Sun Microsystems, NVIDIA and products that found their fit later

When I used to teach basic Java, in the first class I liked to spend a bit of time explaining why Java exists. I would tell the whole story of James Gosling and the group that left Sun on a retreat. They wanted to come back from there with a big idea. I romanticized it a bit and certainly told a different story from what it really was.

But it’s interesting to see that, when they come back with the idea of this language that was going to be called Oak and ended up being Java, they were watching the growing number of electronic devices and the need for those devices to have software. The VCR, the television, the blender. Before, it was all machine language, and that machine language depended on the processor and the electronics. Maybe it made sense to have a programming language for different devices. In 1992, 93, that was quite audacious. And even more: to have a language that would be understood by all of them, they were thinking of really creating a piece of hardware, a board, a circuit, who knows, that all these devices would simply need to have, including video games, and they would pay royalties to Sun Microsystems. Anyway, it didn’t work out, it wasn’t the right timing. And this matter of timing, of the idea, is something very relevant in the world of entrepreneurship, not only in tech.

It’s interesting to see that this idea of Java was shelved for a little while and soon came back with the idea of being applied to the web. Because the web had this problem at the very beginning, and always has: there are thousands of different browsers. At the time it was Spry Mosaic, I even installed that thing, running on Linux, on Sun’s Solaris, on Windows, on Unix. It would be complicated to give the power to run code inside the browser, when what was underneath, to understand the code, varied a lot, both in hardware and in operating system. So they looked at that idea of the programming language for devices and said: hey, I think it fits here.

That’s where applets were born, which a lot of people don’t even know about, didn’t live through. The browser was very weak at executing code. It really was just HTML, there was no dynamism at all. Nowadays there’s a whole deep JavaScript machinery, not to mention CSS itself, which evolved along with HTML, plus thousands of other possibilities. The applet gave Java this chance to emerge.

But look: nobody knows Java because of applets anymore. Java ended up being used a lot on the server side, so much so that there was TheServerSide, a site that talked a lot about Java, with that nice name. Why? Because there was also this big problem: big companies had mainframes, giant computers, some had Windows, others Linux, simpler computers, commodity. It was a problem to write in C, for that machine, a service that would be consumed at the other end through the web. Java was applied in a completely different place. Look how interesting: it was conceived to run on all sorts of electrical devices, it launches because of the web, of applets, but it really takes off because of the server side.

This is a phenomenon of several companies that are born with a product and it ends up being a success somewhere else, sometimes after a long time. There are several cases. NVIDIA, which I remember from the days of 3dfx, of the Voodoo cards, was totally focused on video games. The GPU had this explosion first for data science and then for artificial intelligence. NVIDIA, which was already a big company because of video games and graphics, rode the wave and was applied to something they didn’t even imagine had this market fit.