6 ms·

I owe a great deal of my own personal success to Patrick McKenzie and Thomas Ptacek, both of whom have been steadfast, consistent and generous advisors (both in

by dsacco 10y ago

I owe a great deal of my own personal success to Patrick McKenzie and Thomas Ptacek, both of whom have been steadfast, consistent and generous advisors (both in public comments and in "hey can I bounce this off of you" emails). After following Patrick's writings and stories for a number of years now, I can confidently say that his relentless transparency has been one of the greatest gifts I received in the industry. His advice may not strictly work for everyone in the literal sense, but I believe that diligently attempting to use his suggestions as a template is, itself, a highly productive exercise in programming and business. There is one particular note I want to make about patio11's success: Patrick is a phenomenal marketer with remarkable business savvy who happens to be a programmer. He is not primarily a programmer, which is evidenced by his recent work at Stripe and the work he is best known for on HN (essentially, writing about shipping software, not the software itself). This is not to say he is not a good programmer - I simply can't comment on that, though I have reason to believe he is after seeing Starfighter's game. Rather, he leverages that skill set as a means to an end, not an end in itself. I think this is a really important point to make because I see many people who try to pursue significant career success by e.g. ranking up on TopCoder, or open sourcing impressive software. While those things can lead to success, there is a vast, long tail of people who are very capable programmers with no recognition doing those things. Healthy self-promotion and efficient improvement/maintenance of one's technical skills has a much higher probability of success than attempting to become Fabrice Bellard. This is demonstrative - in my opinion, the sum of all of patio11's advice can be summarized as follows: Don't be a programmer, be a $SOMETHING who happens to program, and program well.

sanderjd 10y ago
I've taken this lesson too, and I'm curious about what set of values for $SOMETHING is. In theory it is "everything", but in practice the two common values I see are "marketer" and "scientist". I really can't get excited about marketing, and it's difficult to justify going for a PhD. What are some values for $SOMETHING that others here have seen success with?
- __derek__ 10y ago
  Interestingly, neither of those came to my mind until reading your comment. I would fill in analyst, strategist, or operations specialist. Honestly, though, outside of the self-described "tech industry," understanding software acts as a force multiplier as long as you are able to talk about it. It's similar to having a finance background, except that your special power tends to produce actual gains more easily.
- SatvikBeri 10y ago
  Cases I've seen done successfully: –Accounting –Risk analysis in banks or fintech companies –Statistics ("Data Science") –Lots and lots of other jobs that have "analyst" in the title, e.g. marketing analyst
- thefalcon 10y ago
  Graphic design has worked well in my case. It's something I'm passionate about and truly enjoy, and dovetails very nicely with programming.
- ufmace 10y ago
  Depends on what you think "marketing" means. A lot of engineers think that it means basically building ad campaigns, running adwords, etc. Business people have a bigger definition, more like understanding the market that you're trying to sell to and building things that the customers want and need. That definition probably covers a lot of different possible job titles.
- cosmie 10y ago
  The reason "marketer" seems so common is because those who are good at marketing are inherently good self promotors. Making them far more visible than other, less common examples. Anecdotally, supply chain and business operations are two big areas where I have success. You don't even really need any domain knowledge to get started in those fields. I did a 6 month internship at one of the world's largest companies that has a consistently top ranked supply chain organization. They had the expense of a 10 person recruiting team (all expensive line managers and directors, not HR personnel) coming to campus for 3 days, and only gave offers to two of us. And I explicitly said during my interview I knew jack all about supply chain. Their response was that supply chain was their expertise and they had training for that if I needed it, but my general analytical skills and Excel knowledge were far more rare (in their applicants). Turns out they were piloting a new inventory management methodology and were using Excel-based "models" to trial it manually at a dozen manufacturing plants before deciding if it was worth the 8- or 9- figure cost to have SAP make that an official model in their system. It involved running batch exports from SAP every morning, putting that into Excel, updating the figures, and sending that out to the plants. But the pilot had scaled to a high volume plant that basically maxed out Excel, and they were struggling to make it work. All I did was essentially refactor a lot of logic to do batch operations (made a few VBA scripts that connected to the Excel spreadsheet with an ODBC