3 ms·
I think there is an interesting question here but I'm not sure I follow the train of thought. First of all, I doubt that whatever happened relatively recently
by 4cao 6y ago
I think there is an interesting question here but I'm not sure I follow the train of thought.
First of all, I doubt that whatever happened relatively recently with regard to certain design choices in hardware and software is related in any way to how different implementations of BASIC fared against one another in the 1980s. Regardless of the conclusions, there are simply too few people involved in making these choices now that also know what the situation was then.
As a matter of fact, one of the perpetual problems in the field seems to be that so many already identified problems with certain design choices and solutions to them are being forgotten, and have to be rediscovered again by every new generation of programmers.
And I'm not sure how all this can be framed into the Europe vs America debate: most of the big things happen in the US. Arguably, having a huge single market and less regulatory overhead helps but it's also because this is where the financing is. However, there are people from all around the world working on these things, so to what extent the final outcome can be called "American" is debatable. The founder of Commodore himself was a European Holocaust survivor. And we really have to be careful lavishing praise on "elegant" and "efficient" European products lest we forget about whom to hold responsible for monstrosities like Systemd. Could it be just that we tend to remember the most succesful products from the past, and those originating outside the US had to meet a higher bar to become so?
Perhaps there is an argument to be made that at some point while stuff in Europe was still being "built to last," American businesses have already been discovering the allure of planned obsolescence. This would pertain to the hardware side of course. As far as the software is concerned, I think everybody just got lazy the moment they could get away with it due to the rapid advancement in hardware performance. So from the era where you could have Jet Set Willy or Dynamite Dan (platform games with dozens of levels) fit in 48 kB, we've now "developed" ourselves into a corner where even winver.exe (a program whose sole purpose is to display the Windows version and an "OK" button) takes more than this (56 kB on my system).
While I concur this outcome is not desirable, I can't really agree it happened because of "moving close to the metal." On the contrary, I think this was only possible because of moving away from it, one layer of abstraction away at a time. To paraphrase the quote attributed to Churchill, BASICs of the future will no longer call themselves BASICs but I think if a BASIC was being invented today, it would look like another JS and it would be part of the problem, not the solution.
- lproven 6y ago(Blogpost author/submitter here.) It's a fair cop. It was only a blog post, dashed off on the basis of a FB comment that grew overlong. :-) I do not have a coherent thesis that I was expounding; it was merely a passing thought. I find it interesting, though, how the computer market has developed in my lifetime -- I first touched a computer keyboard in about 1981 and owned my own a year later. In the early 1980s there were dozens of different manufacturers of almost-totally different and incompatible systems, and it was quite normal for the larger developed countries to have one or more indigenous computer makers, often with their own architectures. Most used one of a small number of CPUs, yes, and those CPUs were mostly American, yes, but I knew of English, Welsh, French, Spanish, German, Norwegian, and Swedish makes of computer, to pick the first few to come to mind. I'm sure there were many more. Now I live in the former Communist Bloc, I've also learned of hordes of Sinclair clones and other copies of Western home computers that thrived over here, often after they were dead and gone from their homelands. I covet a Didaktik Kompakt. :-) By the 1990s things were starting to consolidate but there was still a lot of variety, and major innovations came from other countries -- as other commenters note, from Denmark, from a Brit at CERN, from a Swedish-speaking Finn, etc. As I said in the piece, I used a Finnish cellphone with a British OS, and so on. But when the world market consolidated down to a handful of companies and technologies, the winners were pretty much all American. The European software luminaries relocated their and often became American. And yes, I think the overall design ethos has changed as a result, and I find that interesting. I have rarely seen it discussed. I often see people bemoaning modern bloatware, but I rarely see analysis of why and how it happened. This was my 2¢ worth on the subject.