5 ms·
I don't really get who stuff like this is aimed at. Either you know no code so this works for the first 40% of the stuff you need and then you just need to prog
by royletron 9y ago
I don't really get who stuff like this is aimed at. Either you know no code so this works for the first 40% of the stuff you need and then you just need to program the really unique and incredibly hard stuff. Or you already know code and you end up wrestling whatever comes out of this into a form that you can actually support going forward.
Tools like (i.e. WYSIWYG editing suites) this have a yield point of usefulness and I am not sure open source can get there. There are very few exceptions (Gimp and Audacity spring to mind).
- laurent123456 9y agoQt Designer for instance is extremely useful to quickly put together an interface and even connect it to code via signals and slots. It generally pays off to go as far as possible purely using Qt Designer and add GUI code only when needed. If a similar tool could be created for React that would be great, but yes it might be hard for a one man project to get to this point of usefulness.
- TuringTest 9y agoI have been waiting something like this for the last 20 years. We had a WYSIWYG editor in Visual Basic and Delphi, but procedural languages are not adequate to build the logic of the presentation layer. The hope is that using Reactive programming, you won't need to wrestle what comes out of the tool, since the underlying language and programming model is finally capable of truly supporting components without lots of glue code, boilerplate and exceptions.
- zackbrown 9y agoI agree that a reactive paradigm is key to unlocking a powerful WYSIWYG view-layer. In fact, I'd argue this has been proven with the visual design tools available for Verilog/VHDL. Hardware is as 'reactive' as it gets. You may be interested in what my team is building: https://haiku.ai https://haiku.ai Haiku is a UI-builder modeled after Flash, FPGA design tools, VB, Qt, even a dash of HyperCard and Unity. We've sought to 'invent' as little as possible, while adapting proven solutions to the many quirks and challenges of the Web. If you're curious at all, I encourage you to join our beta list and shoot me an email for access: zack@haiku.ai
- pjmlp 9y agoThose of us that enjoy Delphi, C++ Builder, Qt Designer, Matisse, Blend, Android, XCode and feel like in the stone age when doing Web projects.
- goatlover 9y agoIt's as if web developers think the web is cutting edge in everything, and there can't possibly exist better tooling for some things on other platforms, particularly older ones.
- pjmlp 9y agoYeah it feels a bit like the bulb programmer, as if they never used anything else.
- mattferderer 9y agoA lot of newer developers probably haven't.
- enraged_camel 9y agoI actually put the blame squarely on older developers. If you don't want younger developers to reinvent the wheels you have used for ages, take some under your wing and mentor them. But hey, complaining on Hacker News about "kids these days" is edgy and gets you upvotes, right?
- pjmlp 9y agoIt is not easy to teach when the students don't want to learn and think they know better.
- TuringTest 9y agoHaving a universal platform with several competing implementations ported to multiple hardware and operating systems, easy deployment, and a standardized user interface model with very high configurability, resolution independent, allowing end-user tweaks to presentation, and responsive resizing of the user interface? No, I haven't, and I was developing before the web existed. I don't doubt there may be much better tools for some things outside the web, but none for a platform with all those features.
- goatlover 9y agoThere is also Smalltalk Pharo, but I don't know how good it is as a UI builder. Still, you have the idea of a programming language environment inside a GUI where everything is a customizable, live object that talks to other objects.
- xemdetia 9y agoYou forget the interactive nature of WYSIWYG editors- if you have tooling enough to throw together a UI in front of a group of people as part of requirements gathering and just 'do you like how this looks' it can be very powerful- especially when you can show up a few hours later and have the broad strokes mostly working. It is also a natural design constraint- if it is not easy to do with a the generator it's probably going to require to do extra work so you can give a near equivalent that is less expensive but easier to do. Without WYSIWYG and current web dev you can bring it to a whiteboard and hash out a design and then find out that for some reason that it is really annoying to make work well on everything you need to support. It's really the consensus driver that makes WYSIWYG always a very helpful tool.