In this entry, we explore "The Hundred-Year Language" by Paul Graham. At the start, Graham makes quite a compelling question, What language will control the world in one hundred years? Beats me, I thought, I'll be dead (now that I think about it, I could also very well be that code, living an afterlife in an Amazon server).
In a very strange way, I went into this article expecting one thing, but the more and more I read, I thought, "Graham and I are thinking very different things". The start, however, was quite good. If we think of the evolution of programming languages as many different branches, and in the future only certain branches will prevail, it's smart to stay on the branch that will still be around in the future. COBOL is on it's way out, it's branch didn't prevail, good. Java will probably perish as well, Brilliant! So far, a good assessment.
The following got me pretty confused. Graham begins to speculate on how processing power in the future will probably be over a million times more powerful than it is today, sure, that makes sense. However, making programming languages more efficient by flattening data structures, and the suggestion of getting rid of Strings, makes less sense to me. I feel like a language in the future might not be like that, but highly inefficient to make it easier to understand.
When Graham dives into parallelism, I'm on the fence on that one. I don't know if such a great opportunity for faster processing can keep getting dismissed that far into the future. With the possibilty of quantum computing, parellelism should be quick to rise.
The number of programming languages, oh boy. I personally think rhat there will be too many. Maybe even CompSci as a whole might begin to split, with emerging technologies, IoT, App development (which I feel like it changes their languages and platforms every 6 months, I tried Xamarin, then Flutter, Java Android then Kotlin Android, Objective C then Swift, what a mess) and upcoming technologies that we might not see coming, there will be an obscene amount of languages.
So in the end, Graham states that the 100-year language might be able to exist today, and it might be a good idea to code in it. I'm sure that no matter how much of an expert you are in a certain field, the amount of foresight necessary to make predictions like that is simply non-existent in the mind of a mere mortal. As the author stated, though,it's not like our ghosts will stick around just to say "I told you so".
In a very strange way, I went into this article expecting one thing, but the more and more I read, I thought, "Graham and I are thinking very different things". The start, however, was quite good. If we think of the evolution of programming languages as many different branches, and in the future only certain branches will prevail, it's smart to stay on the branch that will still be around in the future. COBOL is on it's way out, it's branch didn't prevail, good. Java will probably perish as well, Brilliant! So far, a good assessment.
The following got me pretty confused. Graham begins to speculate on how processing power in the future will probably be over a million times more powerful than it is today, sure, that makes sense. However, making programming languages more efficient by flattening data structures, and the suggestion of getting rid of Strings, makes less sense to me. I feel like a language in the future might not be like that, but highly inefficient to make it easier to understand.
When Graham dives into parallelism, I'm on the fence on that one. I don't know if such a great opportunity for faster processing can keep getting dismissed that far into the future. With the possibilty of quantum computing, parellelism should be quick to rise.
The number of programming languages, oh boy. I personally think rhat there will be too many. Maybe even CompSci as a whole might begin to split, with emerging technologies, IoT, App development (which I feel like it changes their languages and platforms every 6 months, I tried Xamarin, then Flutter, Java Android then Kotlin Android, Objective C then Swift, what a mess) and upcoming technologies that we might not see coming, there will be an obscene amount of languages.
So in the end, Graham states that the 100-year language might be able to exist today, and it might be a good idea to code in it. I'm sure that no matter how much of an expert you are in a certain field, the amount of foresight necessary to make predictions like that is simply non-existent in the mind of a mere mortal. As the author stated, though,it's not like our ghosts will stick around just to say "I told you so".
Comments
Post a Comment