Software Engineers: Advanced Math & Differential Equations in Demand?

[andrew339]

Are you a software engineer using differential equations or more advanced math everyday to solve problems?

I would like to hear about what you do--what projects do you work on? Do you consider that differential equations and advanced math skills are useful/in demand in the realm of software engineering?

I am just starting this month my first semester of software engineering in university. I am doing right now my first course on differential equations, along with other courses.

Right now I want to assess if taking more math courses and even doing/reading some math on my own is a sensible idea, career wise. And if yes, where to start? (I constantly have (unwelcome) questions in class, some of them resolve when I read textbooks and watch videos, but some not, should I start posting here? experimenting on my own?... when I don’t even have the basics?). How did you learn math? Khan academy is awesome but I need to go beyond that and connect with other people who are interested in math. (and software). that’s why I’m here.

My goal, originally, in getting into soft. eng. was to contribute to the development of the web in some way, ie provide a valuable web service of some kind that could improve people’s lives.

Of course adv. math could be almost completely useless if I’m doing web design or working on user interfaces for example. But I have the vague intuition (but almost no concrete example) that advanced mathematics could be essential for my work as a software engineer later on. I think I base my reasoning on the idea that “there will always remain harder problems to be cracked” in the race for technological progress, and that the harder ones will require more math, in the future, simply because the easier ones (like the most obvious web services, not requiring any type of math) will have been deployed today (and in the past). it seems to me that the web will evolve into a huge global brain, basically, and the services available on it will match and surpass even the aspects of human cognition that seem hard to emulate with computers right now. in other words I see it as a mix of:

• VR and AR (the matrix? videogames?),
• the semantic web (tagging and reusing info efficiently for various purposes/services)
• wikipedia and wolfram alpha and other AIs/knowledge engines.

So where is advanced mathematics useful in software, now and (according to you) in the future?

• encryption
• data mining, statistics, semantic web??
• videogame graphics eg (improving Open GL?)
• artificial intelligence??
• … ?

 

[andrew339]

I received an answer by email but it's not showing here on the forum, so I will copy it now (below).

Yes this answer below is what I'm looking for. --I need to hear of mathematicians who are also programmers and what project they work on, even if they are really random. the more diverse examples the better for me, so I see what you can do when you master math concepts.

I want to estimate what "a true mastery of advanced math" will bring to my "software creation capacities" (concretely, by providing direct tools, not just "making me smarter").

I need to decide if I should invest a lot of time mastering adv math (and taking more classes in the next semester), or if I should simply learn what I am taught, to pass my obligatory math classes, and no more.

also I would also like to get some advice about what I should read or do to master, for eg, differential equations, besides learning procedures in class. I would like to get more intuition and general understanding (graphs, real world applications) on those types of problems. a book you liked?

are math nearly useless for programming? is biology more relevant to building software than mathematics? software is functions, organs, survival. how can math help?



basically to sum it up, I am very curious about math (and what it can do for software), but I don't know where to start to learn.

book suggestions, (to start accumulating math insights)?

math+software project examples to help me identify what is hot?

I would like to believe that if I master more math concepts I will see new possibilities, new problems and new ways to solve problems. maybe I am wrong and these conceptual tools won't really help for software.

it's true that from a certain perspective mathematics can seem like a somewhat sterile entreprise, and some famous mathematicians' lives and accomplishments can seem relatively dull from a certain point of view, compared to the exciting world of technology.

math feels like classical or contemporary art/music and software feels like film/pop music. it is obvious to me that something is wrong with the former. similarly, "could mathematicians be wrong about how awesome math are?" as Stephen Wolfram famously said in a TED talk?

is math sterile?

I think I retained the preconceived idea (from my "culture" and childhood) that math is the way to go, that if you can do it, it will be the most productive route.



please, examples. tell me what to read, where to start.

throw big picture ideas at me, too.

thanks in advance.

==
his thread is located at:
https://www.physicsforums.com/showthread.php?t=569731&goto=newpost

Here is the message that has just been posted:
***************
What exactly are you looking for?
Advanced maths are everywhere from the device you are using right now to software physics engines.

If you want to learn about a projec using advanced maths, you can check a car performance simulator, accoarding to the website it does more than 1000 different math operations each second all by runing on web, check it at http://www.nxgtrsim.com

Another good project I have used a lot is Kstars it is for the KDE desktop but I have seen somepeple runing on Windows, its is a planetarium programa I use on my Linux box, learn about it at http://edu.kde.org/kstars/
***************


There may also be other replies, but you will not receive any more notifications until you visit the forum again.

All the best,
Physics Forums

 

[jedishrfu]

Many engineering firms develop hardware and software solutions that require knowledge of advanced math. The programmer doesn't always need to know why only how convert an engineer or scientists algorithm into code and to then extensively test it under all likely conditions.

The open source physics software allows students to write physics simulations using a variety of ode solvers. They are already provided you don't need to understand how they work only how to use them. That's the way it is in industry.

Mos programming math is algebra estimating the amount of memory used or the time it takes to execute a program. Best ways to improve the speed. Next would come graphics wher you might rotate an object or vector draw some object. Usually there's API to do the task and you have to know how to do it not necessarily how it works.

Lasltly there's scientific/engineering programs that might use the ode solvers and some differential eqns to simulate a bridge or ship or something complex. A knowledge of diff eqns helps a programmer understand and write an effective program. Beyond that I haven't
Seen any other math beyond college sophomore year in use in industry.