connection and ran a series of SQL statements instead of in-cell functions). It's been about 5 years since then, and I still get calls every few months for contract work essentially doing the same thing (Excel got bogged down -> needs a refactor). Part of my day job involves essentially doing the same thing, but in a more generalistic sense on business operations. Process engineering is a portable skillset across verticals. Whether it's in supply chain, sales, marketing, or operations, there's always a bottleneck that's preventing growth or efficient resource utilization. I am a piss-poor programmer. I have never once been hired or paid explicitly to program, and I don't have enough experience in any particular language to pass even a junior level programming interview. I get paid to make irritations or problems go away while causing the least amount of friction utilizing whatever constraints exist. In the course of doing so I've tweaked C# and PHP programs. I've written bash, awk, perl, VBA, Python, and NodeJS scripts. I've written a whole lot of SQL. And I've refactored an incredible number of Google Sheets and Excel workbooks. I'm not valuable because I can program; I'm valuable because I can reliably and consistently resolve whatever issue is thrown my way in a self-sufficient manner, and ramp up on whatever knowledge or tool is needed to do so incredibly quickly whether I have prior exposure or not. I'm not a consultant that provides a report that then requires additional resources to execute; I provide a working solution (or prototype) that immediately solves the issue. And in the case of a prototype, provides a stopgap solution that (crudely) solves things in the immediate term and provides an explicit, working example of the functionality that their engineering resources need to incorporate into their internal tools/processes for a long term fix. One piece of advice that seems to serve me well: people hate change. And people also tend to hate relying on others to perform their job. The greater the change, or the more reliance on a programmer to make modest changes to a process, the more resistance and ill will you'll create. Engineers thrive for perfection, and invariably lean towards "optimal" solutions that are very invasive and supplant existing practices, rather than supplementing them. patio11 didn't create an entire Customer Management System or appointment management system that had automated appointment reminder functionality. He created a simple system that supplemented whatever process and system was already in place and added incredible benefits with minimal additional steps.
henrik_w 10y ago
Scott Adams has similar advice (for anything, not just programming): "Become very good (top 25%) at two or more things." http://dilbertblog.typepad.com/the_dilbert_blog/2007/07/career-advice.html http://dilbertblog.typepad.com/the_dilbert_blog/2007/07/care...
- dsacco 10y ago
  Yes. I've also seen this referred to as a "T-shaped expert" - someone who has a breadth and depth of useful expertise. Develop broad expertise in a variety of areas and develop deep expertise in a few specialized subfields.
  - lumpypua 10y ago
    "Paint Drip People": https://www.facebook.com/notes/kent-beck/paint-drip-people/1226700000696195 https://www.facebook.com/notes/kent-beck/paint-drip-people/1... HN Discussion: https://news.ycombinator.com/item?id=12226242 https://news.ycombinator.com/item?id=12226242
pryelluw 10y ago
This is not to say he is not a good programmer - I simply can't comment on that, though I have reason to believe he is after seeing Starfighter's game. Rather, he leverages that skill set as a means to an end, not an end in itself. Fucking this. I battled for years thinking I was less of a programmer because Im also a marketer. But holy shit does the combination work like a charm. It is like a 1-2 punch that knocks out the competition. Downright unfair. I love it. When companies that work with me realize that I can work in defining, building, shipping, and selling the product, they dont want to let go. Its such a powerful combination of skills that I encourage everyone to acquire them. If you code and learn to market there is a crazy amount of things you can do to make money. Its ridiculous how much.
- rjeli 10y ago
  Do you have advice for how a programmer can learn to become a marketer?
  - pryelluw 10y ago
    Sell something. Anything. Not just software or online. Just like in code, it takes 10,000 hours of practice. If you need help getting started shoot me an email. No bs or strings.
  - jtwebman 10y ago
    Build something small and try to sell it. Doing it is really the only way to learn it fully.
    - pryelluw 10y ago
      Building something is a common approach devs take. But there is also the option to just selling something you did not make. Its a shortcut to learning because you can jump right into the action.
      - minhaz23 10y ago
        What are some examples of selling something that you didnt make? Where can I go to begin? Can you also recommend any resources that helped you become a better seller and marketer? Books, lectures, etc? Thanks!
KirinDave 10y ago
It seems like wherever there is social and technical progress in our industry, Thomas Ptacek's shadow was cast nearby. I dunno how that guy has so many hours in the day.

