From my previous experiments and communication with @rodlie there is “official” interest in a new website. The current website mockups are much better than what exists design wise in my opinion and once implemented with Jekyll for news and releases should be a superior platform to build upon for the future as well. IMO the website redesign here is probably in the realm of “officially wanted but not endorsed yet by the two developers that actively work on the program”. I have full confidence that if we develop it to the point where it’s implemented in code with Jekyll it will be easy for them to approve!
The program UI is a more complicated matter, I’ve experimented with styling Natron’s QT Stylesheet and have made a whole bunch of improvements by duplicating and implementing styling from Blender, but QSS is stuck in CSS 2.0 and for any fancier features you’ll have to make some larger changes to the program that I haven’t bothered looking into yet. I am also of the opinion that energy is best spent making Natron’s existing UI better rather than completely scrapping it and starting over. There’s no reason to stick to “this is how Nuke does it” for everything but the general layout, system of panels / tabs, node-graph, and properties panel is good. I’ve outlined my other problems earlier in this thread if you really care about my opinion there.
Website mockup thoughts for @Songtech-0912: It looks like a good start but not something that I’d find a huge amount of value in doing right now. Would much rather we focus on getting Figma set up and do a mockup of every piece of new content in there to really figure out the web design aspect of it first. It may seem attractive to get right into live websites but this really isn’t where time is best spent if you want to nail down how it should look and behave, there are plenty of small typographical things I’d do differently, mostly small sizing changes.
Right now you’re writing everything with npm and Tailwind and not using Jekyll or GitHub Pages’ build system. Creating everything with tailwind and just exporting it to SASS files may not be the best way of creating a maintainable codebase because we need to deploy it on GitHub Pages which, as far as I know, does not support npm and has no build process other than Jekyll’s. If we can write it with Tailwind and have GitHub Pages’ Jekyll somehow work to build it and output it to the _site directory then that’s great and I’m fully on board! If it is a “write it in Tailwind once and then export it and then that exported single CSS file is how the website works from now on” I do not think that is a good idea. The whole point of Tailwind is to minimize the amount of CSS that you have to write but if SCSS is what we have to maintain and build upon in the future I find the benefits of that system to be somewhat short sighted.
Any reason for just including SVGs right in the code over an icon set? Personally I like Boxicons, it’s got everything we need since Atisa added the tux icon for Linux! Makes for cleaner HTML.
I can program you this website in a few hours but there is no point when the design (although very nice) is not 100% locked down and agreed upon. DM me when that’s done if I stop following the thread.
I for one think that the website needs more content showcasing Natrons features. It also needs a video, like a showreel videos. We need to find out if anybody has used Natron in a professional, just cool or viral way/project and ask for a cool shot in different stages of the Natron pipeline.
Or maybe we just need to get off our buts, go out and shoot shots for the purpose of showcasing Natrons features. It wouldn’t be as cool as if we had shots from a Marvel movie or Avatar set but maybe that’s the point.
Maybe Natron should appeal to small, one man band wedding, event, indie videographers, students and video enthusiasts alike?
Maybe that’s the way to traction for the software. To market it to that part of videography market that would never use something like Nuke because it’s too complex and too expensive for offering many features that this particular market doesn’t even need.
Anyway, I’m just now learning Natron I might not know what I’m saying yet.
@Shrinks99 While I totally agree on polishing the website (probably through Figma, which I am already experimenting with in my spare time, I do have an automatic build script that uses npx for building the webpage. However, I am highly aware of the limitations of Tailwind compared to SCSS, at least for us. My current understanding is that Tailwind will only be used for prototyping, not for design or for the final website, and Figma with a few plugins can fill that niche as well.
Meanwhile, my experiences with learning Figma have been positive so far, and the collaborative features are great (even though I haven’t used them yet). I’ve created a Figma project for any of you to hack on: come and edit it here.
