If you have ever had the chance to work at any or both of the two, which would you rather work at again - a research laboratory or a fast paced startup? In which of the two does one learn more and gets exposed to new ideas and concepts?
I worked for the Air Force Research Lab once as an intern and later while in a uniform. And now I'm jumping into the startup world.
Here are my thoughts:
1. If you're in a hot research area--the feeling of being a part of something bigger than yourself and the potential for impact is huge.
2. If you're in a hot research area and you're good--FYI good doesn't mean PhD... are you smart and do you get things done? Bringing ideas to the table helps too. If you fit these qualifications you'll soar to leadership in your area very quickly. I saw this happen with contractors, government program managers, and military personnel. I should note, the bench scientist part of a research organization is mostly flat so soaring to leadership is usually an informal designation. Still, its nice to see your ideas living and other people working on them either way.
3. Again, if you're in a hot research area (and the passion is there), then you'll get to work with cutting edge concepts and live there. Installing an SVN dump of open source package X is not cutting edge after working in research. At the same time researchers are much like the open source community, a big bulk tends to go in the same direction at the same time until the horse is shot, beaten, and then the bones picked clean.
4. Beware of the bureaucracy in a research environment. It is soul crushing and it will get you. It can really get in the way of everything you do.
5. The research world does offer you the opportunity to be entrepreneurial. Yes, really! Do you think you're good? Do you want your ideas listened to? Then go write a proposal and get funding. Getting someone to fund your ideas is incredibly validating (as I'm sure startup folks here will attest to). If you get enough funding, you can then afford to have people working on your team moving your idea forward. Maybe you start with a little money and then prove your concept enough to justify more funding later. Of course making relationships with parties interested in using your research helps too. Sounds a bit like growing a business doesn't it?
6. Oh last thing about research world--if you're willing to lead (if you're willing to start a company, you're willing to lead) then you won't just sit in a cubicle writing code. You get to write proposals, papers, reports. You get to go speak. You get to travel and reach out to other organizations. You also get to act as door to door salesman, I once went to Korea knocking on doors at US Forces Korea HQ peddling research. The number of hats you can wear is amazing.
I don't advocate doing one or the other. I do know if I make it in the startup world, I'll probably continue to spin multiple tops on the side, but I'd like to stay in the research world almost as a hobby. If one could eliminate the bureaucracy then it would be nearly a perfect job.
There is a big difference between being someone who starts a startup and working as employee #72 in said startup.
Likewise in the research world there is a big difference between someone conceiving ideas, trying to get funding, executing on them, and trying to transition the research (sales). Compare this to enthusiastic computer programmer who just wants to code and is unwilling to take on those other roles. YMMV
Then again, in professional sports, there is a big difference between being a player on the team and selling tickets at the door. Both are part of the same organization and both have a role to play.
In either environment you have to fight to matter but once you've had some success, then you get the freedom to make things happen.
Well said. There's one thing I would add if you're doing defence work which is about security. The complications brought on by working with classified data can be very frustrating: limited or no internet access, limits on the software you can use, hassle presenting results at conferences etc, having to double check everything you write/say to the outside world to make sure you haven't said something you shouldn't. And no, knowing secret stuff isn't worth the hassle - most of it isn't even interesting anyway.
I've done both. Could/should write an essay on this question.
Most important: I think that the specific qualities of a particular startup or particular lab outweigh the importance of the category "startup" or "lab". Good labs, bad labs; good startups, bad startups.
You can often leave a bad startup faster than you can leave a bad lab.
Anecdote: I'm in a lab now and I'm really glad I'm doing what I'm doing. I like it a lot. It has the feel of a startup, but I'm doing research.
I've worked in both and I prefer the startup. In a startup, you're a fast-moving cutting-edge company with people who really believe in your product or idea. Every meeting is energetic and your input matters.
In a research-heavy institution, it can be fun, but the scientists get most of the glory and creative work while the engineers just put the tools together for the customer or scientists. Research money tends to be less than startup money if the startup prospers.
I have been working part-time in fairly fast paced Japanese startup, and next year I graduate and choose to enter a company in research area. My decision is mainly because of culture and demand from my parents (the type of parents that ask you why you work at home instead of applying to "real" company like IBM, etc). One will learn more in area that suit him, no matter it's a startup or a company's research laboratory. One thing for sure before you make the decision is that you have to know exactly what you're going to do/face and whether you're going to like it. I learned a lot from startup but I felt that you don't get much time to learn deeply about new concepts and implement them. In startups, you get limited budget, limited devs team and basically you don't have enough time to implement/test new ideas. The most important factor for me is: since I'm not the kind of top level coder who knows everything, I feel insecure working at startup with lack of assistance from someone who really knows what they're doing.
> Nobody can guarentee the startup will be fast paced, if it's indeed a startup.
This point isn't repeated enough. I've worked with startups where the pace was crazy-fast: like, finish that impossible feature that 3 other devs have tried and failed at by tomorrow, because we've got a demo. I've also worked at variants of this where the pace is the same, but the direction ends up changing every 2 weeks, so you never actually get anything substantive done. I've worked at startups where you work really fast, in one focused direction, for a long time and then find out it was the wrong direction and jump off a cliff. I've worked at startups that were really fast in the beginning, and then bogged down as we discovered the problem was much harder than expected. And then I've worked at startups where the pace was the same or slower than a normal company.
Startups are like a box of chocolates: ya never know what ya gonna get. ;-)
Remember, almost all companies start as startups. For example, startups get to claim Ford, inventing the assembly line.
Of course, the potential for work in a research lab is huge too. So I think it's more about the specific opportunities. A web2.0 tagging social network is unlikely to change anyone's life right now, but startups like Kiva or 23andme might.