This paragraph relates to a thought I've had lately:
What is the software "end game"? There has to a point where there's nothing else worth adding to most commonly used software. Obviously we're in a climate of rapid advancement and meta shifts currently but it seems as though that will inevitably end at some point. Proprietary software that can profitably leverage personal information is bound to hit the market first, but even if it takes 50+ years you have to imagine that equally competitive open source and freedom respecting alternatives will eventually become available.
I would dare say that there's no end game, as long as we are making money out of it, without that much effort. We are terribly good at ensuring our job, at the expense of the end user.
We will keep doing software, even after the point where there's nothing to add. I mean, we already do so, right? How many features are added because engineers think it would be cool? Or, even worst IMHO, how many features are added because engineers think they are what users want/need? How many releases look like they were made just to keep programmers busy (and happy)?
We should be aware of this: most of the time, we are working for ourselves. We are even entertaining ourselves. How much % of the effort spent on a given software is founded on solid and systematic analysis of end user needs?
Unikernels could extend the shelf-life of applications, reducing the need to comply with OS pressure for upgrades. Mature and battle-tested applications need security upgrades and the occasional minor bug fix.
Imagine the societal chaos if buildings rerranged themselves without warning or just disappeared ("end-of-support"). We are missing an analytic framework to calculate the economic benefits of stability and costs of pointless change.
Windows XP has proven that customers will pay for not changing what works perfectly well. With virtualization to provide drivers for new hardware, mature OS/apps can live for decades.
Microsoft Word effectively reached this point many years ago. They keep making new versions, people keep buying new versions. It's by no means the only example.
Who believed that?
You discontinue any sales of permanent licenses to your software, charge users $50/month to access it, pitch it as "Cloud" because clouds are the cool new thing, and hope people just suck it up and pay because your products are the industry standard and there aren't good alternatives.
But seriously, Adobe and Autodesk are textbook cases of this. People frequently stuck with years old versions of their software because it did everything it needed to. Development of 3ds Max has slowed to an absolute crawl over the last 5-10 years, with speculation that Autodesk is deliberately trying to kill it in favor of Maya.
But if you're stuck using it, you get to pay $185/month to support them not working on any notable improvements. Perpetual licenses to nearly all of Autodesk's products will no longer be sold after January 31st 2016, and Adobe has already made the transition.
IMO, the point at which the software can "understand" human articulation and make improvements; improvements to itself, to the system it operates within, to the world we live within. I'm thinking along the lines of the Star Trek universe as presented in TNG and later. They just ask the software for something and it complies. Human intelligence and creativity are still required, but no one has to "write software" any longer.
If you limit your question to current technology (I mean those things that we have now, and we can foresee - not those that will require some unknowable leap in scientific understanding), then the answer is 'no time soon.' This is because every founder, every manager, every customer has some "unique" change or feature request and some other human will have to build it.
Zawinski's Law
I guess when we reach the point where everything can read email (pretty close, if you replace email with 'receive messages'), there will be another function to be attained by every software with their newer versions. Maybe something like having its own neural net to adapt to the user.
"Version 12 of our monitor calibration software now can learn when you actually want your monitor calibrated, and when you want it somewhat offset, for those mornings where there's no work to be done and you just want to read email, which, by the way, can be read on the same monitor calibration software you've come to trust since version 5!"
The "software end game" looks a lot like the show "How It's Made". Frankly, most of that is just PLCs, which are barely computers at all.
As to phone/desktop/laptop applications which do things for people, I think we're largely at satiety. That's always a dangerous prediction, but it's been at stasis now for quite some time.
Software is relatively easy. Networks effects are hard. And there's no indication for free/open projects to be good at achieving network effect and stealing network effects from incumbents.
The most exciting part of working in technology in general, and software in particular, is that you get to be at - or at least better understand - the cutting edge and see those changes before they become more broadly accepted. For example, machine translation used to be cutting edge AI research in top secret intelligence projects, yet today you can translate entire webpages in Chrome with right click -> "Translate to English". For free. Hell, your portable Star Trek-like computer can take a photo of the Japanese menu for you and convert it to something understandable, then translate your response back to Japanese for your waiter. The Babel Fish is real! Given a bit more time, some better brain-computer interfaces and a larger corpus, we might well be able to converse in any language without learning it...
