PB's law: "Sooner or later every software will be FOSS or perish."
Posted: Mon Sep 27, 2021 9:33 pm
I claim the following:
It's simply a lot of effort (developer/admin/tech-staff resources) to support software over time.
Therefore, any piece of software that stays proprietary can only be kept alive by its proprietor. Even if this is a big and wealthy company, the income generated by that version of software will diminish over time to a point where even major companies drop support of tool X in version Y.
Especially considering that over time, each developer/company will pile up more things to maintain & support.
If being FOSSed:
And its userbase is rather...
small: it's very likely that someone of its users will try to find a way to keep it alive. Usually by either adopting the project or finding a way to fund developers doing so.
Usually, it may become adopted by a company - often ones that specialize in certain domains (like preservation, AV, retro-hardware, custom stuff, etc)
large: If a software has a large userbase, it's very likely that a community will form around it, transforming it into a supportable FOSS project, possibly not only maintaining it as-is, but improving it beyond its original purpose and value (e.g. Mozilla, RPi, WRT54GL, etc).
Usually, it may become adopted by a company - often ones that either succeed as startup thanks to a existing code- and user- and knowhow-resource, or happily increase their profit by including FOSS tool X in their portfolio.
none: If for some reason, tool X isn't even interesting enough for an individual to support/maintain as a whole, having its code out in the open, it will very likely become a source of code snippets/libs for other projects/tools. This is in fact a digital recycling flow which happens naturally in FOSS ecosystems, whereas having a copy of proprietary binary X is like having a piece of molded plastic that you can use as-is. Or not at all. Proprietary software is in fact like plastics, whereas FOSS is like bio-degradable, regrowing resources. Plastic may seem cleaner and more modern, yet there is a reason why certain materials are timeless and increasingly preferred.
If staying closed:
Sure, proprietary closed source software/devices can be kept alive beyond their shelf-life, but knowing how software works and considering the increasing frequency of updates of everything around it that it interfaces with, experience shows that doing so can get quite tricky, cumbersome or in some cases even impossible (without the ability to patch/adapt it).
Sure, one may engage with the original vendor/proprietor and ask them for help, and may actually get infos or even patches or inofficial versions. I've done this quite a few times over the years now, and not only is it getting harder to almost impossible to even be able to contact the vendor, the willingness to support not-profitable-anymore versions seems to be noticeably decreasing. Additionally, the knowhow about former products is often linked to individuals who work(ed) there. In my experience, dealing with software/hardware in the archival preservation sector, I've received the reply "sorry, there is noone here who knows that anymore." more than once.
And in institutions, it may require a data/knowhow repository to find/retrieve/apply these fixes over time, too. Been there. More than once.
Sure, there are things like emulators and virtualization, yet we already see that with growing complexity and inter-dependencies of code/programs, creating a simulated environment not only gets harder, but you have to consider that the vm-host itself will pile up legacy support requirements. Since the vm-host environment itself is software, PBs law applies to it, so it will become FOSS - or again, perish in that version.
Therefore my hypothesis stands to be challenged
It's mainly based on observation from different IT sectors, like System administration (SysAdmin), development, desktop - including AV and gaming - as well as embedded systems."Sooner or later every software will be FOSS or perish."
It's simply a lot of effort (developer/admin/tech-staff resources) to support software over time.
Therefore, any piece of software that stays proprietary can only be kept alive by its proprietor. Even if this is a big and wealthy company, the income generated by that version of software will diminish over time to a point where even major companies drop support of tool X in version Y.
Especially considering that over time, each developer/company will pile up more things to maintain & support.
If being FOSSed:
And its userbase is rather...
small: it's very likely that someone of its users will try to find a way to keep it alive. Usually by either adopting the project or finding a way to fund developers doing so.
Usually, it may become adopted by a company - often ones that specialize in certain domains (like preservation, AV, retro-hardware, custom stuff, etc)
large: If a software has a large userbase, it's very likely that a community will form around it, transforming it into a supportable FOSS project, possibly not only maintaining it as-is, but improving it beyond its original purpose and value (e.g. Mozilla, RPi, WRT54GL, etc).
Usually, it may become adopted by a company - often ones that either succeed as startup thanks to a existing code- and user- and knowhow-resource, or happily increase their profit by including FOSS tool X in their portfolio.
none: If for some reason, tool X isn't even interesting enough for an individual to support/maintain as a whole, having its code out in the open, it will very likely become a source of code snippets/libs for other projects/tools. This is in fact a digital recycling flow which happens naturally in FOSS ecosystems, whereas having a copy of proprietary binary X is like having a piece of molded plastic that you can use as-is. Or not at all. Proprietary software is in fact like plastics, whereas FOSS is like bio-degradable, regrowing resources. Plastic may seem cleaner and more modern, yet there is a reason why certain materials are timeless and increasingly preferred.
If staying closed:
Sure, proprietary closed source software/devices can be kept alive beyond their shelf-life, but knowing how software works and considering the increasing frequency of updates of everything around it that it interfaces with, experience shows that doing so can get quite tricky, cumbersome or in some cases even impossible (without the ability to patch/adapt it).
Sure, one may engage with the original vendor/proprietor and ask them for help, and may actually get infos or even patches or inofficial versions. I've done this quite a few times over the years now, and not only is it getting harder to almost impossible to even be able to contact the vendor, the willingness to support not-profitable-anymore versions seems to be noticeably decreasing. Additionally, the knowhow about former products is often linked to individuals who work(ed) there. In my experience, dealing with software/hardware in the archival preservation sector, I've received the reply "sorry, there is noone here who knows that anymore." more than once.
And in institutions, it may require a data/knowhow repository to find/retrieve/apply these fixes over time, too. Been there. More than once.
Sure, there are things like emulators and virtualization, yet we already see that with growing complexity and inter-dependencies of code/programs, creating a simulated environment not only gets harder, but you have to consider that the vm-host itself will pile up legacy support requirements. Since the vm-host environment itself is software, PBs law applies to it, so it will become FOSS - or again, perish in that version.
Therefore my hypothesis stands to be challenged