Ask HN: At 34, do I go back to college or not?
So here is the deal: I want to build a startup in 2-3 years dealing with custom-made prostheses, and I'd like to hack the technical side myself. My problem is education/background in robotics to pull it off.
From my initial investigations, I'll need a basic knowledge of mechanics and electronics (nothing fancy, it is basically a solved problem) and a more mature grasp of computer science (possibly some LISP to wrap everything). This is quite a stretch from my technical background (basic Calculus, basic FEA and some Python).
Here are my options: 1) Self-study (more targeted and possibly faster; requires self-discipline; no feedback; isolation) 2) Follow a 2-year technical college in mechanics and/or electronics (good practice; requires complementation) 3) Get a 5-year degree in mechanical or computer engineering (nice, but too long; lacks practical side?) 4) Combinations of the above (e.g. drop college after 2 years + self-study)
It certainly looks challenging, but my goal is not only making money in a fixed time frame, titles or jobs, but the journey itself. Can I get some advice?
74 comments
[ 12.7 ms ] story [ 265 ms ] threadWhile on your own time, you can dedicate 30-40 hours a week on your studies.
There are times when a structured class gives you more benefit than unstructured. Specific examples off the top of my head are like sciences where you need lab equipment. Or if you're the kind of learner who needs lectures rather than readings.
School is not as worthless as HN makes it out to be.
Is it stealing? The professor would be doing the exact same thing with or without you there. As long as you don't take up a seat in an otherwise full lecture.
I think the lesson is choose your professors, classmates, and class sizes carefully.
Eventually, you'll develop a sense of what's right and wrong (from your own experience and looking at other people's work), and then you'll have the tool you need to do the project you actually want to do -- experience and intuition.
It reminds me of that essay, "Teach yourself programming in ten years". You can't read a book and become an instant expert. You have to read books AND fail for yourself AND learn from your failures and success. You aren't going to get this skill in college, you can only get it with your own practice.
My problem with graduate studies is that your mentor always has an agenda (a paper, a prototype, etc) which may compromise his advice (e.g. skipping really important courses/books to save time). Therefore, I'd much rather focus on technical/undergraduate courses, if following any at all.
That will give you a view-from-the-bottom perspective on CS; perhaps then you'll figure out if you need to climb to the top.
For example I prefer to understand quantum mechanics from a book, but I would rather somebody showed me how to truss a chicken.
EPID and all that, but I think there is a huge shortcut to be had by spending a little time with somebody who has the tools and skills you are trying to pick up. The rest you can figure out for yourself.
There's lots of fascinating systems/control theory that you can 'climb up' from there. You'll also find that part of what the body does is similar to those crude electronic systems. And a lot of the programmed digital control is emulation of the behavior of analog, but often with bells and whistles impossible in simple circuits.
My thoughts are that you are financially stable to afford the formal education, but you are limited by time - once you finish up a 4-5 years degree, that puts you are almost 40. I mean, your life is far from over but I assume that your priorities will change. Instead of thinking about your startup, you would probably be thinking about your children's educational future.
But less talked about it is the "journey" part.
As long one is in touch with personal motivations and goals, then college seems fine.
You're in no rush, financially comfortable, and as eager for experiences as you are your innovative product. I'd say college sounds like a cool idea. I'd just be careful to pick a good program, otherwise it could be painful/frustrating.
I'm about to go to college part-time with no professional necessity at all. I'm in a different spot than the OP. I don't have a first degree. But I do have a fine geeky career that I'm about to step out of to start a business (way less capital intensive than his idea :D ), and am going to college with the hope of learning a subject matter, but no real expectations. Just mostly for the experience.
I've had mixed response from the schools I've talked to about this. The admin folks seem mildly confused/insulted, but the profs are cool usually. It does take more overhead to explain your situation, meet w/profs to get out of prereqs, but it's pretty do-able.
Hah, yeah. Mostly because of each side's interests: the job of the administration is to sell the idea that $university's degrees are a valuable thing you should want, so it is completely crazy to them that you don't want their degree. But the profs ideally want to teach stuff they think is important/interesting to people who want to learn it, and to them, the people taking the class only because it's required for the degree are sort of the least desirable students, while someone taking it even though they don't have to, because they actually think it's interesting, is great.
