The content/meta-content concept is pretty forced. The reason we technies know that something like Spotlight search must be creating "meta-content" (ie. an index) is because we know through experience that grepping 100GB HD is not going to be performant, so we know something must be going on under the hood.
I think the divide is simply that techies automatically think algorithmically about how everything works. Everyone expects technology to work the way the way they need for their use case, it's just that techies are quicker to think about what is logically feasible or why it might not be working. We are natural debuggers as well, so we rarely get stumped with technology for long.
There is also one other class of people... the technophobes, like my mother, who is scared to do anything with technology lest she accidentally sever her email connection and erase all her data by double-clicking something she was supposed to click or vice-versa.
I agree that thinking algorithmically about things is a part of what divides techies from non-techies, but I think a part of that is what information an algorithm has access too.
This is what I think the author is saying. With the TiVo example, techies know that the only information available is 1) when the show was recorded and 2) if there is some flag about saving on the show. Take that information, and we quickly come up with the most natural algorithm (FIFO + check "save till" date).
I don't think the TiVo example is a good example though. Here, the problem is the data to use exists, but isn't being populated. The example with the picture is better. Think about explaining how google images comes up with pictures. Lots of people think its searching the contents of each image so when you search for "Saddam Hussein" it comes back with pictures of him. In reality, its searching what the author calls "meta data" and just doing it extremely well.
This quote is a great example of how to help friends and family with tech issues w/o pissing them off in the process: I've banished from my vocabulary in dealing with users the question: "what did you expect to happen?" (intoned inquisitively with stress on the word happen, not sarcastically with stress on the last syllable of expect). Because what the user hears when you ask that question is, "you disagree with the computer. The computer's right, you're wrong. So 'fess up on your idiocy, and I'll have a nice laugh with my techie friends later at your expense." Instead, I said to my friend, "oh, no! Still... I bet there are some people out there who really love their Project Runway reruns, and don't care for The Shield that much. How do you wish the TiVo had worked differently?"
Honestly, I think that is the best part of the article, much more important than the divide he discusses. When trying to explain why something works the way it does, say three things:
1) Some people might want it the way it works now.
2) How should the technology know you want it this way.
3) Show how the user can tell the technology what it wants.
I'm not convinced of the premise either. A major tenet of computer theory is that programs are data. The distinction between "data" and "metadata" is merely for human convenience, as is the distinction between "code" and "resource".
If techies are the ones who accept this false dichotomy and non-techies are the ones who don't, I'd say it's us who need to adjust our position, not the other way around.
I found that paper really interesting, but didn't see multi-layered abstractions mentioned at all. It just said people must be able to deal with the "meaninglessness" of the syntax of computer programs.
Anyway, I'd love to read more about this if you have any more links to share.
The TiVo story feels to me like a particularly egregious case of non-techie misunderstanding. I'd be very surprised if any of my friends or family, even the least technical among them, expected their TiVo to choose what to delete according to their valuation of its content.
I can imagine someone not realizing TiVo will delete shows at all, and perhaps expecting a 'Your TiVo is full!' warning, but it's hard to believe someone complaining 'It deleted x when I would have preferred for it to delete y.'
I think the divide is simply that techies automatically think algorithmically about how everything works. Everyone expects technology to work the way the way they need for their use case, it's just that techies are quicker to think about what is logically feasible or why it might not be working. We are natural debuggers as well, so we rarely get stumped with technology for long.
There is also one other class of people... the technophobes, like my mother, who is scared to do anything with technology lest she accidentally sever her email connection and erase all her data by double-clicking something she was supposed to click or vice-versa.