I tried to follow along with that documentation but it has the user create a deployment on github codespaces which looks like it is on the cloud, where would I find it on my PC following the docs?
Ok I better try and get docker working once again on Windows 10 or just bare metal an old box for proxmox as it'll probably be quicker 😆 ...
Once I do I'm assuming connecting to the localhosted CMS is going to be nice and easy, like literally just picking JSON/REST/GraphQL for the data flavour in Webstudio and it'll all work nice and easy without requiring an internet connection on the server?
Are there plans to have an easy to use and setup local environment for designers using Webstudio in the roadmap that we can use to work on even without internet like Herd/XAMPP/Node etc have let us do for years?
Ok after a few hours getting Docker working on Windows I've followed the guide to install Webstudio with Docker and VS Code but I get this error when trying to start it.
I am assuming that the following in the docs means downloading a .zip of the cloned fork of Webstudio and opening that in VS Code? Apologies if this is basic stuff for developers but it is not very clear
Forking is more about contirbuting rather than just running locally
@TrySound so downloading the zip file from the forked repository of Webstudio is the best practice method for running it locally? I'm following the docs, perhaps a guide for developers (who may want to contribute) and a guide for designers (who want to use it locally) might be useful?
The best is git clone command
Well, let's say webstudio is not ready to selfhost for designers
It will be improved in the future though
Every designer will want to just download the software install it on a mac or a pc and start working locally and connect it to existing CMS's easily that are either running locally or on another server, especially if working under NDA's etc.
And it will be possible in the future. Though we are very small team and should prioritise some features over others.
What is the best advice for Designers wanting to test the software with various CMS's from a local install of Webstudio (as in a program you click on from Program Files on PC or Applications on a Mac)? It is quite confusing as the term self hosting is assumed to mean traditional software?
The best way to test all features is to start free trial of pro in our cloud
I want to work local first and cloud last though
I like to keep it as simple as possible
Local setup is not simple at the moment
It must be easier than doing cloud based functionality? From a development perspective?
Windows can be the issue. We all work on mac os.
You don't test for x86 then?
It's not about x86, but windows specifically
The errors a few users have had using the instructions in the documentation in failures for the local dev container that runs on Linux is a Windows problem?
How come you don't just sell Docker containers already set up connected to various CMS's as examples for like 20 USD/Euro/GBP/AUD/Localised equivalent based on PPP for localised development (and proof of concept)? Surely you have them already for development of the software. I'm just floating ideas as a potential customer and I appreciate the time answering. Best wishes
Or just sell one for designers with a blank connection to a cms but with say 3 flavours of server infrastructure to add your own cms if its a licencing thing. Include LAMP for compatibility for the vast majority and it's a useful boost for development funding and adoption.
It's simple - we are a small team and we are focused on the things on the roadmap
Local dev is mostly for people who want to contribute.
I was under the impression the products target market is designers so apologies if it is not the case.
Why didn't you just make a front end open source locally installed and hosted first? Rather than joining the crowd of going straight to the cloud. Your target market has been expecting some developers to do it since 2010?
I'd pay for that but no one's made it
I'd also pay for a cloud hosting on top, but never first
designers and developers are the target market
They've made a great product here. You can get a lifetime license for insanely cheap, and you would never need to worry about hosting, servers, bots, etc. They give a very generous plan for free to test with lots of features and the ability to self-host and do everything.
I bought the product a while ago and tried setting it up for self-hosting afterward. It was a little complicated, but I got it running (with community help). I will be sticking with the cloud version as there's zero hassle for me.
You can do everything you want FOR FREE but you need to have some ability to understand the docs.
I don't think you represent a majority. Also we will get there at some point.
I'll happily pay for decent local first perpetual licence software and for decent hosting as a designer. My main consideration for any software tool I may be in favour of is how to use it in production working on projects that require not to be hosted on other peoples computers. I'd only be able to suggest it for project work if it meets those requirements.
Perpetual licence as in that major version and pay to upgrade worked well for design software imo.
I went through the whole downlaoding the builder and docker stuff and after analysis and some experimentation I can tell you that the most simple solution is like they suggest using their pro licence.
Now about your hsoting concerns, what is a little confusing at first is how this software works. The builder and the resulting app are in the end two different things. You can use their cloud plan to build stuff and then you can host the result wherever you want.
What I'm planning to do is for clients to get their own license for the builder so that if you need to just give them all their stuff they just use their account. In the end the application itself will be hosted wherever your client deems necesary (and the included limits on the pro plan are itself quite good for a lot of businesses)
The only real benefit of a local builder instance is if you are planning to be off the grid.
You can host this on your own cloud or personal server. It meets those requirements right now, but it might not be on the specific server you're looking at. No need to bother with all the extra MAMP stack stuff.
Oh i am planning to use that
but for the moment its just easier and faster to get them going with their own
specially with no "multiplayer" yet
I'm looking to use it to do decent work not repeat the worst of Webflow and validate a shit business model...
it's clearly not for me if my original question has not been answered.
We are currently in the phase of building features and making this software powerful. Making it also easy to use on a self-hosted instance is quite a big task on itself, so currently we can't say this is a great idea. We just don't have enough engineers to support this type of usage in an easy way. I would suggest to stick with wordpress if you insist on hosting the entire platform yourself, at least for now.
