drpancake

I can relate to a lot of this. The customer emails in particular. I was experiencing flashbacks while reading the article.

Back in 2014-2015 I created a platform for building crypto trading bots in the browser with Python scripts. You could write a bot, backtest it and trade live on multiple exchanges. It was a lot of fun to build, initially. I kind of stumbled onto the idea without any trading background.

It got some early traction. So I raised $100k of funding and went all in on the idea. I worked like mad.

But I slowly realised that all of my customers were "technical analysis" bros. Basically horoscopes for trading. I received endless support requests for exotic new indicators and obscure features.

I added a marketplace that let you sell your bots to others. It quickly jumped up to $10k/mo in commission after the fee I took. That seemed bananas to me as a 24 year old.

But the bots being sold were pure snake oil. I distinctly remember that the guy who sold the most didn't even run his bot with his own money.

I felt like was running a gambling site. After 14 months of this I was completely burned out. I quit and I wasn't able to work at all for four months.

Glad to see you figured out a smarter solution.

They're in the business of encouraging users to post content to get more eyeballs and therefore more ad revenue. So commoditising a way of generating a lot more content is in their interests.

I commented on this above but I think this is because ASV users need to be screened for heart issues first!

While I agree in general, people who are candidates for ASV have to be screened first by a cardiologist. Certain traits can lead to a much higher risk of death while using one.

Hi there. The app surfaces dates in various ways.

In the search results there's an "Active" column that indicates whether a newsletter has published an issue within the last 45 days.

Then when you click into a newsletter it will show you the "Founded" date (when it started) and the dates for the most recent issues.

Hi there. It's all publicly available data.

Right now, no, but you can claim* your newsletter and soon I'll add the ability to edit your listing.

The current pricing is the cheapest it will ever be! I'm aiming this at businesses and agencies. I have a similar and quite successful app for podcasts that starts at $99/mo.

* https://reletter.com/claim-your-newsletter

Sure, it's Elasticsearch.

It's 54GB on disk.

Hey there. Yes the search isn't perfect yet for certain terms but I'm working on improving it.

Did you use the topic search for Javascript? I get quite a lot of relevant results. Note that we only show the top 6 for free users.

When you know you know. It's like the difference between pushing a boulder uphill and downhill. Very hard to predict ahead of time - you just have to keep shipping stuff.

Sorry to hear that. How long were you trying for? A lot of these guys were at it for years and years before they found success. It can be brutal in the beginning.

I think it's unfair to characterise the "digital nomad influencers" on Twitter as making it seem easy. I know a few of them personally and they work incredibly hard. You can tell by the speed at which they ship new products. They work a lot.

That said, building a Twitter audience of indie hacker types is massively overrated. It's a huge amount of work and the space is utterly saturated. These people generally don't buy stuff and if you try a lot of different ideas then the chances are that many of them are not in your target market.

Feel free to email me. Details in my bio.

I built very simple tool in an industry that I wanted to be part of and was surprised to find that a few people wanted to pay for it. I mean very basic - done in three days.

A few months improving that and soaking up everything I could find (newsletters/press/blogs) led me to noticing a larger opportunity in the same industry so I built that and now it supports me full-time.

So I would advise picking an industry or niche that you're interested in (preferably one that's growing and not particularly tech-savvy) and then spend some time reading up on what people in that industry talk, write and complain about. There are opportunities everywhere.

A few people have tried that e.g. https://web.stanford.edu/class/archive/cs/cs224n/cs224n.1184...

Thanks! I wasn't aware of django-db-queue. As far as I can tell it runs in a single process so I guess running tasks will block any others from running. That's probably acceptable for some use cases.

Interesting that django-q has an ORM broker mode. Looks like it forks workers using multiprocessing under-the-hood.

Exactly. For many projects it's overkill to set up an external broker like Redis or RabbitMQ. That's where I think Chard fits in.

Do let me know how it goes :)

Sure but for 99.9% of Django projects you safely assume there is a database available. I'm not entirely sure why that's a no-go.