I've been tweaking this side-project for the last week, it's getting a bit more useful I think.

Features:

- It's free - Generate app spec with AI based on your idea - Generate business strategy with AI - Edit the scope list in Markdown, app flow diagrams update automatically - Share your spec and strategy with a link

Check it out and let me know what you think at https://appflows.ai

Comments URL: https://news.ycombinator.com/item?id=41929884

Points: 2

# Comments: 0

I made AppFlows to help our clients here at Pocketworks. Most app ideas start with boring spreadsheets, and I craved a visual way to scope, estimate and explore ideas without getting bogged down in visual design.

It's a bit rough n' ready, but any feedback would be greatly appreciated.

Comments URL: https://news.ycombinator.com/item?id=41846739

Points: 3

# Comments: 1

https://appflows.ai

It doesn't save anything to the cloud. It supports markdown. It currently doesn't using an LLM but I'm sure that will sneak in at some point. One day, I'd like to have it generate wireframes using AI.

I've spoken to a CTO who are using no-code tools, so here are some views:

1. Cost - Will depend on compute power needed, but for internal tools that will be low.

2. You might want to pick something you are familiar with and is actively developed and accepting pull requests. A lot of the low code tools do allow you to write custom code anwyay, which is worth bearing in mind.

3. That's a tough one, some of these tools have a commercial model that runs alongside their open source model, which may help give some security. I remember that Facebook pulled their low-code backend tool many years ago, so just because it's by a big company, doesn't mean it has a long future.

Here's the list:

https://pocketworks.co.uk/blog/open-source-no-code-tools/

To save you a click, here's what's in the list:

NocoDB - Turn any database into a smart spreadsheet Baserow - Create your own online database without technical experience ToolJet - Build & deploy internal tools with minimal engineering effort BudiBase - Build apps for your workplace in minutes AppSmith - Build admin panels, CRUD apps and workflows 10x faster Saltcorn - Point and click database web applications Lowdefy - Build web apps, admin panels and BI dashboards with ease Directus - Wrap your existing SQL database with a GraphQL+REST API and admin panel Frappe Framework - Metadata-driven, full-stack framework in Python and Javascript Basetool - View and manage all your data in one place like a pro Rowy - Manage Firestore data in a spreadsheet-like UI GDevelop - Free, and easy game-making app PocketBase - Backend for your next SaaS and Mobile app in 1 file NocoBase - Build internal tools in minutes

Cheers

T

In 2020, The World Health Organisation started working on a COVID-19 app for Nigeria, built with Flutter. I've been actively interested in Flutter since 2017, but for some reason never saw this in any of the Flutter news.

First commit was March 17th 2020, which is 13 days before the first Nigeria lockdown. As a mild tangent, this was 6 days before the first UK lockdown, 2 days before the first USA lockdown and 4 days after the Albania lockdown (the first lockdown in the world).

https://github.com/WorldHealthOrganization/app

It's also a monorepo, which seem quite popular in the Flutter community where Dart is used on the backend.

Comments URL: https://news.ycombinator.com/item?id=32930708

Points: 12

# Comments: 3

I'm a developer but have studied sales quite a lot over the last 10 years. The most important thing I learned is to find clients that fit you really well when it comes to skills, culture and budget.

To do this you have to give a shit and not be afraid to ask potentially off-putting questions. Stuff like:

"Are you sure you want to build this app, your business case doesn't add up so why bother?"

"Can you demonstrate you're in this for the long haul? There's no joy in building an app that fails, which is what happens if you don't budget to iterate your app."

"How come you don't hire freelancers or contractors, it will cost you less?"

Obviously I ask nicer questions too, but these are good "give a shit enough to risk losing the sale" kind of questions.

For that, the Traction book would get you off to a very good start in implementing your company "Operating System".

I do understand your interest in tools though, but I spent far too long obsessing over tools in my business. Perhaps it's because I'm a software developer and I wanted to believe that tools could fix the problems, rather than facing up to the more human challenges.

Right now our 15-person business uses Notion, Slack, Google Docs, Figma Jam, Semrush and Productive. These cover all the things you mentioned.

