5 ms·

Author here. Happy to answer any questions about this post.

by mtlynch 8y ago

galfarragem 8y ago
Interesting and honest read. Freedom is costly. Mistakes are costly. It's the price of learning.
- mtlynch 8y ago
  Thanks for reading!
aberoham 8y ago
What conferences did you have in mind to speak at? Are there any videos out there of you speaking to groups?
- mtlynch 8y ago
  Last year I gave a talk called "How to Do Code Reviews Like a Human" at NERD Summit[0] and PyGotham[1]. Here's a video of the PyGotham talk: https://youtu.be/0t4_MfHgb_A https://youtu.be/0t4_MfHgb_A I'm scheduled to speak at NERD again in March to talk about incrementally applying mature programming practices (unit tests, CI) to a codebase that's missing them. It's based on a series of blog posts I wrote last year called "Resurrecting a Dead Library." [2] I'd also like to speak at Python or DevOps conferences about why I think good developers fall into traps that cause them to write bad tests.[3] 0: http://2018.nerdsummit.org/ http://2018.nerdsummit.org/ 1: https://2018.pygotham.org/ https://2018.pygotham.org/ 2: https://mtlynch.io/resurrecting-1/ https://mtlynch.io/resurrecting-1/ 3: https://mtlynch.io/good-developers-bad-tests/ https://mtlynch.io/good-developers-bad-tests/
jeanlucas 8y ago
Very interesting. I took the same path a while ago. My biggest mistake was not validating business properly. Coding before checking if it was commercially viable. I think you can do this business/validation part better.
- mtlynch 8y ago
  Thanks for reading! Yeah, I think my biggest mistake on Zestful was that I spent too long on the product before validating that anyone would buy it. I think for some products, it's unavoidable to build a little first but I've definitely learned how important it is to get to validation ASAP.
  - montenegrohugo 8y ago
    How would you go about validating a product at such an early stage though? For example, what could have told you that no one had an interest in an API like Zestful? EDIT: There is actually an extremely similar service that does have a business use case: https://dialogflow.com/ https://dialogflow.com/ (from your past employer) Does exactly the same thing, take a short input of words and identify overall intent as well as entities (in your case ingredient, amount, format, etc...) On top of that, it has a great UI, export functions, integrates with other services, has an API, etc...
    - mtlynch 8y ago
      >How would you go about validating a product at such an early stage though? For example, what could have told you that no one had an interest in an API like Zestful? I wrote a blog post last year[0] about what I think I should have done differently to validate it sooner but the tl;dr was that I think I could have pitched it to potential customers before I built anything. I imagine a lot of them could have told me off the bat that they wouldn't be interested. But even if I didn't do that, I spent way too long making a demo site even though the API marketplace I was using had its own (I just didn't like theirs). >There is actually an extremely similar service that does have a business use case: https://dialogflow.com/ https://dialogflow.com/ (from your past employer) I've never tried DialogFlow for recipe ingredients, but I'd be very surprised if it could match Zestful in accuracy. I'm using NLP too, but recipe ingredients are so niche and specialized. But their business is general-purpose NLP. I couldn't achieve that and I definitely wouldn't be able to compete with Google on that battlefield. [0]: https://mtlynch.io/shipping-too-late/ https://mtlynch.io/shipping-too-late/
stuxnet79 8y ago
First off, I just wanna thank you for being so candid and sharing your experiences so far. I wish you'd blog more, but I understand this eats into time you'd prefer to spend on your projects :) Apart from your rent and private health insurance what else about the cost of living in Manhattan (apart from rent, food) pushed your expenses to 7K a month. I'm assuming you moved to a lower cost of area without modifying your lifestyle too much, so it's crazy that you'd get such a dramatic decrease in cost just by moving.
- mtlynch 8y ago
  Thanks for reading! I wish I would blog more, too! I write so slowly that I have way more ideas than I have time to write. >Apart from your rent and private health insurance what else about the cost of living in Manhattan (apart from rent, food) pushed your expenses to 7K a month. I'm assuming you moved to a lower cost of area without modifying your lifestyle too much, so it's crazy that you'd get such a dramatic decrease in cost just by moving. Good question! One thing to keep in mind was that I don't have especially expensive tastes, but I also didn't adjust my lifestyle at all when I quit until I moved, so I could have lowered it from $6-7k if I put more effort in. One big expense in New York is just the cost of socializing. In other areas, I feel like people socialize more at each other's houses. In New York, because housing is so expensive, it's hard to have a group of friends over, so people tend to meet at restaurants, which are expensive. And nobody has a car, but everyone lives spread out, so you spend a lot on Uber. But if you're living in Manhattan, everyone around you is living this relatively high income / high spend lifestyle, so you don't really notice spending so much. And I'm not even talking like bottle service, five-star restaurant dining. Just like get brunch together, go do some activity that costs $30-50/person, have dinner, that could be like $100-200 in a single day. Part of what keeps my costs down in Western Mass is that a) costs are just way lower and b) people around me don't have such a high income so normal activities that people choose are relatively cheap.
ghostbrainalpha 8y ago
How many items are listed on Is it Keto? I'm not seeing results for my first two searches Pickles, and Mayonnaise. Also its not imminently clear after doing a search that the search bar will let you do another "Is it Keto" search because of the change in style. Did you do a "Is X Keto" keyword search in Adwords to come up with your initial list of foods?
- mtlynch 8y ago
  Thanks for reading, and for the good questions! >How many items are listed on Is it Keto? It's currently at 131 foods. At the beginning, I deliberately made browsing hard to obscure the fact that the database was so small. Now that it's a bit bigger, I'm considering making it possible to browse by category. >I'm not seeing results for my first two searches Pickles, and Mayonnaise. Yeah, those are in the queue. My writer actually wrote a draft of Mayonnaise but I still need to edit it. Mayonnaise is the #21 most 404'ed page and pickles is #45. But it's a long tail. "Beans" is the #1 most 404'ed and it's only 1.7% of 404s. >Also its not imminently clear after doing a search that the search bar will let you do another "Is it Keto" search because of the change in style. Good point. I'll tweak the UI to make it more obvious that the user can keep searching. >Did you do a "Is X Keto" keyword search in Adwords to come up with your initial list of foods? I do it a few different ways. At the beginning, I would just look in Facebook groups and wait until someone asked "is X keto?" then I'd quickly write the article and link it. Then I started looking at which pages got the most 404s and I'd write those. Now that the Twitter account is more popular, I have more control over what people read (as opposed to only reacting to what users search for) because I can drive people to a page by sharing it on Twitter. And if a product is keto and I @-mention the manufacturer, sometimes they'll RT it and give it more exposure, so I look for manufactured foods that are keto friendly. Because for something like "lettuce," it's hard to make money because they have to click the link, then click something else that's keto-friendly and that they can buy on Amazon. For SEO, I also do look at Google Trends and look for "X keto" keywords where there isn't already a good result. A query like "is xylitol keto" is hard to win at in SEO because a lot of big keto blogs already have dedicated articles to it. But I think Is It Keto gets top spot for "Is Propel Keto?" because none of the bigger sites ever bothered to write an article about such a specific product.
shaman1 8y ago
Congrats for your courage! On a slightly personal note, how was your social life been for the past year? If you want to have a family do you see yourself staying in this city?