Self-hosting the site, should be easy enough though. I am explicitely talking about self-hosting the builder platform. Builder and sites are decoupled.
To recap: You can download the builder and host and use it locally to connect to whatever backend you want.
The app built can be hosted wherever you want.
No, the builder is not a software you download an installer so the process is not at easy, but getting a docker with the git clone to work is not also that complicated.
They are developing a platform with priority on cloud use because it is objectively easier to use. And at least I can say I also prefer their team focused on growing the tool than impleemnting niche uses.
They are developing a platform with priority on cloud use because it is objectively easier to use.
Exactly. Also we are buiilding towards very powerful features which need to be done first before figuring out how to provide an easy way to self-host that.
It would have been far easier and I dare say more profitable from a development perspective to just build what the market desires. I've not used wordpress since about 2008. I've only looked at it recently and core hasn't changed much IMO, it's still a pain and code to do any other form of page than a blog as that's what it was designed for from the outset.
Your main concerns as a brand is how to make sure you get adoption and mindshare of creatives whom may see the value in your product, which fundamentally comes from trust.
You're upset that an open-source project (with 100% free features when hosted locally) hasn't made a self-hosted version on your preferred tech stack. Then you criticize them because their business goals don't align with what you want and then think you have a mic drop moment. If you don't like the product, you don't need to use it. It's that simple.
It's "open source" in Webflow branding terms. Here's how the Webstudio business model is intended to evolve that we've all seen before:
- Bring to market an MVP inline with whomever provided seed funding. This is not aimed at providing a market need, it is more involved with stage 3.4.
- Get designers/creatives enthusiastic about you MVP so they're trapped into the platform. But only host on your terms.
- Use these entrapped customers as proof that the product works with figures for further venture capital.
- use all that VC second stage funding to spend on marketing trying to sucker new customers into the entrapment scheme
The number one thing that 'developers' need to get into is design thinking. Read it, understand it, do it.
over engineering is a detrimental coefficient on productivity.
keep it simple and Kaizen
Odd. I don't remember nominating anyone to speak for "every designer". It sounds like Webstudio is not a good fit for what you want to do for YOU.
That does not mean there is a flaw in Webstudio or the choices they make. It's their platform. For many of us designers, it works for what we need. And pretty darn flawlessly for being fairly new.
The other part of this thread I find confusing is your statement: "over engineering is a detrimental coefficient on productivity. keep it simple and Kaizen".
I agree, but this entire thread sounds like someone spending hours trying to force a round peg into a square hole. Then yelling at the peg for not fitting.
Maybe that time is better spent looking for a square peg.
there is an absolute flaw in Webstudio, going cloud first
the whole premise is reliant on people like me to work for it, nah blud thats not running tings proper.
I mean that multiple US companies have all managed to make decent front end design tools so that designers can get on and work. Yet, alas they are all on developer terms so don't do what we need. I get it, I get y'all feel superior and love some new abstract bullshit that's been hidden from normal people for well over 35 years. Imagine how long it would take to make Quark express, or Maya, or PhotoShop if web developers were involved? None would exist.
That's my point though. You're essentially asking for advice and then taking a piss on those trying to help. I have no connection to the WS team but I can get behind a crew this passionate. Creating something that many are already finding useful. I've been designing since the early 90's and what Webstudio is doing here is the first time I've been this stoked about something for a good long while. The designer vs. developer is a tired trope. Take your advice of Kaizen. Always improve. Maybe leave that designer superiority checked at the door. Seriously. Speaking from a 30+ year career in design myself. I've been there too. Life is better working in tandem then apart.
If you want something that lets you develop in a local environment that's not WordPress, check out Bootstrap Studio or Pinegrow.
I'm saying this as the web really does need a decent front end design tool, that has the ability to be trusted by both designers using it and at board level. The only way that's doable is at a local level.
What's the best way to set up a local development environment then without internet access for use on confidential client projects then?
it should be a simple Docker container if rather than an abstracted one via VS code IMO.
Why is the back end choice infrastructure correct and why should we trust it John?
I was just addressing this
it should be a simple Docker container if rather than an abstracted one via VS code IMO.
I'm out
If it runs on my shitboxes I'll trust it more locally without internet access
@Deleted User your point is valid regarding the ease of hosting the builder with docker, we have NOT made this easy yet and I already told you that.
Overal tone of this thread reads more like an attack, so I am out. Majority of things you said about what we do and why are your personal assumptions that have nothing to do with how I think or how we work.
Our priorities are 100% value focused.
Since you chose that tone, I am going to quit this discussion and never respond here again. Rethink how you talk to us.
@Oleg Isonen I really do appreciate what you all are trying to build, I may come across confrontationally but in real life I'm not, I'm just in favour of what people like me may need (and will pay for). The main thing is making something that critical yet supportive people like me will support and advocate for. Whether you embrace that or not is up to you. @degreyghost it is a dogshit peice of software and why I'm here.
I got annoyed post purchase as from the name I assumed it was a tool to make your own design language. But alas it was someone else's CSS framework which is not what I'm interested in. I've not used it for a while though...
