MaknMoreGtnLess

Happy to expand on this, but unless, the founders fit the exact beachead market the MVP is testing to help solve problems of, finding alternative ways to discover the beachead market than through an MVP, is a way more effective way to learn.

Definition of MVP: an actual product being built to "validate the product" with the beachead customers needs.

It's important to call out and define what an MVP means because other people might mean "cheapest market discovery tool" instead of "cheapest product validation tool" - very different things

> I then have to go out of way my to prove to this person the code is doing what is expected/asked, either in long winded responses on the PR, or sometimes as far as pulling them into a video call to screen share and live debug the code to them

Process issue. Fix:

1. Write tests

2. Ensure tests can run in a well defined/containerized environment so that they can run tests and ensure they pass

The JVM is a blackbox that fails in terrifying ways at the worst time possible.

You won't know this until you take your service to production and expose it to real life traffic.

Now days, unless there's an absolute non-alternative to a Java project, I stay away from one.

For example, for most cases, Apache Kafka can be replaced by RedPanda (C++). Other potential solutions are LiftBridge (or core NATS)

> I'm interested in frugal computing, and I want my software to run smoothly on a Raspberry Pi if at all possible. So I do have some demand for a native program or, rather, set of programs.

I understand. My point is that it's likely you're a fraction of the overall "home PC" market who are fine purchasing a regular PC/Laptop/Tablet with 16GB+ RAM and multiple cores (because it's so cheap now) if that can support apps that simplify their overhead.

SO while a Raspberry Pi with a NAS that backsup to rsync.net might consume way less power and be more cost efficient, it might require technical overhead that majority of people are unwilling/unable to support/incur.

> Besides, on a principle level, I fail to see how something as simple as e-mail should need (or be allowed to use) something as complicated and resource-hungry as a web browser

From a developer of a product POV, it's much simpler and less of an overhead building a web app that runs in a web browser than a native app that runs on 5 different platforms and architectures.

Essentially the developer of products are outsourcing the headaches of 5 different platform and architecture compatibility to the web browser.

It seems to me that majority of users don't want to deal with a native application - specially one that:

1. They would have to install, maintain, backup and configure themselves 2. Has a reasonable web application, that, as a major perk, they don't have to install, maintain, backup and configure themselves

Due to the lack of demand for such a product, I am not surprised the existing products arn't polished/well maintained.

> I personally see a minimum of three jobs, done well, at 30 hours a week.

Concur!

It isn't that enough people arn't wildly smart enough/don't have the mental capacity to be a full-stack developer AND a penetration tester simultaneously.

It's that it just don't make sense to enough of these wildly smart people to take on what's essentially three jobs, that when done well, individually consume 30+ hours a week each.

Businesses that hire these frankenstien's monster as employees instead of contractors just don't have a clue.

It's also extremely likely these frankenstien's monsters will burn themselves out as employees essentially juggling three jobs as a salaried flat rate employee.

The only scenario in which being this frankenstien's monster make sense is to bootstrap a consultancy as a founder or a contractor, sign up enough clients and scale out by hiring specialist employees.

The only reason why these frankenstien's monsters happen (outside being a founder or a contractor and I am talking exclusively about employees here) is that employee started off as one type (full-stack developer OR a penetration tester), failed to find employment and cross trained into the other, where they found employment.

For example, without knowing any details, it's likely the full-stack developer AND a penetration tester started off as a penetration tester, failed to find employment and cross trained into full-stack developer. Of course, devil lies in the details, so if the full-stack developer failed to find employment as one but noticed significant demand for penetration testing, they could very well have become a penetration tester.

Empirically a penetration tester doesn't have either time or energy to be an effective full-stack developer or vice versa, especially outside a founder or a contractor role in a bootstrap stage.

> I didn't know we were this rare, but it's nice to hear.

I don't think you're drawing the right conclusion.

It isn't that enough people arn't wildy smart enough/don't have the mental capacity to be a full-stack developer AND a penetration tester simultaneously.

It's that it just don't make sense to enough of these wildy smart people to take on what's essentially three jobs, that when done well, individually consume 30+ hours a week each.

Businesses that hire these frankenstien's monster as employees instead of contractors just don't have a clue.

It's also extremely likely these frankenstien's monsters will burn themselves out as employees essentially juggling three jobs as a salaried flat rate employee.

The only scenario in which being this frankenstien's monster make sense is to bootstrap a consultancy as a founder or a contractor, sign up enough clients and scale out by hiring specialist employees.

The only reason why these frankenstien's monsters happen (outside being a founder or a contractor and I am talking exclusively about employees here) is that employee started off as one type (full-stack developer OR a penetration tester), failed to find employment and cross trained into the other, where they found employment.

For example, without knowing any details, it's likely the full-stack developer AND a penetration tester started off as a penetration tester, failed to find employment and cross trained into full-stack developer. Of course, devil lies in the details, so if the full-stack developer failed to find employment as one but noticed significant demand for penetration testing, they could very well have become a penetration tester.

