You Are Not Late

>You could pick almost any category X and add some AI to it, put it on the cloud. Few devices had more than one or two sensors in them, unlike the hundreds now. Expectations and barriers were low. It was easy to be the first. And then they would sigh, “Oh, if only we realized how possible everything was back then!”

It's a bit easy to imagine once something's already survived it. A device with a hundred sensors probably survived hundreds or thousands of failed released or unreleased iterations. Every other mature field has probably gone through this same sentiment (and maybe still going through it, and perhaps will always go through it). If only astronomers today could imagine what new, novel techniques are just lying right in front of them with all of this computing power and automated manufacturing ability. Imagine how much further ahead astronomers would be in 2044 if they had discovered thingamajig back in 2014 instead of 2040! I think the greybeards of 2044 would be a bit biased.

But this feels like we'd be backpedaling on choosing the right tool for the right job. Slapping AI onto something and watching it fail is great for our knowledge, but not our wallets. This is when business leaders create friction with development. This is where you get "Just make it work" commands which gets the snowball of technical debt rolling. The value gained is "This solution sucked in reality! Or maybe we just sucked delivering it! Don't do what we did unless you have a good reason!", but that's not what we're really after, is it? We could sit around and brute force a bunch of combinations of ideas and markets, I guess.

And finally, if it was true back in 1985 with domain names, and it's true today with AI and the cloud, why wouldn't it be true for 2044 with VR contact lenses and holodecks? It seems like the future will be like then and now in this regard. So.. there will always be a wide open frontier somewhere? Do we really say anything substantial by saying "Today's an open frontier for internet ideas!"

I think there's a lot less low-hanging fruit than this article might have us believe.

Couldn't disagree with you more. To name just a dozen new internet things from the last 10 years that I did not have in 1990s.

  o Google
  o Wikipedia
  o Netflix
  o iTunes
  o Spotify/Pandora
  o Stack Exchange/Overflow
  o RSS (Google Reader/Feedly/etc..)
  o GitHub
  o Facebook
  o Instagram
  o Whatsapp
  o Gmail
  o Google Earth/Maps.

I can only imagine what the next dozen new categories will be invented in the next 15 years.

Not sure what newsgroups you use these days, but every one I ever made use of has turned into a cesspool of spam and low value content. Usenet has been dead for at least 5+ years for me - there was a 4 year period where my fingers had memorized every shortcut available to nn...

Correct. Note: US centric dates: My point being that starting up a railroad in 1850 was a great idea, not so good by 1900.

Or there were about 150 startup automobile mfgrs around 1900, not quite so many by 1925.

Starting up a telephone company in 1900, good idea, in 1950, not so good idea.

We still have railroads and car mfgrs and telcos in 2014, they're just not realistic startup fields.

Every other field of recent human activity has gone thru the same stages of startup success, startup failures/consolidations, oligopoly oscillating with semi-regulated public utility, sometimes a long term decline in the industry after that.

Someday its economically inevitable that "Internet and HTTP" will not be the place to do startups. I don't think its today, but my grand kids are highly unlikely to be doing internet startups.

The barriers were still high back then, but they were different barriers. Bandwidth, CPU and storage all cost a lot more than they do now, and the tools available were less sophisticated. Building an online store or CMS back then was a much bigger piece of work.

Today, you don't need to focus on those kinds of problems, generally. The problems are the next level up.