3 ms·
>A next problem is that software replaces physical objects by small computer files. Such files have the twin attributes of what economists call public goods: >
by i_am_proteus 8y ago
>A next problem is that software replaces physical objects by small computer files. Such files have the twin attributes of what economists call public goods:
>The good must be inexhaustible (my use doesn’t preclude your use or reuse).
>The good must be non-excludable (the existence of the good means that everyone can benefit from it even if they do not pay for it).
Software may not be inexhaustible, but it generally does require regular maintenance and updates, as well as support for the end-user. For me, these facts weaken the value of the thought experiment tremendously.
- colechristensen 8y agoMuch software needs much less maintenance than it gets. Or at least it could. Constant upgrade cycles are expensive and increasingly unnecessary. As a thought experiment: could you take the current generation of iPad and keep producing a 100% compatible version for 100 years? If so how much software maintenance would you need? Would that be useful? People are very accustomed to a fast upgrade cycle and so thing something a few years old is ancient, but it doesn't need to be like this. A whole lot of software could be engineered to completion and then left only to occasional maintenance requirements. It isn't. It could be.
- ineedasername 8y agoBut we aren't at the level of automation the author envisions as necessary to take us out of free market dynamics that underpin capitalism. Until we get there, software will need to be upgraded and maintained to continue the march in that direction. I work with ERP systems, and this is very much the reality. What we have now is significantly better than 20 or 30 years ago. While they are arguably still filled with cruft and bloat that has accumulated over that period of time, their ability to automate tasks has greatly improved, with enormous room for improvement still present. In the case of an ipad, your argument might hold if we were to hold hardware features stagnant. But better cameras and screens are obvious upgrade paths that will require software upgrades. Folding screens will undoubtedly require new ways of thinking about UI and UX with concomitant software upgrades. Your right, we don't need to keep chasing such hardware upgrades, or improvements to software defined features. However, if we don't, we also don't really reach the post-scarcity society that the author of the article posits either.
- skybrian 8y agoAt the very least, browser vendors would need to keep shipping new versions because web standards continue to change rapidly. It's already the case that some web developers complain about Safari not keeping up. I suppose someday web standards could be considered "done" but we're nowhere near a consensus on that. I couldn't see Chrome ever doing this and imagine the complaints if someday they said "okay, good enough."
- p1mrx 8y ago> could you take the current generation of iPad and keep producing a 100% compatible version for 100 years? TI calculators sort of follow this model.
- pdonis 8y ago> A whole lot of software could be engineered to completion and then left only to occasional maintenance requirements. The iPad isn't software, it's hardware. It's very hard to predict what hardware upgrades users might really want a few years from now. So even if all the software on the iPad could be engineered to last a hundred years, it would still need to be updated every time the hardware changed.
- pdonis 8y agoMaintenance and updates are just more bits, which can be copied just as easily as the original bits were, so the same two attributes apply to them. User support is different, yes. But user support is a service, not a product; users will pay for it if they gain enough value from the support to make it worth it. But for many pieces of software, the set of users for whom that is true is small or even empty.