This is just one man's opinion. Let me know what you decide.
A surprising amount of those who have their head around making things have done the thinking and the math to find that, hey, the risk/reward ratio of a startup might be worth it, particularly at a young age.
Don't go back to school though. I think its a bit overkill. Having a PhD you already possess the skills to find the right resources and just begin doing what you want...no need to relearn the theory (unless that is half the fun for you). To address the isolation concern, I would suggest to find a hardware company (perhaps medical) that you could intern at. Try and find a mentor who can show you the ropes quicker than if you were to be by yourself.
Missing/Better versions from the list are: http://see.stanford.edu/ http://education.jimmyr.com/ http://itunes.extension.harvard.edu/
San Jose State University keeps a good list of local surplus depots here: http://www.engr.sjsu.edu/bjfurman/courses/ME106/componentinf...
If you are good at motivating yourself to learn you don't need it. Also most colleges you can just attend the lectures for free anyhow, you only pay if you need the degree. Just get some books on the subject, and find an online community to bounce your ideas off and your set. That way you can learn at a faster pace and not be dragged down by the speed at college.
I don't remember the names of early stage life sciences VCs off the top of my head, but if you're interested, I'd be happy to do some digging this weekend for you.
That being said, a younger co-founder with the engineering and /or business skills, good stamina and character would be certainly great. I hope to find someone like this during my preparation.
If you do go for a degree, I'd suggest electrical engineering. I know many great programmers whose degrees are EE, and it sounds like you'll need some expertise in that area as well. (No, I'm not referring to people from schools where CS technically falls under the EE department, like MIT -- though a joint program would probably work too).
My problem with the full EE course is that it will eventually require me to go into areas absolutely not necessary at all. But the same can be said for all other options as well.
This has a couple benefits - you can churn out a basic prototype, and live "the journey" as you want. You can work hands-on, directly with top students, and go through the learning process with them. You can learn from them, they can learn from you - it'll be more relaxed that starting a company from the get-go.
This might also have practical advantages later on. If you decide to go for VC funding later on, you'll know both the software and hardware sides more intimately, and you may even get a higher valuation for having a legitimate product rather than a paper product.
You may even be able to talk the professor into letting you audit a few classes that you might need. Maybe you'll even take classes with the students helping you!
Sourcing components and tools as a hobbyist, garage inventor, or even small design firm is very different from sourcing parts as a large institution or as an enterprise. The factors of scale change the playing field drastically. Many of the smaller kit builders and maker types have a shared knowledge regarding this sort of limited run development that you are endeavoring to pursue.
That being said, the key benefits for you that a major scholastic environment can provide you outside of tutelage is access to a robust machine shop, capable electronics facilities, and more important than all of that... protection from litigation in the patents you file.
Many graduate and even undergraduate programs offered by major universities actually will stipulate that any patent you file while at school, regardless of whether it's a personal patent or something completely unrelated to school, will be protected by them, their legal war chest, and their patent warchest. This is a huge selling point for some schools. ASK ABOUT IT.
Also, many schools have start up incubators of their own, or a network of people who can help you find funding. A friend who left stanford graphics lab took advantage of their start up program where they offer you use of facilities and space on campus in return for a percentage stake in the venture... they also will provide you with assistance in using their alumni network to score capital.
These are just some immediate thoughts. But I definitely see some values in going back to university that have no relevance to the actual classes you might take, but more with getting things done in your R&D.
Definitely regardless of what you decide your level of commitment is, don't get discouraged. There are far too few people with medical certification and experience turning their eyes towards prosthesis and other technologies that assist the disabled. And while there's money to be made there, more importantly there is a lot of good that can be done. The things that help someone who is confined by their disabilities can be for more huge to the people who use them, than us.
I know a number of engineers, good ones that would be interested in helping people the way you are suggesting. It's a job that one can draw immense satisfaction from.
Whatever people on this community say, CS is significantly different from Hacking or writing some web app in PHP.
A good MS program at a top school will lead you to deeper understanding of CS. The general opinion on HN is very anti CS education.
However if you truly want to build for future (think of Google and Intel and other) then nothing can replace firm knowledge of CS.
