Lol. I actually saw it a few weeks ago. It was unfortunate for both of us to have the same name. Mine is being renamed Rudy by the way. Not because of that name similarities. Honestly there so many graveyard attempts at this that I never thought Redux-First-RoutING would see the light of day.
Unfortunately he’s doing the Redux Little Router way, and you already know what I think about that. All action types are basically the equivalent of PUSH. That’s not to mention Rudy has an exhaustive feature list. I hate having to say this sorta stuff cuz I know he and we all have put a lot of work into this stuff.
Besides the long feature list OF THINGS YOU NEED, Rudy has a lot of nuances in the implementation that have been thought through and rewritten over 8 months. The project is literally 2 months of 70 hour weeks worth of work. I’ve tried many different things and synthesized it the perfect combination. For example RLR and ING likely dispatch actions before URL changes and then when a URL changes dispatch an action. That little inconsistent has possibility to cause u problems. Mine always results in the URL changed first, and then the state changed. This solves problems specifically related to Scroll Restoration.
Next week’s code splitting and chunk/thunk prefecthing capabilities in conjunction with my “Universal” family of packages are going to cement Rudy as the one and only :). Everything is done nicely in a decoupled way, which is the draw of using this stuff over Next.js. Frameworks are bad. Plain and simple. Maybe not always, but when React and Webpack get the level of abstraction low yet powerful enough it removes the need for frameworks on top. Or another way to put it: With tools like Universal + Rudy, a framework better bring a lot more than Next.js. Rudy + Universal offers their proclaimed core feature with far more flexibility.
Bottom line: if another routing package decides to go the types route, then perhaps we have a competition. Hopefully we just collaborate. But basically then it’s just a matter of who has more features and a simpler API. Rudy has the simplest most obvious API. Building a competitor is something that would happen in the dark or would be a waste of developers’ time if they have already seen this. This is the right way to do all this stuff. After next week’s release and after I put out all the tutorials the world needs, I’m moving on. It’s a solved problem. Finally. Not that I won’t support it. I will stay very active for a long time. But there’s not much more to achieve feature complete, similar to how React Router 4 is basically feature complete. A few things were found in the issues to the repo and I’m on it: query strings we do need after all. Well that’s it. I’m also going to add prompt confirmation like a React Router — a feature I’ve always known about but never personally needed. All this is easy. The real challenge is getting the idiom for code splitting and prefetching right. I have a few insights there. The clue is that ur Link components and ur universal components must be coupled by URL (or type in the case of Rudy) to enable the correct idiom for prefetching. Like Next.js has. They discovered that.