3 ms·
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 ind
by dasil003 17y ago
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.
- roundsquare 17y agoI 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.