Roast my startup - Open Lowcode - The next step when a spreadsheet is not enough

[exoregonsoldier]

Open Lowcode is a solution for rapid development of specific enterprise software. With Open Lowcode, you can develop quickly specific software by assembling pre-existing software bricks to your exact requirements.

In a few hours or at most a few days of work, you get a real multi-user application that really fits your needs with all the security, data segregation, and scalability that you currently get from multi-million dollar enterprise solutions or more rigid cloud offerings.

Open Lowcode today is a perfect fit to digitalize budget planning and program management processes involving hundreds of people. With Open Lowcode, you build the app you really need for your special process, suppressing any need for costly pre or post-processing. Apps using the technology version 1 are alreadying running at Airbus.

Current challenges are to build version 2, which will be much more friendly to non-coders, and sell the product to other companies. I am in the middle of the painful process of learning enterprise sales, leveraging my network. One of the main challenges is to sell the "lean" / "low-cost" aspect of the tool, which is much harder than I thought. I plan to scale the project (currently mostly a one-man-show) when I find enough customers to make it sustainable.

Low-code (rapid application development) tools are in competition with spreadsheets at the low-end, Saas Software (very convenient, but rigid and captive) and classical enterprise software (quite complex and not always flexible). The low-code scene is quite crowded now, it seems we all had the same idea at the same time, with both specialized players (Appian...) and big actors (Amazon, Google).

In this crowded scene, Open Lowcode positioning is:

• full open-source: so that companies will never be a captive customer again
• stable technology stack: Open Lowcode only uses proven technology (java, sql...) that will still be there in two decades to avoid the utmost waste of technical migrations.
• developer-friendly: I strongly believe that coding is the best way to solve some type of business needs, so the product has all the 'extension points' required for coders to develop the 'last mile' of the solution
• ergonomics for real work: Focus is 100% on 'real-work' (done on a 'real' computer with keyboard, mouse, big screen...).

Links: Project site, github repo

 

[magda0316]

@exoregonsoldier

full open-source: so that companies will never be a captive customer again

How are you going to earn money, if you give away the code for free? Don't get me wrong. You're addressing a real problem (vendor lock-in) here. But if it's open source, anyone can take the code and compete against you without participating in the development costs. There are business models that work with open source (e.g. Red Hat Linux), but you should give some thought about how your startup will reliably generate income from the software. Wouldn't be the first time that the innovative company ultimately lost the game while everyone else benefited from them. Just look how the story about Docker went.

Apps using the technology version 1 are alreadying running at Airbus.

Already having a big customer is good. A few concrete show cases would be nice still. What exactly did they build with it. Tell us more about what can be done with the platform (and where the limitations are). Except for being opensource, how is it unique from the competition? Or do you solely distinguish yourself with being opensource and on a "proven tech stack"? These are good features, but how do you monetize them (see first question)?

ergonomics for real work: Focus is 100% on 'real-work' (done on a 'real' computer with keyboard, mouse, big screen...)

This isn't necessarily a feature. It will be understood that the apps don't really work on mobile devices (e.g. tablets) but those can be relevant in many "enterprise" scenarios.

 

[exoregonsoldier]

@magda0316 Hi,

I really appreciate your comments. Here are my current thoughts on the points you raise:

• Monetizing open-source : I am not worried about this one. First, I enjoy running operations super-lean, so I am expecting my needs to be limited. Also, when the platform gets adoption, I expect many opportunities to monetize, considering the habits of IT departments in companies
  • selling support: This is more or less the redhat model: companies happily pay to have someone to call about the software they buy. There are also trainings, developer certifications...
  • selling extensions to the framework: I expect this one to be big. A company building a significant application with the framework will very likely encounter points where they are blocked because a feature is missing in the framework. I plan to sell at high price development in a short timeframe of the extension a customer needs.
• Unique positioning:
  • I have a question for you: do you think a positioning of delivering something good enough and open source could work (Kind of Ikea for enterprise software) ?
  • Pilots at Airbus are around budget forecasting and planning.
  • I think a unique feature I have is bringing spreadsheet-like power features (mass figures edition, pivot table) to the next level (complex data model, security, advanced versioning features).
• 100% focused on desktop:
  • I completely believe in this one. For people managing complex topics (say more than 1000 lines of data in excel), desktop ergonomics is key. It is not beautiful, it is not trendy, but man, it is so much more productive than trendy stuff with mobile-like ergonomics. I have seen significant regression in productivity when applications with mobile in mind are deployed to users.
  • I do not rule out having access on tablet / mobile phone, but as my resources are limited, and I do not need 100% of customers to use my application to go to next stage, it is clearly not the priority. So today, I focus on topics that are done at a desk (and there are a lot: your accounting department will not work on the factory floor).

 

[bkuc9810]

@exoregonsoldier

Javafx

need I say more

 

[exoregonsoldier]

@bkuc9810 Do you mean javafx is substandard ? To be honest, this is the part of the architecture I am the most in two minds about, but all alternatives have issues. Here is the rationale:

• I need the software to work well with limited networks, be fast (target time for loading a simple page is 100ms) and also the client to have very good ergonomics with datasets of 1000 to 10.000 lines of data. I understand it is impossible, or very complex to get such properties using a web interface (I am not a super specialist on the latest stuff though). So I developed my own thin client with all primitive I need, hoping to get the best of two worlds.
• All the software is in java (I am clear on the choice of java, it is the type of language I want - hard-typed -, good enough for performance, and second to none for libraries. So having the front-end in the same language is somehow a plus (sharing libraries between client and server)
• I need the software to be stable. I do not care if what I do is not the latest marginally cutting edge stuff as long as it still works 5 or 10 years from now until whenever we have a significant ergonomics paradigm change from the windows / mouse interface invented in the 1980s
• javafx is not so easy to work with (at least with my skill level), so, yes, I struggled with tuning some components.
• The most obvious alternative for what I want to do for the client would be Microsoft stack, but, somehow, for an open-source product, I fear putting Microsoft inside, and it would not work on secondary platforms for enterprise stuff (mac, linux, mobile...), whereas javafx can.

 

[bkuc9810]

@exoregonsoldier I might be ignorant on the target demographic for this software but as someone who regularly writes a lot of frontend Javafx is a bit of a dealbreaker, I think front end is at a stage where your choices are Angular/React/Vue for most modern apps and having support for that could help sell to the newer crowds