6 ms·

I resonate with this a lot. There is a time horizon beyond which the work we do doesn't really matter. No code I wrote five years ago is in production today. A

• normie3000 2y ago
  > No code I wrote five years ago is in production today. This is wild! How can you be so sure?
  • taspeotis 2y ago
    He's a JavaScript dev
    • solarized 2y ago
      ROFL !
    • jaggederest 2y ago
      Funnily enough my first gigs around the turn of the century were in ASP/JScript! As far as I know that code is still running in production, NT 4.0 or something?
      • joquarky 2y ago
        ASP/VBScript is also still a heavy lifter in some industries.
    • Jerrrrry 2y ago
      My javascript may had been gutted, but my HTML and CSS live in Word documents across clipboards immortal.
    • ralphc 2y ago
      Then it would be 5 months.
  • raverbashing 2y ago
    That's a funny statement I'm sure-ish some of my code from 10 yrs ago is still in production. Though I'm far removed from where it actually runs. Maybe even some of the older code I wrote (which at this point is ~ 15yrs old) might be still around (at a different place) 5 yrs ago? No problem. I even have to still maintain the darn thing. Of course, different companies, different cases, etc.
    • rblatz 2y ago
      Agreed, I believe code I wrote over 20 years ago is still running. It wasn’t good, but apparently that’s not as important as we tell ourselves.
      • GrumpyNl 2y ago
        If its still running after 20 years, its great code.
        • Clubber 2y ago
          Non-web code probably lasts longer because it isn't on the front line of ever more sophisticated attack vectors.
      • dangedbyidiots 2y ago
        [dead]
    • jaggederest 2y ago
      I'm in the weird position of code I wrote 15 years ago still being a common thing I use in production on applications I run today. I worked for a monitoring company that still uses the distant ancestor of the code I wrote, so if I am monitoring applications using their library, it's still got a lot of my code in it. The oldest code I think is still in production I wrote in 2002, so it's old enough to drink. Still powers a web form as far as I am aware. Funny story - it was in ASP/JScript, so I've been doing "full stack JS" for close to 25 years now?!
    • 2y ago
  • 8organicbits 2y ago
    Now I need to track my code down. I think I can find open source code from 14 years ago still in use. Unless the product is defunct or all your code is rewritten, I think it's common that code just keeps running.
• self_awareness 2y ago
  You think like this today. Would your life be better if you had this stance for your last 30 years? Would that be your advice for young adults who are just entering life? I guess you were lucky that you didn't have any cases of cancer in the family (if you had, then I apologize and I'm sorry). No important problems that are only easily solved by lots of money. But if it's just luck, then it's really inappropriate to build a world view based on what you were lucky enough to avoid.
  • djtango 2y ago
    Could you help me explore your view a bit more? Parent's advice is about favouring team and striving to be an enabler/multiplier rather than sweating over individual perfection of craft. It's a fairly common conclusion that some people arrive at. What are you contrasting that to? How does that relate to money problems and health problems? Is becoming a multiplier mutually exclusive from chasing financial success in your opinion? EDIT to elaborate on the concept of a multiplier. Say you can become a mythical 10x engineer or you can become a mentor. If you can mentor 5 normal engineers from 1x to 2x then you've now got 10x engineering and the org also has reduced bus factor. What about if those 5 2x engineers can bring another 25 0.5x engineers up to 1x engineers. Suddenly your org can become much more than a single 10x engineer can accomplish. This is quite spurious and 9 mothers can't have a baby in 1 month but it's a useful mental model for the value of individual contribution vs coaching/managing. In reality while people like to balk at the idea of the 10x engineer, the truth is there are probably 100x and 1000x engineers out there. Von Neumann is probably a 1000x or more. Linus and Page&Brin are probably at least a 100x etc - those multiples are made up, power laws make numbers go big
    • self_awareness 2y ago
      I never had the pleasure of working with a manager that cared about me, nor I had contact with any other form of a "multipler" person, other than my parents. So I don't think I'm a good candidate for discussion in this subject, because obviously I won't really know what I'm writing about, and I'm certainly not knowledgable about what you care about. I only had the impression that the OP tells his subordinates that "they shouldn't oversweat, do some meaningful stuff but remember it won't really matter in the end". This is while working for a company that was created by a person who has probably worked their ass off, and maybe even sacrificed their private life for pursuing a dream of having a large company. All I'm saying is that money, success, power, connections are so important in life that they give happiness even for other people. If we're poor, then where do we go for help? To those who are richer or more powerful. They enable our happiness in our time of need (that is, if they choose to help us).
