Warning: this is a rant.

Important: the following reflects my opinion and my opinion only. I report the facts at the best of my knowledge/understanding but they are not necessarily accurate.

Of course I’m going to talk about /dev/tty keycaps set and the MT3 profile. The two are actually very different projects.

MT3 is the new profile we designed. It was a long process but it went relatively smooth. We had many iterations of the keys, I got plenty of prototypes both 3D printed and injected. We were able to fine tune the shape up to the very last minute. Indeed the profile came out beautifully, better than my expectations actually. The result is a wonderful new and original profile.

I was expecting the same process to happen for /dev/tty which is the first color scheme released on MT3.

The color selection has been painful, with color samples coming on and off without logic. Sometimes completely wrong, sometimes missing some pieces. I asked to see the colors on an actual injected keycap because the result changes from the color chip (which is like a small tag of plastic) and the final product. That did not happen for all colors. I also asked to receive at least three options per color (one slightly lighter one darker). That also did not happen for all colors. I also discovered that color shades are not absolute but very much cultural. That added a layer of complexity dealing with China.

Anywho. I did the best I could with what I had. At the end they turned out pretty well, except for maybe orange. I believe the problem is with the pigment that is not dense enough and the keycaps turned out not intense as I thought. That has been very difficult to explain because apparently nobody I interfaced with knew what I was talking about or what color pigment is.

Dye sublimation is another story. I only got one partial sample in a very early stage. I was able to outline some problems and I was expecting to receive a final prototype before going full ballistic. That –again– didn’t happen. We were able to tune in with the manufacturer with some photos, but that’s it.

Unfortunately in such complex projects organized between US, Italy and China some hiccups are bound to happen, but I can’t help but think that we should have done more. I’ve been told that Quality Control was set up but they need to be instructed first otherwise they don’t know if a rotated or a bolder legend are actually an issue. I’ve never been asked to train QC.

Perfection is not a state of things, it is a process and it has costs. At one point you have to release and move the improvements to the next round. They reached the point where they couldn’t procrastinate anymore or the whole thing would have collapsed.

What I am upset about is that I never received a final sample to review before the actual production started. I got a set –that I thought was still a prototype– the same time everyone else did. I was left with the assumption that the errors could have been still fixed but that wasn’t the case.

Why didn’t I get a pre-production? Because making one sample and holding the manufacturing lane for a week is expensive and you may risk to lose your place in line and the Chinese new year was dangerously close. Still, the lack of communication from the first dye-sub sample up to last week is upsetting.

I can honestly say that MT3 is heading where I wanted it and I can call it my creature. I can’t say the same for /dev/tty (that is spelled “device terminal”, btw) that is something I hinted and MassDrop and the manufacturer interpreted.

Good news? The set is still pretty darn good and if you got a flawed kit just ask for a replacement, I know they are already working on it. So at the end it’s not that bad, still when you work on a project for years and you reach 99%, I can’t help but feel that bittersweet aftertaste for that missed 1%.

Thank you for sticking with me up to this point. I can tell you that we are already working on a second release and this time we are making sure that everything is in place… At least so I am told.