No, it just feels like the right thing to do, and I'm hoping that others could benefit from the work. While I would like to release some libraries under permissive licenses, I really have no plans to shepherd collaborative open source projects (which I think might turn into a huge time sink). Others would obviously be welcome to fork and maintain these projects, but I would prefer for them to simply be read-only versions of what we're using in production.
A would-be competitor, in a country not respecting intellectual property laws, might get a few customers by launching our product and charging only a fraction of what we charge. However, what will this competitor do when their customers request new features or report bugs? Obviously, we'd be a lot more nimble in responding to such requests, so maybe this shouldn't be such a big concern after all.
Also, if this does becomes a problem, we can just stop releasing new versions publicly, at which point the competitor's offering would stagnate, while our offering would continue to improve.
No. Worst case scenario is that the competitor has more resources than you and can be even more nimble at adding new features and fixing bugs. Because you made all the source code available, you lost all the 'head start' you had against such a competitor.
The service is named Calcapp, an app builder for spreadsheet-savvy users (https://www.calcapp.net). The codebase weights in at half a million lines of code and documentation, with lots of generic utility libraries I have written over the past ~20 years, in Java and JavaScript.
I don't think that releasing the user-facing app creator and runtime libraries under a permissive license would be smart. However, I would still like that code to be out there, and for the utility libraries to be available for anyone to use under a permissive license.
The one downside I can see myself is that by making the source available, we might reveal security vulnerabilities. Sure, security through obscurity is not a good idea in and of itself, but revealing that we haven't updated a key, vulnerable library might still be problematic.
I am on my phone so couldn't really test it out, but it sounds sufficiently niche that it could both benefit or be harmed by releasing the code.
Usually, the hardest bit is getting the infrastructure set up, so if you do have dozens of intertwined components, just pushing the code might not really help anyone.
Still, I think it depends on the business you have and market you are covering, more than the app itself. Eg. if your market is huge, somebody will want to jump in. If it's tiny, nobody else is probably going to bother even if you made it trivial to deploy and run. And there is a huge continuum of options in between :)
One thing I've seen was default to AGPL after, say, 3 years after release. Depending on the development pace, you might make that 1 or 5 years. In general, this means you could AGPL the version from 5 years ago today.
> You're probably referring to the wonderful nXML mode
But wasn't there an issue where early nxml mode for Emacs would not support the Relax NG using the XML syntax? I vaguely remember the compact Relax NG syntax being supported but not the XML one. And for whatever reason that XML syntax for Relax NG was supported under XEmacs.
It was really a long time ago so my memory is very fussy but something like that. Since then I switched back to Emacs and I do use nxml-mode (I don't use Relax NG anymore though).
Does anyone know if this works well (to the extent that it does), because the documented Blender Python API was part of the GPT-4 training set, or because the Blender API is fairly predictable?
I'd love to add this capability to our SaaS product, but I've waited for OpenAI to make GPT-3.5 or GPT-4 available for fine-tuning. (Cramming an entire API into the prompt does not seem feasible, not even with support for 32K tokens.)
He doesn’t need to supply the API, GPT already knows the API from its training data. I’ve written a similar tool for Unity and it works the same way. Any program that allows scripting/plugins (and has a sizeable community posting code online) will be able to work the same way.
Back in 2010, I used plasTeX (http://plastex.github.io/plastex/) to convert my thesis to HTML (http://www.polberger.se/components/). plasTeX is "a Python package to convert LaTeX markup to DOM." If memory serves, plasTeX worked rather well, and still seems to be maintained today.
My father asked me to create Windows apps for the hospital department where he worked as a doctor (calculating drug dosages and the nutritional needs of preterm infants). I figured that there might be others with a similar need and eventually created a SaaS app builder that is currently my full-time job. It's been almost 20 years since that original request. https://www.calcapp.net/blog/2018/04/09/launching-after-15-y...
Thanks. I did lots of other things in parallel during the early years. I left my last consulting gig in 2014, so I have been doing in full-time since then.
Yeah, it takes a lot of perseverance to get a startup going. (Or rather, a lifestyle business in my case -- it's not a rocket ship, but I get to work on interesting things and interact with friendly customers, so it really is a great job.)