• m463 2y ago
  Most of the companies I've worked for no longer exist, let alone the code. Even stranger, I've never had anything to show as the code never "existed" in the physical sense.
  • drmpeg 2y ago
    None of the companies I've worked for exist today. Even some of the buildings are gone. I worked at 1188 Bordeaux in Sunnyvale (at a kooky company called Microtest where I used to smoke weed with the CEO and CFO), and now it's been replaced with a Google parking garage (and Sunnyvale Fire Station #5 next door).
    • scruple 2y ago
      Every software engineering company I have worked for (5) still exists today. One of them was started in the '40s, so it clearly wasn't always a software company, and wouldn't strictly define itself as one today, either, but software is a critical component of their product lines so whatever. My current employer has been around since the '80s. Another was started in the late '90s. The others were startups that are younger.
• masfoobar 2y ago
  > I resonate with this a lot. Agreed. In my 20 years as a dev.. I am sure my fingerprints (code I wrote) still exist in past jobs... the past companies I have worked for. Sadly, going back to my first job, I know the project I wrote came to an abrupt end a few years after I left. I am sure that codebase has long since disappeared... though I might still find some code in USB sticks, somewhere. Even my previous job, some front end gui projects are likely terminated due to structural changes in the company. Going all-cloud with a product which would have changed bespoke applications. All comes down to experience. If I could go back in time to my first job, I could "hack" out that code in days.. rather than months, but I am diverting the topic a little. lol. > These days, I take the most pleasure out of building careers and mentoring younger engineers Again -- agreed! Doing my bit to help find their feet is very rewarding. Being in my 40s and seeing "kids" that are 10, 15, or near 20 years younger than me is a reality-check for me! It comes a point when my humour in the room no longer lands... because they have no idea what I refer to. I nearly get a heart attack when I hear someone say "I have not seen Robocop" The other reality-check, and a more serious one, is that I am at an age of guidance... for them. I remember being their age, with little help or guide from the older, senior devs in the room. I swore I would never be like this. Well, I am now at that age they once were. I know at least a couple of guys I have worked with, for different reasons, have moved on and are successful. If I can claim atleast 2% of their good decisions.. I will take it. Very happy for them! Getting good pay is one thing. Mentoring new careers is heart warming.
• _steady 2y ago
  Really reminders me of the following phrase: "When the work is done, it is forgotten. That is why it lasts forever." - Tao Te Ching (Chapter 2)
• the_biot 2y ago
  > No code I wrote five years ago is in production today. Same. Sometimes the company disappears, sometimes they just stop maintaining your code after you leave, and it eventually dies. The only code of mine that has stuck around is the open source stuff. And even then... a large open source project I started 15 years ago was wildly successful, but after I left (with another maintainer in place) it languished, and one of these days I expect even its website and git repo will disappear. It's discouraging :-(
  • joquarky 2y ago
    Accepting impermanence is difficult but liberating.
• RealityVoid 2y ago
  Based on your post, I was curious how long the stuff I write will be around for. I write software for cars, so that means a lot of legacy stuff, I've seen code written by coleagues hang around in the next generations of product for about 10 years. Then you have the life of the vehicles that varies greatly, but at least _some_ of the vehicles will reach 25 years, so I'd expect at least some of the SW I write to linger around for ~35-ish years. Not bad. A bit unsettling when I think about the varying quality of software though.
