exploded in popularity with millions of activations per day... "Who profits" indeed! :)

This is a fair point; but it comes under the 'too complex to use easily' provision which I mentioned in the essay. Sure if your software is hard to fathom or use then there may be a percentage in being paid to maintain or develop it. This case was covered in the essay.

However note that the OS developer's position is precarious. If the corporation decides to make the software in-house or decides that they understand it well enough or that it is stable enough not to need maintenance then the OS developer is likely to be dropped in the next round of corporate spending cuts. BSD/MIT licenses give you very few rights over your work.

5. The lousy quality of FOSS is disguised by pointing to the few success stories. So much of it is (badly) copying commercial ideas.

False. I'd say not only FOSS, but MOST of ALL software is of a lousy quality, and FOSS often tends to be somewhat better where it matters, though, admittedly, not always. This is of course due to lack of resources in both cases, but for different reasons. As for copying ideas it happens in all ways. I know just as many instances of commercial software copying open source as vice versa!

I think there are specific reasons why FOSS is often worse, because there is often no financial incentive for people to maintain and improve it.  Any commercial software at least begins with this incentive.  However the 'if you don't like it too bad' attitude is symptomatic of a lot of FOSS work.  I'm not being paid so why bother?  Also the atrocious quality of the doc because OS hackers do not like writing up their work because it's not 'fun'.

6. Corporations like FOSS because it allows them to exploit ideas without recompense.

Yes, and no. Corporations are smart enough to know that FOSS enthusiasts need to eat too, and often will hire them or fund them.At least in cases where it makes sense economically to said corporations.

To the minimum degree to get what they want which may be zero if they can get away with it.  Or they may prefer to make the project inhouse and hire their own having grabbed stuff for free.  Who says they have to be nice or fair?  

7. We do not need the Open Source Movement or the FSF.

False. While I do think that the FSF is a bit extreme, I think they embody a positive spirit of giving that ultimately is better for mankind than the short-sighted model of maximum profit at the expense of anything else. While economy is important, the current systems we have are not the end all be all. The problem is that the state of the economic science is dismal, often it is not better than a religion where everything is taken on faith and facts are ignored (see, for example, the Austrians). Now why would that be so?

Well that begs the question as to whether OS is a good thing. I've argued it may not be. The message of the FSF is that closed source is evil.  The OS site simply refuses to engage in any discussion on the weaknesses of their model.  This is not to say giving is wrong, but that it is an individual choice and not a matter for organisations to promote - particularly if they are pushing something that is flawed or which they cannot justify.