Author here. Happy to answer any questions about this post.

galfarragem 8y ago

Interesting and honest read. Freedom is costly. Mistakes are costly. It's the price of learning.

mtlynch 8y ago
Thanks for reading!

aberoham 8y ago

What conferences did you have in mind to speak at? Are there any videos out there of you speaking to groups?

mtlynch 8y ago
Last year I gave a talk called "How to Do Code Reviews Like a Human" at NERD Summit[0] and PyGotham[1]. Here's a video of the PyGotham talk: https://youtu.be/0t4_MfHgb_A https://youtu.be/0t4_MfHgb_A I'm scheduled to speak at NERD again in March to talk about incrementally applying mature programming practices (unit tests, CI) to a codebase that's missing them. It's based on a series of blog posts I wrote last year called "Resurrecting a Dead Library." [2] I'd also like to speak at Python or DevOps conferences about why I think good developers fall into traps that cause them to write bad tests.[3] 0: http://2018.nerdsummit.org/ http://2018.nerdsummit.org/ 1: https://2018.pygotham.org/ https://2018.pygotham.org/ 2: https://mtlynch.io/resurrecting-1/ https://mtlynch.io/resurrecting-1/ 3: https://mtlynch.io/good-developers-bad-tests/ https://mtlynch.io/good-developers-bad-tests/

jeanlucas 8y ago

Very interesting. I took the same path a while ago. My biggest mistake was not validating business properly. Coding before checking if it was commercially viable. I think you can do this business/validation part better.

mtlynch 8y ago
Thanks for reading! Yeah, I think my biggest mistake on Zestful was that I spent too long on the product before validating that anyone would buy it. I think for some products, it's unavoidable to build a little first but I've definitely learned how important it is to get to validation ASAP.
- montenegrohugo 8y ago
  How would you go about validating a product at such an early stage though? For example, what could have told you that no one had an interest in an API like Zestful? EDIT: There is actually an extremely similar service that does have a business use case: https://dialogflow.com/ https://dialogflow.com/ (from your past employer) Does exactly the same thing, take a short input of words and identify overall intent as well as entities (in your case ingredient, amount, format, etc...) On top of that, it has a great UI, export functions, integrates with other services, has an API, etc...
  - mtlynch 8y ago
    >How would you go about validating a product at such an early stage though? For example, what could have told you that no one had an interest in an API like Zestful? I wrote a blog post last year[0] about what I think I should have done differently to validate it sooner but the tl;dr was that I think I could have pitched it to potential customers before I built anything. I imagine a lot of them could have told me off the bat that they wouldn't be interested. But even if I didn't do that, I spent way too long making a demo site even though the API marketplace I was using had its own (I just didn't like theirs). >There is actually an extremely similar service that does have a business use case: https://dialogflow.com/ https://dialogflow.com/ (from your past employer) I've never tried DialogFlow for recipe ingredients, but I'd be very surprised if it could match Zestful in accuracy. I'm using NLP too, but recipe ingredients are so niche and specialized. But their business is general-purpose NLP. I couldn't achieve that and I definitely wouldn't be able to compete with Google on that battlefield. [0]: https://mtlynch.io/shipping-too-late/ https://mtlynch.io/shipping-too-late/

