Reading 05: Watch your Language





Language is an incredibly important tool in all facets of life. Different languages have different ways of expressing concepts such as numbers or the future, all of which affect how we think about these concepts. In some languages, the future is described more in vague terms rather than exact terms, which can lead to differences in how cultures that speak that language view the future. In written language, we can see that the Arabic script is far more efficient than the Roman script in terms of numbers, which is the reason we now use Arabic numbers instead of Roman numerals.


The same applies to programming languages. Choosing the programming language that you are going to use when starting a project is one of the most important design choices you could make. You must consider the efficiency of the programmer in addition to the efficiency of the program. For a small script that is unlikely to see use past a very few people, it is perfectly fine to choose a language like Python that is quick and easy to write a script in. However, for projects that will be used by many people, or perhaps a project that requires an incredible amount of computational power / data, it becomes more important to decide which language to use in development. A language like Go offers powerful tools for concurrency, which could be useful depending on how much your program uses that. A language like C or Zig, while a bit more difficult to program in offers incredibly fast performance. There are other languages that are harder to initially learn, but are incredibly safe, such as HAL/S which is used by NASA for the Space Shuttle program.


Thus, the biggest thing is not whether a language is more powerful than another (they are all mostly the same in power; people have even written operating systems in Scratch), but the different characteristics and strengths each language has to offer. One strength could be seen as accessibility - sure, a language may be less efficient at doing things than other languages, but if it is easily accessible it can carve out a niche in certain communities. Even if it is inferior in other aspects such as safety, efficiency, if a language has more documentation and is easier to learn, it will be used.


I think of Lisp, which the author of these essays loves. It is a bit trickier to pick up though, than high level languages like C++ or Python. However, it is very powerful and allows you to do some pretty cool stuff in an easier manner than other languages (once you manage to think in a Lisp way). When I was browsing /g/ on 4chan after setting up Linux on my computer, I saw an argument between two anons (which before degenerating into insults that I can’t repeat here), touched on some of the same topics that Paul Graham writes about regarding Lisp. One anon argued that learning Lisp was the only language that you needed to know (speaking from a purely hobbyist standpoint) because it was more powerful than other languages, and once you learned to think in Lisp, you would be able to do much greater things than in other languages. Overall, reading Graham’s articles and remembering that argument has led me to think more about learning Lisp this semester (especially considering Emacs uses a variation of Lisp as a scripting language).

Comments

Post a Comment

Popular posts from this blog

Reading 04: Programming is an art (Donald Knuth approved!)

Paul Graham sees hacking (and by extension hackers) as closer to an artform than anything else. Paul describes hackers as makers, people who are creating something beautiful, not necessarily something new and original, but something beautiful whether in output or in design. I believe his definition is compatible and in fact fits in very well with Levy’s description of hackers. The hacker ethic even mentions that computers can be used for art and beauty! Additionally, from the stories we have read of early hackers trying to impress other hackers through code shaving to tales of hackers choosing to program an astronomically accurate star chart for the background of a game, we can see that beauty has been part of hacking since it’s very beginning. There are certainly diminishing returns to code shaving - once you’ve reached the point where you are spending hours upon hours to try and shave off a line or two of code, you’re wasting your time if you purely think of your efforts as being to...
Read more

Reading 02: ABC Always Be Selling out

  In this week's reading and discussion, we learned about the second wave of hackers, who were more focused on hardware than just software. These hackers were of a very different type than the hackers who were shaped by the culture of the RLE lab. While the first hackers were systems hackers, who were obsessed far more with the software they could make on the hardware they had, the second wave hardware hackers were very interested in building their own computers, especially with the rise of microprocessors. While they both emphasized the importance of the Hacker Ethic and the hands-on imperative, the main difference was that software was not the end all be all for these hackers. They did write and give away programs for free (at least, in the beginning they did), but they also focused on building their own computers and increasing their performance. The fun for these hackers was often in the actual construction of the machine and knowing why it worked, which was not as important fo...
Read more

Reading 09: Open source? Sounds communist to me

Open source can still make business sense today, even if it isn’t necessarily the most profitable choice (by %). As ESR has mentioned, by providing software for free, you can increase your market share more than proprietary software, and then sell add-ons / support / premium versions of the software in order to monetize the open source software. I think another element in favor of releasing open source software is the potential for it to be widely used by developers, who then encourage their companies to use it (and their companies will pay for an enterprise license / support). I believe winRAR used a similar model of selling, since it allowed users to use the free version of it indefinitely even though it warned them that they would have to buy a copy of it eventually (albeit not being an open source piece of software). I think that services instead of software as a business model is definitely a viable model, potentially even more viable than the traditional model of simply ...
Read more