@KristijanZic I’m actually doing a scifi-esque photoreal shot with Blender and Natron right now and I’ll release it with creative commons licensing for the renders and project files sometime soon. The pandemic has been slowing things down a bit but I’d say that it’ll be ready in a month. I’ll try to do as much as possible with Natron only (e.g. matte paintings instead of 3D backgrounds, Blender will only be responsible for the 3D models). And considering Natron’s limited 2.5D capabilities, it is unlikely to directly compete with Nuke for some time, so that might be a step in the right direction.
Why use Tailwind at all then? Perhaps I misunderstand you when you say “prototyping”? Creating fully working HTML and CSS is not a prototype, it’s a real webpage that you could deploy!
Will have more time next week to work on things in Figma, glad you’ve got the project set up I see you’ve got it set to anyone with the link having editing access, pls restrict to set users and add people individually, I’m not about to put a whole bunch of work into web design just to have some completely random person delete it all Anyone who has commented on this thread or is active on this forum seems cool tho, not trying to gate keep or anything.
As for Natron’s place in the market, because it parallels many of the same workflows used by Nuke it also (by nature) gives you a bit of a lower level access to your images compared to a tool like After Effects or Photoshop. This is good in some ways because of what it allows you to do, but also adds to the learning curve of the software. Right now artists are all pissed off at Foundry for their ridiculous licensing and cost of their software that prices out freelancers and smaller studios, there’s definitely a need for some competition and open source software is in a better place than you might think to deliver at least some of that.
You don’t even need figma. It just adds more overhead. It’s a very simple design and will most likely be a one page one. You can design it just in inkscape and then a web developer can just pick it apart for icons, images etc and develop a working website.
Please, don’t use any frameworks or scss. This is to small of a website to have all those abstractions and will end up being too large and load slow just because of all the unused dependencies that come with frameworks.
But I like the general direction of the design, I also liked the first ones you’ve posted here. And it could be developed as a nice and lean one page site with all the cool stuff to show.
Personally I find SCSS very useful for breaking things out into nav, content, and footer at least, grid system too if you don’t plan on using CSS grid. Using the pre-processor also lets you build in minification and comment removal for free so that’s fun too. Would like to have a figma doc for some other things, this project will probably go beyond the homepage.
@KristijanZic Perhaps “no frameworks” could work, but unfortunately we still need Jekyll for the blog and I don’t think the main devs are willing to take that away. In addition, I have the base design in Inkscape already, but Inkscape is badly suited to web dev e.g. no prototyping, no auto layout. You can see this file if you really need the original SVG though.
I’d rather go with Hugo too but it’s not supported by GitHub Pages last I checked
Agree on CSS grid, comments ehh but sure that’s a super small deal, I’m not a programmer so I can’t comment too much on standards and practices with web dev, maybe I just don’t have my text editor set up in a nice way but I still like separating to different files for organizing things. Change my mind!
@Shrinks99 Actually it seems that Hugo is supported by GitHub Pages (see here for details) and we’re not going to have that many articles, so we could switch to Hugo.
It supports it in the same way that it supports static HTML. Jekyll on GitHub Pages actually renders your Jekyll site for you, you do not have to host or deal with the generated site yourself as long as you replicate their build environment. PRs and updates are only handled on Jekyll files and you never have to touch — and most importantly never have to review — the generated website ever.
It’s quite nice. I like Hugo more don’t get me wrong, just makes a whole lot of sense for this project to do it the intended way so that management is easier.
But it’s a first class citizen at GitLab, so that’s what I’d use
Sure, open a terminal and execute touch ~/.vimrc.local now vi ~/.vimrc.local to start setting up your vi/vim editor just the way you… Jk
Well, for years I’ve been using Atom but since Microsoft bought GitHub I’ve been using Visual Studio Code and Codium (which is VS Code without Microsoft trackers (you can opt out in the VS Code too)).
Yeah, if I had to guess @paperdigits is talking about GitHub actions whereas Jekyll and GitHub Pages just work with no setup. Unless something massive has changed in the last few months that I’m not aware of Jekyll is the only natively supported static site generator with GitHub Pages.
End of day update! Everything above the downloads section I’d consider to be largely finished. There are a handful of things to still tackle, all of which are noted with comments in the Figma project.