stuxnet79 8y ago

First off, I just wanna thank you for being so candid and sharing your experiences so far. I wish you'd blog more, but I understand this eats into time you'd prefer to spend on your projects :) Apart from your rent and private health insurance what else about the cost of living in Manhattan (apart from rent, food) pushed your expenses to 7K a month. I'm assuming you moved to a lower cost of area without modifying your lifestyle too much, so it's crazy that you'd get such a dramatic decrease in cost just by moving.

mtlynch 8y ago
Thanks for reading! I wish I would blog more, too! I write so slowly that I have way more ideas than I have time to write. >Apart from your rent and private health insurance what else about the cost of living in Manhattan (apart from rent, food) pushed your expenses to 7K a month. I'm assuming you moved to a lower cost of area without modifying your lifestyle too much, so it's crazy that you'd get such a dramatic decrease in cost just by moving. Good question! One thing to keep in mind was that I don't have especially expensive tastes, but I also didn't adjust my lifestyle at all when I quit until I moved, so I could have lowered it from $6-7k if I put more effort in. One big expense in New York is just the cost of socializing. In other areas, I feel like people socialize more at each other's houses. In New York, because housing is so expensive, it's hard to have a group of friends over, so people tend to meet at restaurants, which are expensive. And nobody has a car, but everyone lives spread out, so you spend a lot on Uber. But if you're living in Manhattan, everyone around you is living this relatively high income / high spend lifestyle, so you don't really notice spending so much. And I'm not even talking like bottle service, five-star restaurant dining. Just like get brunch together, go do some activity that costs $30-50/person, have dinner, that could be like $100-200 in a single day. Part of what keeps my costs down in Western Mass is that a) costs are just way lower and b) people around me don't have such a high income so normal activities that people choose are relatively cheap.

ghostbrainalpha 8y ago

How many items are listed on Is it Keto? I'm not seeing results for my first two searches Pickles, and Mayonnaise. Also its not imminently clear after doing a search that the search bar will let you do another "Is it Keto" search because of the change in style. Did you do a "Is X Keto" keyword search in Adwords to come up with your initial list of foods?

mtlynch 8y ago
Thanks for reading, and for the good questions! >How many items are listed on Is it Keto? It's currently at 131 foods. At the beginning, I deliberately made browsing hard to obscure the fact that the database was so small. Now that it's a bit bigger, I'm considering making it possible to browse by category. >I'm not seeing results for my first two searches Pickles, and Mayonnaise. Yeah, those are in the queue. My writer actually wrote a draft of Mayonnaise but I still need to edit it. Mayonnaise is the #21 most 404'ed page and pickles is #45. But it's a long tail. "Beans" is the #1 most 404'ed and it's only 1.7% of 404s. >Also its not imminently clear after doing a search that the search bar will let you do another "Is it Keto" search because of the change in style. Good point. I'll tweak the UI to make it more obvious that the user can keep searching. >Did you do a "Is X Keto" keyword search in Adwords to come up with your initial list of foods? I do it a few different ways. At the beginning, I would just look in Facebook groups and wait until someone asked "is X keto?" then I'd quickly write the article and link it. Then I started looking at which pages got the most 404s and I'd write those. Now that the Twitter account is more popular, I have more control over what people read (as opposed to only reacting to what users search for) because I can drive people to a page by sharing it on Twitter. And if a product is keto and I @-mention the manufacturer, sometimes they'll RT it and give it more exposure, so I look for manufactured foods that are keto friendly. Because for something like "lettuce," it's hard to make money because they have to click the link, then click something else that's keto-friendly and that they can buy on Amazon. For SEO, I also do look at Google Trends and look for "X keto" keywords where there isn't already a good result. A query like "is xylitol keto" is hard to win at in SEO because a lot of big keto blogs already have dedicated articles to it. But I think Is It Keto gets top spot for "Is Propel Keto?" because none of the bigger sites ever bothered to write an article about such a specific product.

shaman1 8y ago

Congrats for your courage! On a slightly personal note, how was your social life been for the past year? If you want to have a family do you see yourself staying in this city?