by dsacco 10y ago

sanderjd 10y ago

I've taken this lesson too, and I'm curious about what set of values for $SOMETHING is. In theory it is "everything", but in practice the two common values I see are "marketer" and "scientist". I really can't get excited about marketing, and it's difficult to justify going for a PhD. What are some values for $SOMETHING that others here have seen success with?

__derek__ 10y ago
Interestingly, neither of those came to my mind until reading your comment. I would fill in analyst, strategist, or operations specialist. Honestly, though, outside of the self-described "tech industry," understanding software acts as a force multiplier as long as you are able to talk about it. It's similar to having a finance background, except that your special power tends to produce actual gains more easily.
SatvikBeri 10y ago
Cases I've seen done successfully: –Accounting –Risk analysis in banks or fintech companies –Statistics ("Data Science") –Lots and lots of other jobs that have "analyst" in the title, e.g. marketing analyst
thefalcon 10y ago
Graphic design has worked well in my case. It's something I'm passionate about and truly enjoy, and dovetails very nicely with programming.
ufmace 10y ago
Depends on what you think "marketing" means. A lot of engineers think that it means basically building ad campaigns, running adwords, etc. Business people have a bigger definition, more like understanding the market that you're trying to sell to and building things that the customers want and need. That definition probably covers a lot of different possible job titles.
cosmie 10y ago
The reason "marketer" seems so common is because those who are good at marketing are inherently good self promotors. Making them far more visible than other, less common examples. Anecdotally, supply chain and business operations are two big areas where I have success. You don't even really need any domain knowledge to get started in those fields. I did a 6 month internship at one of the world's largest companies that has a consistently top ranked supply chain organization. They had the expense of a 10 person recruiting team (all expensive line managers and directors, not HR personnel) coming to campus for 3 days, and only gave offers to two of us. And I explicitly said during my interview I knew jack all about supply chain. Their response was that supply chain was their expertise and they had training for that if I needed it, but my general analytical skills and Excel knowledge were far more rare (in their applicants). Turns out they were piloting a new inventory management methodology and were using Excel-based "models" to trial it manually at a dozen manufacturing plants before deciding if it was worth the 8- or 9- figure cost to have SAP make that an official model in their system. It involved running batch exports from SAP every morning, putting that into Excel, updating the figures, and sending that out to the plants. But the pilot had scaled to a high volume plant that basically maxed out Excel, and they were struggling to make it work. All I did was essentially refactor a lot of logic to do batch operations (made a few VBA scripts that connected to the Excel spreadsheet with an ODBC connection and ran a series of SQL statements instead of in-cell functions). It's been about 5 years since then, and I still get calls every few months for contract work essentially doing the same thing (Excel got bogged down -> needs a refactor). Part of my day job involves essentially doing the same thing, but in a more generalistic sense on business operations. Process engineering is a portable skillset across verticals. Whether it's in supply chain, sales, marketing, or operations, there's always a bottleneck that's preventing growth or efficient resource utilization. I am a piss-poor programmer. I have never once been hired or paid explicitly to program, and I don't have enough experience in any particular language to pass even a junior level programming interview. I get paid to make irritations or problems go away while causing the least amount of friction utilizing whatever constraints exist. In the course of doing so I've tweaked C# and PHP programs. I've written bash, awk, perl, VBA, Python, and NodeJS scripts. I've written a whole lot of SQL. And I've refactored an incredible number of Google Sheets and Excel workbooks. I'm not valuable because I can program; I'm valuable because I can reliably and consistently resolve whatever issue is thrown my way in a self-sufficient manner, and ramp up on whatever knowledge or tool is needed to do so incredibly quickly whether I have prior exposure or not. I'm not a consultant that provides a report that then requires additional resources to execute; I provide a working solution (or prototype) that immediately solves the issue. And in the case of a prototype, provides a stopgap solution that (crudely) solves things in the immediate term and provides an explicit, working example of the functionality that their engineering resources need to incorporate into their internal tools/processes for a long term fix. One piece of advice that seems to serve me well: people hate change. And people also tend to hate relying on others to perform their job. The greater the change, or the more reliance on a programmer to make modest changes to a process, the more resistance and ill will you'll create. Engineers thrive for perfection, and invariably lean towards "optimal" solutions that are very invasive and supplant existing practices, rather than supplementing them. patio11 didn't create an entire Customer Management System or appointment management system that had automated appointment reminder functionality. He created a simple system that supplemented whatever process and system was already in place and added incredible benefits with minimal additional steps.

henrik_w 10y ago

Scott Adams has similar advice (for anything, not just programming): "Become very good (top 25%) at two or more things." http://dilbertblog.typepad.com/the_dilbert_blog/2007/07/career-advice.html http://dilbertblog.typepad.com/the_dilbert_blog/2007/07/care...

dsacco 10y ago
Yes. I've also seen this referred to as a "T-shaped expert" - someone who has a breadth and depth of useful expertise. Develop broad expertise in a variety of areas and develop deep expertise in a few specialized subfields.
- lumpypua 10y ago
  "Paint Drip People": https://www.facebook.com/notes/kent-beck/paint-drip-people/1226700000696195 https://www.facebook.com/notes/kent-beck/paint-drip-people/1... HN Discussion: https://news.ycombinator.com/item?id=12226242 https://news.ycombinator.com/item?id=12226242

pryelluw 10y ago

This is not to say he is not a good programmer - I simply can't comment on that, though I have reason to believe he is after seeing Starfighter's game. Rather, he leverages that skill set as a means to an end, not an end in itself. Fucking this. I battled for years thinking I was less of a programmer because Im also a marketer. But holy shit does the combination work like a charm. It is like a 1-2 punch that knocks out the competition. Downright unfair. I love it. When companies that work with me realize that I can work in defining, building, shipping, and selling the product, they dont want to let go. Its such a powerful combination of skills that I encourage everyone to acquire them. If you code and learn to market there is a crazy amount of things you can do to make money. Its ridiculous how much.

rjeli 10y ago
Do you have advice for how a programmer can learn to become a marketer?
- pryelluw 10y ago
  Sell something. Anything. Not just software or online. Just like in code, it takes 10,000 hours of practice. If you need help getting started shoot me an email. No bs or strings.
- jtwebman 10y ago
  Build something small and try to sell it. Doing it is really the only way to learn it fully.
  - pryelluw 10y ago
    Building something is a common approach devs take. But there is also the option to just selling something you did not make. Its a shortcut to learning because you can jump right into the action.
    - minhaz23 10y ago
      What are some examples of selling something that you didnt make? Where can I go to begin? Can you also recommend any resources that helped you become a better seller and marketer? Books, lectures, etc? Thanks!

KirinDave 10y ago

It seems like wherever there is social and technical progress in our industry, Thomas Ptacek's shadow was cast nearby. I dunno how that guy has so many hours in the day.