4 ms·
I worked on a Logo computer in 1983 but never launched it. Even so I tried to keep up with the Logo community over the years. I was very disappointed to see wha
by jecel 5y ago
I worked on a Logo computer in 1983 but never launched it. Even so I tried to keep up with the Logo community over the years. I was very disappointed to see what schools did with Logo through the 1980s. To me it seemed that the worst problem was that the teachers were gatekeepers - if there was a book they were the ones who had it and they only passed a small fraction of that to the students. Every time I would see a school showing of their student's Logo projects I would see how many pages of the "Apple II Logo" book you would have to read to do that. It normally would be around the first 8 to 12 pages of the 280 page book.
The solution would be to put the book in the system. That is what Smalltalk-80 came close to offering. But 8 bit floppy-based computers were not up to that. Eventually the Internet would become a way around the gatekeepers, but by then the idea that Logo had failed had become widespread and we had to wait for Scratch to occupy (not very well) that space.
- andai 5y agoCould you elaborate on the current situation? Did Logo do some things better than Scratch?
- jecel 5y agoBy "current situation" I mean that with larger disks and the Internet in the mid 1990s it was possible to package a system like Logo with a whole book and a nice library of examples. In the case of Scratch there was a shared website where students could post their projects and look at projects from all over the world. As Alan Kay would put it, being literate is not only being able to read and write but having access to literature. For some reason many people don't take Scratch seriously because of its visual nature. So they do a similar system where you can press a button and see your block as JavaScript text so you can know that you are doing "real programming". The real reason Scratch is limited is because it was created that way on purpose to make it easy to make simple games and animations. It could easily be expanded into a full programming language and the Snap! project at Berkeley did exactly that. https://snap.berkeley.edu/ https://snap.berkeley.edu/ Logo was a full programming language from the start. Early versions didn't even have the famous turtle and focused on text applications. By the time of the 8 bit microcomputers it had become associated with the graphics (and so was used as the world's most awkward drawing program) and limited implementations. 64KB was a huge amount of memory for Basic or Forth, but only allowed the simplest Logo programs to fit and ran them very slowly. This made people think that Logo was just a toy (which was not true) like today they think Scratch is a toy (which is true, though not for the reasons they think).