I have a friend who's set up an offshore team and he saves 70% on developer salaries from what I can infer.

We employ within Europe and the UK, our average dev salary is around $65,000.

The previous year was 29%, but last year was lower as I made some expensive decisions during the COVID period. Basically, I grabbed all the revenue opportunities I could without too much concern for profit

1. Finding clients is hard. Think hard about what an awesome customer looks like and then figure out how to reach them. This is the most important thing IMO. I'm reading Standout of Die, it has good advice on this.

2. First attempt, I started with about $40,000 and blew it all quickly as I didn't have enough sales. Better to get the sales first then grow the rest. You can be transparent about this with potential clients.

3. If your costs are $400,000 a year, and you'll sell 44 weeks of the year allowing for holidays, sickness and wiggle room, then you need to sell $9,000 a week to stay in business. If you want to make 25% profit then you'll need to sell closer to $500,000 a year and $11.3k a week.

4. I'm in the UK, we pay someone to politely chase clients for payment. This was a game changer. Most agreed to pay monthly but never do without a good bit of nudging.

5. As for tips... There are so many good books and courses on building and running a service business, I wish they'd been there when I started. To name a few.

- Jonathan Stark.com - Hourly Billing is Nuts

- Gareth Healey - Standout or Die

- Jason Swenk - Agency Playbook

- Blair Enns - The Win Without Pitching Manifesto

- Blair Enns - Pricing Creativity

- Traction - Gino Wickman

Running a dev shop can be incredibly rewarding and fun, especially if you love the work you're doing. Building a business will be like learning a whole new career, so don't underestimate the learning curve the number of mistakes you can make (if you're like me haha). It's great that you're asking for advice, it will save you some headaches :)

You can see it on this page. Note that I haven't decided if I'll launch the app yet, so there's not even a domain name. https://qzj2yr24qf8m.umso.co/

GTD also recommends you keep a projects list to remind you of all the open projects you have.

It includes the ones mentioned here - AppSmith, ToolJet, BudiBase and SaltCorn. I'll add NodeRed even though it's a bit more hardware/IoT oriented.

https://pocketworks.co.uk/blog/open-source-no-code-tools

When we hired the 2nd, we looked for someone who had good knowledge of Node, Rails, Laravel or whatever and were excited to learn Elixir/Phoenix. Then they completed a Udemy course and started practicing. It worked out pretty great!

1. A company playbook writing tool that helps you document team and company processes. It gives you a nice online browseable playbook, along with .epub and .mobi download.

2. Adding more advanced features to my https://yuml.me UML tool, including text formatting, UML packages, and a more succinct DSL.

3. A contract e-signing tool that doesn't suck on mobile. For some reason, every digital signature tool I use feels yucky.

4. A tool that lets you write out user stories and converts them into example mobile wireframes by parsing the text. You can also do point estimations for relative sizing.

In our team (non enterprise), there's a lot of visual flow models for UX, and still some class/interaction/activity diagrams for explaining how stuff hangs together.

Think I know what you're saying. That said, there are many people using a command line wrapper to generate yUML diagrams as part of an automated workflow.

The paid features are more for people who want better management of diagrams.

I used to be a back-end guy (I once co-authored a book on an ORM and just love middle tier and db shizzle). I had a huge appreciation for products that looked and felt great, but only had the back-end skills. I had a genuine desire to build skills in design so I could make better products.

If you want to take some short-cuts to build great looking stuff I'd do the following

- Build a mood board of stuff you think looks great. Set your own standards bar.

- Play around with Sketch or similar to learn how to get the look you like (this will make you think about UI design problems). This might take years but you have to start somewhere.

- Read "Design of Everyday Things" and "Don't Make Me Think" and a few other design classics. The principles stand strong.

- Get help from designers to bridge the gap between your skill level and where you want to be. When I started my company, I'd find designers who had a visual style I liked and paid them to help out.

Would look great in http://yuml.me which also has a sketchy look.

I need to rewrite yUML using newer tech, business collaborations welcome :)