5 ms·
Generally no, because they do not have large populations of self-taught professional practitioners who never learned theory. Least common denominator writing i
by Kalium 6y ago
Generally no, because they do not have large populations of self-taught professional practitioners who never learned theory.
Least common denominator writing is very useful for a great many things. It is wonderful and ideally suited for items aimed at a popular audience. It just may not always be ideal for efficient and precise technical communication.
- swagmoney69 6y agoHow can a self-taught individual such as myself begin to learn the theory presented in the academic computer science world without enrolling in a college or university?
- rovr138 6y agoIf talking about physically going to university or college, take a class online. If you don’t want classes in general, go to course sites for universities and check the textbooks they use/reference.
- avs733 6y agoto add a term... they have professionalized. With licensure, required training, standards, codes of ethics, legal recognition by government actors, etc.
- bcrosby95 6y agoI don't really buy this. You're framing this as if people don't use this notation because they are self taught. But I've worked in companies full of fresh out of college people and we didn't use this notation when working on projects. And when working on group projects in college, we didn't use it either when discussing possible solutions. We used pseudocode. Beyond that, I think you're selling an ideal here in these other professions that doesn't necessarily exist. There certainly is some vocabulary overlap but a huge problem in e.g. medicine is practicing doctors not being able to understand the science behind medicines and procedures that biased 3rd parties are recommending they offer to patients.
- jacquesm 6y agoBut medicine once you get to application of drugs rather than anatomy (setting bones, surgery) crosses over into another domain: chemistry, and at it's lowest level chemistry in turn is physics. Doctors need to know how to diagnose an issue (which often involves other disciplines within medicine for instance radiology), and then what particular medication or procedure to prescribe to solve the problem. The implementation details of that medication are not that important until you get to side effects, akin to how a programmer might select a library to perform a certain function.
- Kalium 6y agoPlease accept my apologies. I can see I have been unclear. It's been my experience that when attempting to communicate, people use the vocabulary they expect will do the job efficiently for the task at hand. As you say, for many things this is absolutely pseudocode! But I've also been in situations where mathematics, or the many Greek letter transformations of Haskell, or something else was the vocabulary of choice. As the projects I've worked on have grown more complex and occasionally abstract, pseudocode has increasingly not been the go-to answer. Again, you're completely right. It's a major issue that medical professionals often lack a real understanding of how a given pharmaceutical works. It's one that is in dire need of addressing! I also think medical professionals have this shared assumption that they can talk about anatomy and physical processes with one another to communicate effectively and precisely, rather than "the bone next to the other bone next to the chompy thing". Especially when on a chart and the other professional isn't nearby to ask. That's shared vocabulary at work. I've certainly struggled to have architectural discussions with people with whom I shared little vocabulary. I'm sure the failing was entirely mine. You're completely right. There's a lot here and pseudocode is incredibly powerful! I find it worth considering that there might be value to be gained in shared vocabulary beyond that one tool. Your mileage may of course vary. Thank you kindly for the opportunity to better explain my points.