what is next after Ext?

[Bart van den Eijnden]

So this question came up in the MapStory project, say we have to develop a new component from scratch, would we still do it in Ext if the rest of the application is Ext, or would we choose to slowly move things over to our next framework?

And what would that next framework be? jQuery was suggested.

Personally I think the next framework is not gonna be jQuery, but something which is mobile from the ground up. For structuring applications (MVC) Angular seems very popular, although Ember is as well. 

For UI there is a lot of bootstrap out there, Enyo has been researched a bit by CampToCamp as well.

But maybe it doesn't make sense to mix and match technologies in existing projects, and we should only explore new technologies in new projects such as MapMeter?

Thoughts?

Best regards,

Bart

-- 
Bart van den Eijnden
OpenGeo - http://opengeo.org
Expert service straight from the developers.

[Tim Schaub]

I think the biggest mistake would be to pick "that next framework" without any design work to guide the decision (right tool for the job and all that).  I'm personally enjoying being free of geo-specific frameworks and am not too eager to get into discussions about them.  While collaborating on GeoExt proved to be a successful way to share code with Camptocamp and others, I think its also fair to say that focussing on just one UI framework also held us back a bit over the past few years.  I'm interested in broadening our skills for a while, getting up to speed with some of the many solid new choices in JavaScript libraries - even if it means we will have some redundancy across different projects.

This doesn't really help with the decision about what is right for MapStory.  But to talk more about that, it would be good to hear first what the specific requirements for the new components are.

Tim

--
--
http://groups.google.com/a/opengeo.org/group/js
 
 
 

--
Tim Schaub
OpenGeo http://opengeo.org/

[Bart van den Eijnden]

Even though it's personally hard for me not to make a choice I see your point Tim. I have some follow-up questions:

- what do you consider the solid options out there?

-do you think jQuery UI is more future proof than Ext?

For MapStory this was about a new annotations manager, a grid view on annotations for CRUD operations. Also to upload a CSV file for bulk annotations. Personally I see little value of not doing this in Ext since it's part of a composer like application. Also we cam reuse geo row editor stuff from AsBuilt.

Bart

Sent from my iPhone

--
--
http://groups.google.com/a/opengeo.org/group/js
 
 
 

[Ivan Willig]

One thing I think its important to point out is that jquery and ext.js are not at all the same thing. Jquery is a low level wrapper around basic dom creation, event handling, ajax and etc. It is no way a replacement for a batteries included UI library like ext.js. Its comparing apples and beer.. Its really hard to build a large scale application with just "jquery". We tried that with the new uploader in geonode and I really think that you need more then just jquery do these type of applications. 

I would also advise against jquery ui. It seems like no one really uses it and is only good for where you need one or two widgets on a mostly static page. I also don't think its more future proof then ext.js 4. 

To echo a little what Tim said, I think its important to not select a framework that defines how your page looks. The web is a fast moving place, who knows what ios8 is going to look like:). Many design trends have come and gone and ext.js still looks like windows 95. One issue with ext.js, even ext.js 4, is that it defines to much how your application looks. Lets just use html and css to do that. In this way we could use bootstrap css or pure or what else is hip, or even our own css framework. And we could change it down the road. This is almost impossible with ext.js. 

Anyway,

Just my 2 euros. 

Ivan Willig
OpenGeo - http://opengeo.org

[Bart van den Eijnden]

Thanks Ivan.

I know jQuery is similar to ext-core only and not full blown Ext.

So when you said you wanted to do it in jQuery you meant only using the jQuery adapter of ext so you could use jQuery 'style' for eg DOM manipulation? Or not use Ext at all for the UI?

Also think of the cons of just using a CSS framework, I'm sure it will take someone like me tons more work to get something together.

But I agree on having lots of flexibility of what the UI can look like, being locked into a UI that is hard to change is not good. Ext 4 should already be an improvement over 3 in that regard.

Bart

Sent from my iPhone

--
--
http://groups.google.com/a/opengeo.org/group/js
 
 
 

[Juan Marin]

On Thu, Jun 20, 2013 at 1:26 AM, Bart van den Eijnden <bar...@opengeo.org> wrote:

Thanks Ivan.

I know jQuery is similar to ext-core only and not full blown Ext.

So when you said you wanted to do it in jQuery you meant only using the jQuery adapter of ext so you could use jQuery 'style' for eg DOM manipulation? Or not use Ext at all for the UI?

Also think of the cons of just using a CSS framework, I'm sure it will take someone like me tons more work to get something together.

This is why we hired a front end developer. Should be able to help a lot with stuff like that. Our new UX designer also has experience with CSS, she will also be able to help. 

Both start soon :-)

Neither is available for MapStory in the short term :-(

 

But I agree on having lots of flexibility of what the UI can look like, being locked into a UI that is hard to change is not good. Ext 4 should already be an improvement over 3 in that regard.

I agree with what has been said. Maybe what we do is OpenLayers, we basically stop there when it comes to product development. And then use whatever is convenient for us in projects, sharing with the world what we do in the form of templates or examples that others can benefit from. As for MapStory, I don't know all the specific details, but over time it will have to migrate to new technology, in accordance with the goals of the project and with agreement from the stakeholders (migrating to anything takes time, they need to understand the benefits and agree on the investment).

--
Juan Marin