Empirically a penetration tester doesn't have either time or energy to be an effective full-stack developer or vice versa, especially outside a founder or a contractor role in a bootstrap stage.

> HR is complaining about not being able to hire people

Perhaps HR was complaining about not being able to hire those people for the $50k/yr + unlimited PTO budget that they have?

because I can bet that if HR was willing to pay $500k+/yr + sensible PTO, a few of those GCP engineers right out of Google who literally wrote the first few lines of code for GAE might consider joining.

> was the best paid developer in the company with a very competitive rate

OK, so did they die or move on to a better paying opportunity?

I was trying to use Dendron but it struggles with text files larger than a few hundred KBs and my text files can often go into 10s of MBs. Plus I like to use ASCIIDoc instead of MD.

I won't ever use proprietary tools, specially those I have to pay recurring fees for because I would need to use these notes for nearly 50 years.

Unless it's a well defined technical solution or a very deep and narrow (well defined) topic, I've always found "the answer" to be spread across multiple webpages/sites/sources. To tackle this:

1. I have a coarse hierarchy of topic names as a directory tree

2. These directories have mostly text files in them but can also have PDF etc

3. Text files contain my summary/"the answer" followed by the relevant extract from webpage (citation). There usually are multiple such citations

4. Since I mainly deal with a lot of textual context, I have mostly text files but there are situations where I need spreadsheets/pictures/video - in those cases, I save the webpages as PDF but maintain a parallel version of #3 with notes to refer to the native format when spreadsheets/pictures/video need to be involved

> we had a team mate in the past who did exactly this

Did they die or move on to a better paying opportunity?

> something we've needed since the 1970s, but everyone dismissed as too complex, slow, or unnecessary

What are some scenarios where Capability Based Security might be thought of as unnecessary (but isn't/wasn't)?

The first question to ask yourself is whether you want to be an:

1. Entrepreneur, or

2. Investor

> For engineering, for example, big company employees drastically underestimate how much they rely on homegrown tooling.

Are you in engineering yourself?

I can assure you that people in sales, marketing etc also drastically underestimate how much they rely on homegrown or integrated tooling and support systems.

Sometimes, they don't even know what those tooling and support systems were.

They only learn when they either join a company that don't have those tooling and support systems, or start their own

> if you know how to screen for good founders

What are some questions you ask for screen for good founders?

> The point is to find a startup that has proven potential, not a total dud

If you can do this, you can quit your job and start a extremely well paying job consulting for VCs

big tech will always be a safer bet all risks considered

The question you need to answer is, for YOU right now:

1. Are the things you want to learn best served by a big tech co?

2. Or an unproven startup?

> JSON schema and protobuf didn't feel elegant and were too verbose so I thought why not GraphQL

2 questions:

1. Are there tools out there that can translate between GraphQL schema, JSON schema and protobuf?

2. How would you version events

"advice" that's root to why majority of startups fail:

1. Build a MVP trusting your "gut instincts"

2. Buy traffic slapping your MVP on anyone who types in keywords that you think people type when looking for your product

3. Rabidly iterate on random feedback from "customers" who need just "1 more feature" before they can become paying customers.

> We are glad we did so otherwise we'd have ended up spending our time & money on developing something which would've turned out to be totally useless!

The exact specifics of this would be far more valuable than a more general blog post about the importance of talking to customers before building a product - there are far too many variables involved there.

3 questions:

1. Would you like to blog about the exact specifics of this?

2. When would you be able to do that? I am interested in reading it ASAP even if it's a draft

3. Are there other risks to a startup besides:

  - Market
  - Product  
  - Team

> Anyone has experience with working with such a system?

Yes, I have been using this system for over a decade and it's extremely capital efficient.

> but the manager in me feels that this is a naive approach

What risks do you see that make you concerned?

> If you were making a tiny social network site, what stack would you use?

A stack that's already proven to work, that someone else runs for me at as much little cost to me as possible while I collect the metrics I need to decide if there's a business.

In here I propose using private Facebook groups.

If I were making a POC social network site, I would already have had run some user studies for the initial seed group of users to understand why the existing social network sites arn't working for them or why they havn't already used one of the existing social network platforms to create a group.

Let's assume, I find that the initial seed group of users are a group of single mothers who don't want to remarry and are perfectly fine being friends with other women - their main problem is they don't want men to bother them.

I would then create a private Facebook group and invite these single mothers and see if this private social network obtains the engagement I thought it could reach.

> How would you make a decision in this case?

I would interview potential customers and ask if the problem your idea solves, is a problem they resonate with. Then you ask them how badly they would like to solve the problem. Once you find out that actual human beings are interested in you solving a problem they badly need solved, it will strongly motivate you to levels you might have never reached before.

The biggest challenge though is talking to actual human beings because some of them might not be the right customers for you and if the probability distribution is such that you happen to talk to a lot of these at the beginning, you might be badly demotivated.