  • jeffreygoesto 2y ago
    Can relate. My in-vehicle SW is driving for 20 years and counting (and it is not just some standard AutoSar stuff but high performance video processing pre "A" "I" ;) ). When I see cars with it I always remember some funny anecdotes and good fellas that I was lucky to share some journey with. Making a durable dent feels satisfying for sure, but also doing "cyber physical" projects and moving real things around.
    • aYsY4dDQ2NrcNzA 2y ago
      Dent.
  • scruple 2y ago
    Yeah, I worked for an embedded company for ~8 years and there was a lot of old code kicking around that had been carried forward across product iterations. I know, from talking to colleagues who are still there today, that code I wrote ~20 years ago is still kicking around here or there in the current offerings. > A bit unsettling when I think about the varying quality of software though. That's just the business.
• poisonborz 2y ago
  It is not necessary to be like this. A lot of code is used for decades after it's written, a lot of software is influential in its own right. I'm not even thinking of "Fortran for government" situations, but things like IrfanView, FancyBox javascript module, or Magic Fields Wordpress plugin. Not to speak of niche aoftware tools for specific communities.
• jollofricepeas 2y ago
  I don’t know. I had a similar perspective on gratitude when I was broke. I wasnt unhappy necessarily. But when I became “rich” things definitely changed and I don’t even have Woz money. Being “rich” helps in that you don’t worry about money anymore at least not in the way you did before… But as Larry David said as soon as “making money” disappears as a worry than something else tends to take its place in your head. A good example would be “losing money” or “trusting others.” The biggest goal should be becoming debt-free and avoiding consumerism. If you can do this you can reduce money as a worry and focus on gratitude as Woz says. It can be hard to be happy and find direction when you’re broke.
  • madrox 2y ago
    I wouldn't say this is about being rich or not. My father helped shape my perspective, and as he used to say: "I've been rich and I've been poor. Rich is better." [1] He literally grew up in poverty, but he saw his legacy as his family and not his work. I think there's a subtle difference between living a happy life and looking back being proud of the life you lived. Of course having money is better. Money opens up possibilities for the future. However, on your deathbed when there isn't much future left, possibilities don't matter much and all you're left with is wondering what your legacy will be. In the difference lies hedonism. 1. Like most of the things he said, he got it somewhere, though I can't find a consistent attribution. The oldest seems to be Beatrice Kaufman.
• yencabulator 2y ago
  > No code I wrote five years ago is in production today. This is one of the reasons I'm a free software enthusiast. Outside of capitalism and trade secrets, programs I wrote 10-20 years ago can be just fine.
• scruple 2y ago
  I think this is kind of sad, but I guess I'm also very fortunate in who (most of) my employers have been. I've spent a large portion of my career working in codebases alongside stuff from the '90s, and with my first employer the things I was doing ~20 years ago are still in production systems today, right there alongside the older bits from the '90s in many cases. I saw some comments in code last week from the founders of my current employer, written back in the '90s when I would've just been entering high school. That code and much more is still there in a product used and beloved by millions. At an employer in the middle there, the oldest code was from the mid-2000s but it's still kicking around. Code I wrote there a decade later is still in production use today. I can only think of a single employer where this isn't the case and it was a shitty startup that I never should've accepted an offer for.
  • gopher_space 2y ago
    As far as I'm aware the work I've done for institutions that still exist is all largely in place. I wouldn't consider this unusual or surprising in most industries, and certainly not a sign of exceptional competence on my part. Life outside the FAANG-adjacent-or-worshipping community is completely different.
    • scruple 2y ago
      Yes, I didn't think it was exceptional at all. I thought it was just sort of mundane that this happens. But apparently it's rare?
  • madrox 2y ago
    That's great, I guess, but the point is that all work is ephemeral. Whether it lasts five years or 100, there's going to be an end date. When you're on your death bed, is "my code was in production longer than average" going to be what you want your legacy to be? That's what we're talking about here. Sure, it'd be nice if I built a knockout startup with code so great it lived beyond me, and that's a worthy thing to pursue. However, in the end looking back you may, like Woz, find that was less important than other things.