Ask HN: How do you know when it's time to drop your specialty? | Hacker News Reader

Ask HN: How do you know when it's time to drop your specialty?

15 pointsohjeez8 comments

We've all had good career runs (for varying values of "all" and "good") working with one particular technology or another. But at much as we may have loved it, there comes a time when it becomes apparent that that particular string has played out, and it's time to figure out what's next. How do you know when that time has come -- and how do you figure out what your Next Thing is?

Case in point: I was once the "OS/2 goddess." I taught classes in operating systems internals, I wrote books about how to use it, I was OS/2 network administrator for the electric company, I was a lead perpetrator in the user group community.

But when OS/2 went away, so did my career. I didn't want to admit it; after all, I was comfortable with that technology; I had great contacts and it was easy to do what I needed to; and I loved doing it.

However, at some point, there's a "come to Jesus" moment. You tell yourself, "It's been a good run, but now it's time to move on." In my case, it was when I realized I was getting gigs where I was paid to remove OS/2 from a business, not install or maintain it.

...And then you have to think about what to do _instead_, and the lessons you should take away ("I'm not going to hitch my career again to any company's single product").

Sometimes the path is an obvious one (if you can write about one technical topic, you have enough cred to write about another). Sometimes, not so much.

What caused you to recognize "I need to change my specialty" and how did you decide what to do next?

I haven't had a distinguished career in tech, actually I've never even had a job (I build and ship software as an entrepreneur).

I think your perspective of needing to have a "specialization" is a bit overweighted. First, how did you become a "specialist" in the first place? Go back to the beginning of the path. You probably toyed around with some technology, enjoyed it, and kept doing it. You ended up building a relative knowledge advantage over your peers and fulfilled roles accordingly. However, if you were sufficiently educated, you know that the fundamentals of CS/technology always stay the same. That means even if you specialize, you should be able to catch up to anything else without issue.

So rather than looking externally for signals about whether something is becoming dated, it might be easiest to just have a breadth of toys and side projects. You will see things you like, and they will pull more of your time and attention. If those technologies compete directly with your legacy frameworks, you'll know why X is better because you've used both.

Being flexible and broad like this allows you to change direction quickly, at the expense of being a single-minded, deep specialist.

I wonder if many people who experience this make a switch or just leave the field entirely. I started in mainframe cobol around y2k via a financial services companies boot camp program. Switched to java and j2ee not long after. It was a pretty good gig for someone with bs in “field that no one hires for”. The thing that prompted me to jump was the company starting an offshoring initiative. I left, retrained myself, and have a career for 15 years after most onshore devs at that company were laid off long ago.

I'm probably not a good example, I've usually been a "jack-of-all-trades" kind, but it's very often the very same way you did - you see jobs stop asking for a particular tech and start asking for others (and it's probably too late to make a killing by then).

No-one is asking me about my Novell NetWare experience these days :P . I do however work mostly with Microsoft products and they've given me a good run (from vb6 to .NET and my old SQL skills still serve me well).

I do sometimes wish I'd picked up a fad faster (Ruby several years ago, React a few years ago, Blockchain last year and AI/ML this year).

I used OS/2 back in the day, when windows 95 came out, most people knew it was over for OS/2.

Generally you should already know its on the downslide. Easiest method IMHO is job listings. Search on Keywords of what you know.

As for what to do next, is a very tough. Anything you hitch up to can turn south. FWIW I looked at relational databases (before I was using OS/2) and decided there was not enough future in it. OMG was I wrong!

That being said, I suspect the real issue is how do you keep moving? The ideal answer is pick something broad, then hon in on a part, but be able to move to other parts. You were OS/2, can you leap forward to Windows or Linux, then specialize in a particular version of either? Everything you know about OS/2 is not dead knowledge.

From time to time, my wife asks me "What's the next thing you need to learn for the next five years of your career?" It's a good question to ask, say, every five years. (BTW, my current answer is "Android", but I haven't gotten very far on learning it...)

I learned Android but it hasn't served me much (I had no plans on how to use it !!!). I do recommend the Coursera course.

It's a good question to ask.

I was a J2ME Developer when the iPhone launched. Now I’m an iOS developer.

At what point did you say, "I'd better pick up iOS, if I want to have a career"? Was there a moment, or did it just make sense to add another tool to the tool belt?

Reply on news.ycombinator.com