From my journal, about trying Claude yesterday:

I went back and asked it a couple of questions for the options it chose and it explained those to me too in a way that in the past I’d spend hours on IRC (or Reddit or whatever) to try and get this info from some Linux asshat who would only know half of it, if at all. It’s the learning part I like the most. The part that it gives me an answer, and I can ask why, and then why again, and again. I understand why people freak out AI will teach their kids one day, but dude, if this was my programming professor at college… I’d be in a different place today.

It’s yet another example of what a great teacher AI can be, but this time, it really landed home:

There’s a lot going on at school (and in college) in terms of human to human relationship. But those afternoons I fell asleep in the library trying to cram information into my head that I just couldn’t understand for hours, no matter how hard I tried. Those hours of frustration trying to find help on various Linux forums, to get someone to help me boot up my computer, and being told again and again to RTFM (and that’s when I got a response) were a punishment I earned for my curiosity.

It took me years to gain the confidence I have today to justify my unorthodox learning style, that of going from the end, not from the beginning, with tomes of text.

I was discouraged by my teachers and at some point by my parents. I don’t think they had bad intentions: it came from a good place. They saw me struggle and fail again and again and again and suggested (some less gently than others) that I should try something else.

Yesterday I tried Claude with something rather elementary for Linux users: mounting an SMB share from my Synology to my Linux computer.

The traditional approach (which I believe is still best) is to work through the fstab. I didn’t want to “mess with it” because I knew that it’s one of those sensitive system files that might mess up the computer boot if I’m not careful, and I used that fear to have me just use the GUI file explorer.

Claude gave me just the right amount of encouragement from the start:

Your concern about fstab is valid but manageable — the risk is real but small if you’re only adding a new remote mount entry at the bottom. You’re not touching the existing lines for your OS drives, so even if the SMB entry is wrong, the worst that realistically happens is the mount fails on boot and you get dropped into a recovery prompt. You can add nofail to the options to completely eliminate even that risk:

That was right. I was just adding to the file. It wouldn’t damage things this way. What’s more, it suggested the nofail option, which I had in my fstab before, but I didn’t remember why and where I got it from; it was one of those things I copied and pasted because it worked. Claude, with this simple sentence, explained what it was and how it would help.

The rest of the conversation with it felt like a good patient teacher, because that’s how I treated it. Claude could have given me the whole answer, and it did suggest it. I could have just copied and pasted my fstab as it is, let it work its magic, paste it back, and enjoy the magic without understanding it. But that’s not how I work.

I chose to go step by step. I told it I’m going to switch from my Mac to my Linux desktop, which wasn’t needed, but it helped me by roleplaying with it a bit, making it my teacher. And it got better from there, as it assumed that role and waited for me patiently. Of course it would, it’s code on a computer. But it acted out the role nevertheless. It polite and patient, and that’s one of those small things that clicked and made it work.

When we got to a point of testing the fstab and it didn’t work, it asked me to paste the errors I got and worked with me through them patiently. It told me I forgot a sudo one time (duh), and at another time, it asked me if I’m sure I downloaded and installed cifs-utils on my Linux desktop as it suggested, and I forgot. “Oops,” I typed, playing my role; “no problem, you need some coffee,” it replied to me with a wink emoji. Once again: I know it’s not human. I understand it doesn’t have real patience. But how much I needed that patience - even an artificial one - before! How often was it missing!

Suddenly, I got goosebumps. I remembered something else:

The Terminator would never stop. It would never leave him. It would never hurt him, never shout at him, or get drunk and hit him, or say it was too busy to spend time with him. It would always be there. And it would die to protect him. Of all the would-be fathers who came and went over the years, this thing, this machine was the only one that measured up.

In an insane world, it was the sanest choice.