Software is this generation's equivalent to steels and mechanical engineering during the Industrial Revolution, and nuclear weapons and power in the 20th century - we have conquered the physical realm pretty thoroughly but have only made baby steps in knowledge compared to what is possible. Even neuroscientists will admit we know relatively little about how the brain really works. Just look at statistical learning: the field has only really blossomed in the last 15 years or so.
The reason for proprietary software and IP is always the same: to generate via extremely expensive R&D leaps in knowledge and products that allow the researching company to reap outsized returns. Yesterday, it was operating systems and office suites. Today, it's applications that need things like massive GPU clusters for deep learning, or knowing how to scale complex services (like Google Image Search) to billions of users, or weak AI (self-driving cars). I would love nothing more than a glimpse to the technology of 2060 or 2600, although I doubt I would be able to grasp even a fraction of what it represents. Just knowing the problems they are trying to solve then, would be fascinating.
A medieval peasant most likely had more leisure time than you. Having a local technological maxima doesn't mean "constant progress".
Only in technology.
Germany, for example, was a much better state ethically in 19th century than in 1914-1945 (or numerous other 20th century examples).
And it makes no sense to say that 2015 music is necesarrily better than 18th century music or 1960 music, or 1970 music, or that a composer, just because of being born later, is better than Bach or Mozart.
The Whig view of history: https://www.historians.org/publications-and-directories/pers...
See also Fukuyama's "End of history".
Hell, you could almost make the case that the Windows operating system became a "baked" product when Windows XP came out and you no longer had to reboot your computer on an almost daily basis. I mean, that's why you'll hear howls and yelps from grandma and grandpa when some virus eventually comes to polish off Windows XP. Until then, you'll only get it from their cold, dead hands (sorry grandma).
But the concept of computing isn't going away for a long, long time. What's changing is the machinery being managed by code. First tablets and cell phones, and soon the Internet of Things (stupidest name evar). But even if we run out of wires and electricity, I suspect some day we'll be creating living things (tissues at first) that can perform a computation. So yeah, computerize ALL THE THINGS!
This was literally the reason for the design change in 2007 - they abandoned the toolbar system for the ribbon system exactly because 95% of all feature requests were for features that were already included in the product. People just couldn't find these features.
And so they reworked the UI, now users can find the features they want to use. And it generated so much bad publicity that they re-added the toolbar system to Office 2010.
The lede: "The ability to delete yourself from the Web doesn't really matter. What really matters in the age of advanced surveillance is the right to not be correlated. Technology is always watching and capturing you, but the correlation is where the danger lies. Laws can change that, but only if enacted soon."
That's an interesting perspective from the "chief information security officer for In-Q-Tel". [0] But yes, it does seem inevitable: pervasive surveillance of everyone by everyone. Like a global village aka small town.
However, I'm not convinced that laws would be enough. Criminals (large and small) don't care so much about laws. So arguably we're each responsible for our own privacy.
The ubiquity of requisite knowledge and technology, facilitated by leaks, may allow the sufficiently motivated to claw back some privacy. But sadly enough, perhaps the most highly motivated are the criminals. Not good.
[0] http://www.csmonitor.com/World/Passcode/Passcode-Voices/2015...
Now his position makes sense. He wants governments (presumably those who fund his firm) to have a monopoly on surveillance and correlation. While I agree on dangers from other criminals, restricting private institutions and individuals is authoritarian, unless there's an exception for whistle-blowing.
[0] https://cpunks.org/pipermail/cypherpunks/2015-June/008069.ht...
> I, for one, would gladly paraphrase John Perry Barlow's declaration of independence of cyberspace and say that the "weary giants of flesh and steel" should leave me alone but only if the "technology [that] is being imposed on a global scale without restraint" will do likewise. A pox on both; may they fight to a standstill somewhere other than my front room or my backyard.
[0] https://cpunks.org/pipermail/cypherpunks/2015-June/008081.ht...