"An expert is a man who has made all the mistakes which can be made in a narrow field." -Niels Bohr
The matter here is trying to get to expert level while minimizing stupid mistakes, and for practical subjects, demo and feedback seems to be necessary. Making all possible mistakes is very time-consuming.
I guess it ultimately depends on how comfortable you are. Whatever you do, try doing it the "lean" way. Try taking one or two classes first without jumping into pursuing a degree. A lot of universities allow concurrent enrollment so you'd just take a class alongside regular students if they have room. Good luck.
I'm no expert on this, but isn't the role of a design agency to advise on these sorts of problems?
This should give you more information on which to base your decision.
Learn while doing, or more specifically while someone else does and teaches you. You're clearly highly skilled in your technical niche why don't you find someone else with complimentary skills (mechanics, electronics and computer science) and partner with them. You can start building your business immediately in the time that you have and as part of the partnership each teach the other. I know that as a hacker having the opportunity to work with a real specialist in another field and for us both to teach each other would be a massively satisfying experience.
You get two benefits the way I see it 1) You can start building your startup now 2) You get the two skilled specialists to work on it and share the excitement
I see the only drawback is that you don't get to go and spend a few years learning a new thing without any commercial pressure. If this is about learning how to be a hacker that's bad but if you're really into the startup thing then you save yourself a lot of time.
Or as you say it's 'basically a solved problem', maybe you can just find someone and work together. Just lsat week I was recommending HN to someone with a Mech eng degree and who has family manufacturing contacts in Asia to get on HN because he's a great hardware hacker but doesn't enjoy programming so much. Depending on your location, I bet you could assemble a small research team here on HN fairly easily. I wish you good luck with it, as giving mobility to people who've lost limbs or similar is a very worthy pursuit.
As for the headline question of 'college in 30s' I say sure. I turn 40 this year and am returning to study, which will be a 4-5 year process.
I sure wish I'll run into someone like you when I'm done with my Masters.
For now, may I suggest you check out the Springer Handbook of Robotics for an excellent wide overview of the field, both in theory and applications. Maybe seeing it laid out like that will help you decide where you want to go with it.
#2 this is a waste of time for someone with your intelligence imho #3 - yep, too long #4 - possibly, if you're not a self directed learning type.
I would vote for #1, if you've already got a PhD the you have what it takes to pull it off.
As I understand it, you want to work on robotics. Very many of my friends who do research in robotics got their start playing with lego mindstorm robots. This would probably be a good place to start, because you basically can't learn software dev without writing something and it will give you a firm grasp of the basics of writing software for robots.
It's not clear what your level of experience in writing software is, if you haven't gone through basic tutorials then there are plenty of those out there.
If you happen to read this post and there's one thing to take away, it's that the absolute most important thing for ramping up to high productivity quickly is that you have a really solid infrastructure. The reason is that without such an infrastructure, codebases tend to drift and you can write yourself into a corner very easily.
I'll speak to my java infrastructure as this is my primary background.
First, I'm a big proponent of IDEs especially for beginners. I really like java with eclipse (it's unbearable with notepad) because with autocomplete and source code generation, you end up with code that reads more like an essay than a mathematical proof, which suits my personal style just fine. It just helps to offload my memory into autocomplete. Whatever editor you pick, it's extremely helpful to read up on it in advance.
Next is source code management. I personally use git and I create a repository for every project I'm working on, I find the distributed, self contained nature amusing. Commit often and branch when you change features.
Next up is managing dependencies. I use maven, which I absolutely love. Maven basically centralizes your dependencies and separates that concern to help avoid dependency hell. Some people use maven to manage their source, but I do it the opposite way with the maven files inside the directory where I create the git repo.
You also have to manage your own source. It's fine to play around ignoring all this, but once you have "real" code - this means code that you must maintain - then it's important that it be of a certain quality. You should try to have each project capture a particular need (read OCR or access a neural net or calculate some geo stuff for your robots or whatever) and that each product stand alone as much as possible. There should also at least be some functional tests.
imho, a relatively inexperienced dev with such a nice environment can be as productive as experienced devs flying by the seat of their pants. It also vastly reduces the risk of catastrophe.