Web App Opinions?

I tried using a web app to replace a regular web site several years back when the web app was first introduced. I did experience a bit of a tough time deploying it at the time and wrote a lengthy how to on the old forum of how I got it working. I was thinking of giving the web app another try to replace a web site I am currently working on.

Can anyone give me their opinion on the current state of the Xojo web app? Pros and Cons? Security? Deployment issues?

Thanks!

I definitely enjoy the ease of use and speed by which you can create a Xojo WebApp. But Web Edition definitely needs a lot of love.

  1. Styles - Styles are a bit of a disaster, especially when wanting to change/create them runtime. There are a ton of missing features that exist in CSS that would be nice to expose.

  2. Lack of Responsive Design - the web edition was made to mirror desktop apps and this isn’t really a good fit for the Mobile Web world we live in. We really need a responsive design column based system.

  3. Lack of Client-Side code - The great thing about languages like angular, typescript etc… is that it is all client side code. Meaning your server only get hit for API calls, and everything else runs in the user’s browser. Xojo is extremely Server based, adding a higher load to the server and making it less responsive and slow.

The good thing is that all of these things can be worked around using the WebControlWrapper and custom javascript (it’s the web after all). But unfortunately there’s a point where its so much customization and hacking with custom classes and frameworks that you’re better off just using something like Angular, Typescript, MaterializeCSS, Polymer etc… There’s also a lack of Xojo frameworks and publicly available frameworks to make this easier or all play nicely together. At a certain point you need to be just as knowledgable in the Web Languages that you’re not getting a benefit from using Xojo.

I know that Xojo wants to redo large chunks of the web framework (especially styling) but I’m not sure where on the priority list it all falls as I’ve been hearing some of this for the last three XDC conferences while seeing minimal improvements.

I just filed a feature request feedback://showreport?report_id=43531

I agree on all the other points but that is such a huge undertaking I am not sure we will ever see that happen.

I also miss having a Xojo TabControl where I could place my objects when a Container would be overkill.

Graffiti WebWall looks promising regarding the responsive stuff. That requires using Containers too.

RubberViewsWE can provide responsive design, but pass simply making sure a webpage scales to all sizes, there is a learning curve to make things work for all devices rotation and aspect ratios.

Note that the same is true for other web tools. The more sophisticated the design, the more learning it requires.

[quote=260901:@Michel Bujardet]RubberViewsWE can provide responsive design, but pass simply making sure a webpage scales to all sizes, there is a learning curve to make things work for all devices rotation and aspect ratios.

Note that the same is true for other web tools. The more sophisticated the design, the more learning it requires.[/quote]

A column system is pretty much standard for any responsive web. It ensures it scales for all sizes and is the easiest to understand.
MaterializeCSS does a good job at this:
http://materializecss.com/grid.html

Or bootstrap
http://www.w3schools.com/bootstrap/bootstrap_grid_basic.asp

If you allow me this metaphor, Xojo Web is not of the same species as web design tools. It is designed to look and feel as closely as possible to Desktop.

The discussion comes time and again, with always the same mention of Bootstrap and other solutions that are, fundamentally, layout and design oriented.

Xojo Web IMO would need to be completely revisited to use the same kind of design tools as MaterializeCSS, Bootstrap or others. Not to mention a completely different breed that people accustomed to Desktop would have a terrible learning curve to deal with.

Not saying it should not be done. Simply that a completely different product would be in order.

@Brock Nash Thanks so much for the response. Much appreciated.