Ask HN: Is CS research mostly discovery while novel game dev mostly invention?
1amichail55/22/2025, 4:41:38 PM
Should students use this as a guide when deciding between the two?
Comments (5)
sargstuff · 1h ago
Using information provided:
CS commonly involves discovering/developing corresponding enumerations / associations needed to support/to focus on given science field/area. Depending on how 'novel' support is required for given field/area; may have to 'invent' necessary protocols/language/etc. based on required enumerations/associations 'discovered'/needed for a given science field/area. Very helpful if students also look at non-CS areas would want to use CS classes for. aka genomics, web back-end development, AI, etc.
Novel game dev[3] "invention" as in (?) software engineering'(?). Implication of novel taken as 'on own'/free lancing. Game developer more full process/multi-hat development / broader set of process(es) within given field than CS. Still have to 'discover' what 'sells'/'client(s) interested in buying'. (vs. directed discovery bounded by science area/field) . aka ID Software's inventive development approach[0] vs. discovering 'novel ways of using' doom game engine[1][2].
Why do they have to be dialectical in the first place? What is your reason for trying to categorize either career? Do you have any supporting evidence for your claims?
We can't meaningfully discuss these questions without more context. Your title is a loaded question; most people don't try to label each archetype because they can be radically different people. By presupposing that they exist in different worlds, you're already making this question hostile to interpretation and response. Please, if you have to beg the question like this, at least illustrate your thought process so we can understand what you're trying to say.
amichail · 9h ago
I think students don't think enough about the difference between discovery and invention and this could result in choosing an ill-suited career.
While scientific discovery is important, it might not be as fun or intellectually rewarding as invention for most students.
inhumantsar · 8h ago
"Discovery" and "invention" can be fuzzy ideas. In the past I've framed this distinction as "science" vs "engineering" since it links directly to a student's classes and possible careers.
Typically I reach for material (or rocket) science and aerospace engineering as my analogy: "Do you want to build rockets or dream up better rocket engines?" What I like most about the analogy is that it gets people thinking about their options without implying that the choice they make today has to define their entire career.
After all, there are plenty of rocket scientists and aerospace engineers working in industry, just as there are plenty of both doing research at universities.
Side note: The super specific "novel game dev" thing is a little weird to me. There is as much invention happening outside of game development as there is regular ass work happening inside game development.
sargstuff · 23m ago
Stepping back a bit, looking at how Humanities Architecture (HA), Construction Science (ConS), and Engineering (EGR) works as bnf rules of lisp[1]. HA defines/enforces s_expressions. EGR provides/validates the atomic_symbol(s). ConS evaluates the HA directives using EGR atomic_symbol(s).
Less confusing visually if use Autolisp / Autocad framework.
---
CS ~ ConS; HA ~ Game Developer; EGR ~ CS/Game Developer depending at what hareward/software level working at.
CS commonly involves discovering/developing corresponding enumerations / associations needed to support/to focus on given science field/area. Depending on how 'novel' support is required for given field/area; may have to 'invent' necessary protocols/language/etc. based on required enumerations/associations 'discovered'/needed for a given science field/area. Very helpful if students also look at non-CS areas would want to use CS classes for. aka genomics, web back-end development, AI, etc.
Novel game dev[3] "invention" as in (?) software engineering'(?). Implication of novel taken as 'on own'/free lancing. Game developer more full process/multi-hat development / broader set of process(es) within given field than CS. Still have to 'discover' what 'sells'/'client(s) interested in buying'. (vs. directed discovery bounded by science area/field) . aka ID Software's inventive development approach[0] vs. discovering 'novel ways of using' doom game engine[1][2].
----------
[0] : https://charlesboury.fr/articles/id-software-principles.html
[1] : psdoom : https://psdoom.sourceforge.net/
[2] : doompdf : https://github.com/ading2210/doompdf
[3] : https://howtomarketagame.com/2023/09/28/the-missing-middle-i...
We can't meaningfully discuss these questions without more context. Your title is a loaded question; most people don't try to label each archetype because they can be radically different people. By presupposing that they exist in different worlds, you're already making this question hostile to interpretation and response. Please, if you have to beg the question like this, at least illustrate your thought process so we can understand what you're trying to say.
While scientific discovery is important, it might not be as fun or intellectually rewarding as invention for most students.
Typically I reach for material (or rocket) science and aerospace engineering as my analogy: "Do you want to build rockets or dream up better rocket engines?" What I like most about the analogy is that it gets people thinking about their options without implying that the choice they make today has to define their entire career.
After all, there are plenty of rocket scientists and aerospace engineers working in industry, just as there are plenty of both doing research at universities.
Side note: The super specific "novel game dev" thing is a little weird to me. There is as much invention happening outside of game development as there is regular ass work happening inside game development.
Less confusing visually if use Autolisp / Autocad framework.
---
CS ~ ConS; HA ~ Game Developer; EGR ~ CS/Game Developer depending at what hareward/software level working at.
-------------------------------------------------------------
[1] : BNF rules of LISP : https://cui.unige.ch/isi/bnf/LISP/BNFlisp.html