Re:The Cathedral and the Bizarre
by zozbot234 on Saturday December 14, 2019 @09:14PM
(#21794007)
> But above all this is the sheer waste of human effort in terms of the production of rotting software in repositories.
Yup, what about the amount of human effort that has been wasted in the course of producing off-the-shelf proprietary software? How much of that proprietary stuff is now entirely useless, for some reason or another? It's just so weird to make an engineering-focused argument for proprietary stuff, when the only case that's even remotely plausible is about production incentives, and everything else hugely favors FLOSS.
As an aside, I'm pretty sure that, in the medium-to-long run, we'll manage to fund FLOSS development to an entirely satisfactory extent via crowdfunding, and that while saving a lot of dough compared to what it would take to fund proprietary development in the same way.
But in fact, even his point about support revenues is misguided. Well-designed and easy-to-use software does not really draw fewer paid-support arrangements than software that's hackish, unintuitive and the like. The nees for paid support is almost entirely an organizational one, so it matters zilch how easy the software is to support in reality; it's far more important that the organization can point to that support contract if any such need should arise.
Yup, what about the amount of human effort that has been wasted in the course of producing off-the-shelf proprietary software? How much of that proprietary stuff is now entirely useless, for some reason or another? It's just so weird to make an engineering-focused argument for proprietary stuff, when the only case that's even remotely plausible is about production incentives, and everything else hugely favors FLOSS.
As an aside, I'm pretty sure that, in the medium-to-long run, we'll manage to fund FLOSS development to an entirely satisfactory extent via crowdfunding, and that while saving a lot of dough compared to what it would take to fund proprietary development in the same way.
But in fact, even his point about support revenues is misguided. Well-designed and easy-to-use software does not really draw fewer paid-support arrangements than software that's hackish, unintuitive and the like. The nees for paid support is almost entirely an organizational one, so it matters zilch how easy the software is to support in reality; it's far more important that the organization can point to that support contract if any such need should arise.