The story I've heard is that the original lightbulbs filled with soot and darkened well before they burnt out. Edison guessed the soot was coming from the filament and was charged, so he placed a charged plate across from the filament to attract the soot before it could stick to the glass. It worked, but the bulb was much darker than before as the charged plate siphoned a lot of current off the filament.
That device would have three conductors, which you can see in the photos. Two for the filament, one for the "plate." Later, a fourth conductor called the "cathode" was added to be heated by the filament (rather than pulling charge directly from the filament) to make the first diodes.
There's something funny about how modern active components are designed to reduce how much heat they consume, but the first active devices required an active heating element to function at all.
To call De Forest's audion (now known as a triode) an "improvement" of the diode might be technically true, but misleading: being able to amplify signals was a quantum leap in technological capability, and a necessary precursor to there being an electronics industry.
De Forest himself, however, apparently did not understand how his audion worked, at least at first, and felt it necessary for it to contain some gas. There's an excellent short biography that I will post a link to if I can find it...
Edit: This is not the article I had in mind, but it shows that initially, De Forest's audion did not have a connection to the grid, and also that De Forest was unaware that the triode arrangement could be used to amplify, and considered it only an improved diode; it was Amstrong who discovered its amplification. IIRC (from the other article I have in mind) De Forest was able to appropriate the fruits of this seminal discovery by use of his audion patents (Armstrong was later screwed over by Sarnoff and committed suicide.)
I don't think it's even technically true, they're fundamentally different circuit elements that just share a principle of operation. Diodes allow current to flow in one direction (rectification), triodes provide a proportional increase in the energy of a signal (amplification).
Their shared implementation details mean you can wire a triode as a diode but I don't think it's even a better diode.
Cool. I'm old enough that I've actually worked with tubes (final amps for 1970s Ham Radios needed too much power to sensibly use the solid-state transistors of the day. Also Fender guitar amps were just turning solid-state in the late 70s/early 80s, so used tube amps were cheaper.)
I would not have guessed that the first ones happened in 1889; Anyone have a favorite link for tube history?
That device would have three conductors, which you can see in the photos. Two for the filament, one for the "plate." Later, a fourth conductor called the "cathode" was added to be heated by the filament (rather than pulling charge directly from the filament) to make the first diodes.
There's something funny about how modern active components are designed to reduce how much heat they consume, but the first active devices required an active heating element to function at all.