Hacker News Reader

I like that Gary Bernhardt tweets every once in awhile something along the lines of "we have no actual evidence that any software methodology actually leads to better, less buggy software". We have people falling all over themselves to sing the praises of XP, TDD, Agile/Scrum/etc, and yet, we can't show software actually getting any better. I've been around long enough now to see these trends rise and fall. One thing remains constant though. Humans make errors. Humans try to find better ways to fix these errors (faster, more effectively). Humans like trying to notice patterns. It's a good thing. But trying to say any one of those patterns is more effective than another, is a hard sell. The result is always the same. The ONE lesson I wish all young developers could learn early:

You will learn new information that will cause you to question yesterday's "best practices". So quit holding on to what you know now as the "one true way". Be willing to learn from people you consider "old and irrelevant"... they may actually know something.

I agree with most of what you say, but this strikes me as wrong:

> We have people falling all over themselves to sing the praises of XP, TDD, Agile/Scrum/etc, and yet, we can't show software actually getting any better.

Software is absolutely better today than it was in the 80s and 90s. I used to have to reboot my computer 2 or 3 times a day due to unrecoverable crashes and taught myself to constantly save work after every minor change. Now my computer stays up for weeks to months at a time only being rebooted to install updates occasionally. Documents are often autosaved, and I don't even need to save anything when I quit an app. The next time I launch it's all there.

I agree that we may not be able to measure that any given methodology is better than any other one, but we've made significant progress along the way. Better tooling, like static analyzers and profilers have helped, too. I do think automated testing can be shown to improve things when used appropriately. (The drive for 100% test coverage seems fanatical to me, though.)

Damn, from personal experience I echo your feelings about modern SW development bullshit and I'm 30 years old, though I ground my teeth on embedded SW which is more conservative.

I don't think I can make it to retirement at this rate.

First, I had to wrap my head around something very simple: who one is (I am a software engineer) and what one does to acquire currency are orthogonal. This is simultaneously obvious and difficult to accept in modern western culture. I admit this took me a few years to digest.

Next, I had to honestly explore other aspects of myself. What else held interest for me besides electronics and software?

Further, of these other interests, which of them were reasonably "safe" from outsourcing, offshoring, and automation? I do not want to transition into new currency acquisition activities that are themselves a race to the bottom. Tangentially, I considered the perceived prestige of each of these. I could give two shits personally, but I've learned the hard way that the perceived prestige of an activity impacts how others will treat you.

Finally -- as I am a software engineer -- how can I use this skill to augment my effectiveness in performing any of these new activities?

This is an ongoing process for me. I've always had an interest in the law. Law school is out of the question, so I've looked into paralegal work and how I could provide a unique service in this space.

Additionally, there are mundane dirty jobs -- specialized trash hauling, electronic recycle brokering, etc. -- that are poorly serviced in many localities. I believe I'll end up doing something here as well.

I view this transition as open ended. It's no longer about "career" but staying in a lateral thinking mode about combinations of possibilities.

When I faced this question, I read Po Bronson's "What Should I Do With My Life" and it helped me decide. My solution was humanitarian logistics, yours will be something else. Go find it!

I've been through similar "searching" efforts and concluded the same.

What about working in legal or medical software, since those industries will never go anywhere? Or any other industry that is boring.

For me, the only model of acquiring currency from software engineering activities is via product sales (no more selling my time or expertise). To this end, I do have plans in motion; however, some very important points:

- the products I wish to build will take time to do properly and I do not want to be in any kind of rush;

- based on my research, I believe these products could be good earners; however, it will take time and the majority of the revenue will be from the long tail;

- this kind of activity is high risk and I treat it as such. What this means, in practice, is that I put in my 4 hours of focused coding every morning and then go do other, less risky activities;

My experience has taught me to prefer transactional business models. Consultative business models spring from the dark side of The Force ;-).

As a vague teaser: one of the products is a SaaS modernization of a classic desktop database/application dev program complete with a recreation of its programming language; and the other is a game based on the premise of "what is an episode of sleep paralysis and waking dreams like?" I'm excited to be working on both but I have my expectations very firmly in check.

>classic desktop database/application dev program complete with a recreation of its programming language;

I'm picking up what you're laying down and think this could be brilliant, especially given several trends in both tech and market demand today.

I share the same sentiment. I'm not even a graybeard yet, and I'm sick of the work culture/environment/processes. I'm in my late thirties, so I'm somewhere in the middle.

I've been around long enough to recognize that older, more experienced people are incredibly valuable to learn from. They also have one thing you can't shortcut: perspective, and knowing where the bullshit is and how to avoid it. There's shit I'm just not willing to put up with anymore, that I would have been eager to do in my twenties. It wears you down.

>I cannot work in open-office fishbowls. I cannot stomach Agile process and how it has turned something I love into menial factory work (N.B. I get it, your [A|a]gile shop is awesome. I only had such luck once). There are many more things about modern software development shops with which I disagree.

I followed you in the mid-80s, and one thing that bugs me is the lack of 'geeks' in coding these days. What used to be a passion you could make a career, is now a 9-to-5 shuffle for a lot of people. Seems like there's as many coders that are 'just in it for the money' as there are stock brokers. I guess that's to be expected with the growth of the industry over the years, but I still find it sad.

So the question is, what did you move on to? I'm echoing a lot of your sentiments internally, just haven't admitted it to myself yet. But what is left for us to go to?

> However, it does mean I have to move on.

To what, though? I came to the same conclusions a few years back, but I'm at a loss as to what to do next. So I plug along, take what work I can find, shake my head and watch the exact same cycle happen to React as happened to Flex.

I share a lot of the opinions of @nybblesio. By a miracle, I have stayed employed well into my 50s but I think I am reaching the end of my tolerance of conventional software development.

You simply can't expect to remain employed as a programmer until the conventional retirement age. Especially for younger developers, FIRE (Financial Independence and Retire Early) principles are essential to learn.

Amen Brother! I had the same revelation, that things have simply changed and I'm not willing to change with it. It ain't worth it. I made my move at the beginning of the year to a new career and while there are moments where I miss the cerebral joys of software, good riddance to the rest of it. Wishing you a bright future in whatever direction you go.

What have you moved to if I may ask?

I'm right there with you in my 40s. Not sure I was ever on board with the open-office and the managerial mayhem. This has kept me in contract work and startups. The general stability is lower with the rare but occasional lull between projects, but my overall happiness with my work and career is extremely high.

The question is, what are you moving to?

This is the hardest part. I think a lot of people feel this way, but it's taking that next step. Is it sys admin (or Zeus help you Dev Ops)? Is it some kind of database work?

The hard truth is development pays pretty well or at least better than average. Finding something similar at middle age is a challenge.