published on : Wednesday, August 28, 2024
Something I have started thinking about a lot these days is the stagnation and the maturing of the software industry. These days, I find myself wondering:
“What more can they build? seriously.”
I have not come across a software application that’s new/exciting to me in the last five years. The last time I was excited about something new in tech was when Apple came out with the M series chips.
I have expressed this sentiment with a few other software developers & friends and a lot of them agree that it’s become harder to find new things that are interesting.
Before, when I expressed this, I would get back all sorts of answers - mixed reality is coming, self driving cars will make owning a car obsolete, futuristic visions of us interacting with computers by talking to them, AI will put everyone out of work, but it’ll also be so amazing that new kinds of work will emerge.
These days though - there is a growing recognition that a lot of the ideas that were put forth as ‘the next big thing’ are pretty dumb ideas. Those that aren’t, are prohibitively expensive and lack a clear use-case that justifies that investment.
Take for example - TikTok and Discord. These are two new software companies that have become dominant in social media and gaming circles, and are touted as the next big platforms. In a lot of ways, they are. TikTok is valued at $50 billion, and discord is valued at $15 billion. Objectively speaking - they are huge.
But also?.. neither of them is something truly revolutionary that we didn’t have before? TikTok is some combination of Facebook and Youtube with a more addictive algorithm, and discord - well, that’ just IRC from back in the day, or Slack if you’d prefer a more modern comparison.
The more you start looking at what counts as a ‘tech company’ these days, the more cracks you start to see in the story - I once read/heard someone describe ‘We-work’ as ‘Aws like cloud computing for the real world for office real estate’ a comparison that made no sense.
For one thing, Amazon doesn’t lease compute capacity from others and give short term compute leases. They own their infrastructure. Similarly, when Peloton went public, my Linkedin feed was filled with people (mostly product managers) who were gushing about how Peloton had developed a ‘revolutionary way’ to make exercising ‘engaging’ and was building the next ‘social media platform’ - focused on working out of course.
Zoom was the darling company that got the most attention during the pandemic. Again, I remember using a software product called ‘gotomeeting’ which did everything zoom does, and I used it back in 2013 or so. I never understood the breathless excitement around these software products which are just newer versions of tech that’s existed for ages (in tech company years).
Finally, let’s talk about AI. A subject I have been avoiding talking about till this point. ChatGPT came out on November 30, 2022. It’s been ~ 2 years now. The capability is around the same as what existed roughly 2 years ago. Sure, it’s gotten better over time, and I use it once in a while.
When it came out, I became hooked, and was a paying customer at one point. It has certainly boosted my productivity at certain tasks - writing shell scrips and one off things like that. Writing test cases, etc. I used it quite a bit, got a good sense of what things it’s capable of doing, what it sucks at doing, etc and now my usage of ChatGPT is probably on the order of 2-3 queries a day on a good day. Mostly, it’s replaced my typical workflow that went something like this:
“search on bing, and then go to stackoverflow and read answers, see the comments below answers, and figure out which is the right solution to the jest test case error you’re seeing.”
However, the image generation, etc doesn’t really work. The hallucinations are hard to catch in text output when I ask about subjects I’m not an expert at, and it’s mostly serves me in niche, sporadic set of activities. In everything it generates, there is a feeling of ‘AI generated’-ness to it. You can tell.
This lengthy preamble now bring me to the title of this post - As a software developer, I feel that all the low hanging fruit in software application development has been picked. I feel that the middle age of technology companies has started. This is the time when growth has stalled and the avenues for growth are all gone. There will be fewer $10 billion dollar+ software companies in this decade than at any other decade prior is my prediction.
Just look at the actions taken by Microsoft and Google - one is putting ads for Candy Crush in the start menu, and the other has made the default experience of YouTube so shitty that it’s unwatchable without YouTube Premium. Apple is now demanding a cut of donations given via Patreon. Amazon search has ruined product serp page, so I only have prime for the fast delivery from trusted brands, and for Amazon prime video.
If these companies were growing, they wouldn’t be doing things like the above to increase their revenues. They have run out of ideas. New ideas are hard to come by, and the stock market demands continuous revenue and profit growth - so, they do what they are doing.
Don’t get me wrong - I still think there will be a lot of really new and exciting software products that will be developed. They will be more niche ones. That’s a good thing! but ‘Big Tech’ ? well, it’ll still be big, and it’ll still be ‘tech’, but the giant will have finally stopped growing.
What next after ‘big tech’ ? I don’t know. But my feeling is that the time has come for another sector of the economy to replace tech as the next big thing.