New interface

[fischli]

@fhdeggendorf: I'm not talking about transport security. I'm worried about the content. When it gets onto the smartphone, then it will be out of control. Access to confidential information by smartphone may be necessary, but the document (or contact database) should stay on the server.

[fhdeggendorf]

@fischli:
- Did you ever try to access emails from an encrypted Android partition without a matching password?
- Do you think push notifications are working with a web app which is not running?

[pieter_djc]

Hi Merijn,

Please stay with the ExtJS interface. Not only for the customers of the OEM Partners (like our customers) but also for the available custom made modules.
I know that upgrading to ExtJS 6 also ends in upgrading the user-interface of custom made modules but I believe (and now from my experiences) that the time to handle that is much shorter then upgrade and redesign to HTML 5.

Also non of my customers use the WebGui at their smartphone and for tables the Responsive support is OK in ExtJS 5 and 6.

Before bringing the upgrade to the market please give your development partners time to upgrade their stuff too.

Kind regards,
Pieter

[mdhart]

Hi everyone and thank you for a lot of insight on this matter.

I'd like to share my opinion here as well:

At first it was rooting for HTML5 for below 5 reasons:
1. Faster than ExtJS
2. Not bound by components (more freedom for innovation)
3. Maintainability
4. Easier to theme and style
5. Responsiveness.

All of the above was based on my experience with ExtJS3.4 and ExtJS6.0 has a lot of impovements. I've also realized that Group-Office has a huge interface with much features we use directly from the ExtJS framework. If we release the first version with the HTML5 GUI, it would miss out on a lot of features because it would take a huge amount of time to develop everything we've used from ExtJS.

I then reconsidered the 5 reasons:
1. HTML5 is still faster (very useful on mobile devices)
2. We could still create our own components but do we actually need any?
3. ExtJS6 (if used properly) is just as maintainable as HTML5
4. SASS allows easy theming, it might be not the bad
5. Could be optimized for tabled: 9" screen or higher and right-click/double-click will be optional.

For a Smartphone devices (6" screens and below) I believe we could take a different approach...

PRO: Native apps already exist:
- Email via ActiveSync or IMAP
- Contacts -> ActiveSync or CardDav,
- Calendar -> ActiveSync or CalDav,
- Tasks -> ActiveSync or CalDav,
- Files -> WebDav

CON: These apps might miss features in Group-Office:
- Linking
- Customfields
- Projects
- Timeregistration.

Maybe smartphones shouldn't be considered in the choice for the web GUI for the following reasons

1. Personally I hate having 2 applications for the same task because "App A" is missing feature 1, and "App B" is missing feature 2
I have "E-Mail", "GMail" and "Inbox" on my Android phone I wouldn't like to have a "GOMail" as well...
I would rather see an Adapter that would make GO functionality available by sharing E-Mails from my E-Mail app to the adapter instead.

2. I can imagine that a Smartphone application is used in a very different way, only fitting it on a smaller screen wouldn't be enough anyway. For a time-tracking app I only need a "Timer" and a "Save" button. I can still see the full overview of all my time entries on my computer. With a native app it would also be possible to track the mileage at the same time (if traveling with the device)

My Idea:
- ExtJS6 For Desktop and Tablet
- Native for Smartphones (after finishing the Web/Tab client)

Be aware that this is just 1 developers opinion and not a decision. I'd love to hear you opinion in this.

[piterson]

Hi,
Short comment from us.
We like current interface, as our customers. We get the idea and trends to have all "mobile friendly" design and interface, but lets be honest - as many of you where saying, all mobile function needed we have already with current design. If I would like to "do" something in GO, I simply get my MBA and do what I need to do. Eventually I use some screen sharing program (yes, no right mouse button, but I can live with that).

We are appreciate current possibility to sort columns. I tried to imagine how it was be to change it to some sophisticated search / filters, and I really do not like it. One of our suplier just implemented that in his program and guess what... we do not like it at all... and there are some rumors to change the supplier.

About 99% of our time we and our customers working on workstations/laptops. I could not imagine how i.e. ticket module can be operated more efficent using iPad or alternative - for both, customer and our side.

So, our small vote: Keep it, the way it is, please.

[mschering]

At the moment it's a tie but most commenters seem to be in favour of Extjs 6.

One extra thing to note from a developer and open source aspect. Sencha again changed thier licensing and now only the first (buggy) release is open source. This means that anyone who wants to develop a module for GO has to purchase an extjs license for $4,475,-. This effectively means GO will no longer be open source...

[vuo]

At the moment it's a tie but most commenters seem to be in favour of Extjs 6.

One extra thing to note from a developer and open source aspect. Sencha again changed thier licensing and now only the first (buggy) release is open source. This means that anyone who wants to develop a module for GO has to purchase an extjs license for $4,475,-. This effectively means GO will no longer be open source...

God dam..*lots of cursewords not suitable for anyone*.. Sencha!

As a proponent of open source, such a change should hurt the ExtJS community but i guess that Sencha are now large enough to not even care - especially considering the amount of cash required upfront for a license.

Then i guess HTML5 is the "better" choice here. Maybe we can use http://www.semantic-ui.com (MIT Licensed) to save the time of building commonly used modules and also contribute to it's development. When we built our custom site-builder system, we used it for both front-end and admin interface and it's both fast and easy to debug.

http://jmap.io/ might also be interesting for easier/future-proof integration of E-mail server to client communication.

[pieter_djc]

At the moment it's a tie but most commenters seem to be in favour of Extjs 6.

One extra thing to note from a developer and open source aspect. Sencha again changed thier licensing and now only the first (buggy) release is open source. This means that anyone who wants to develop a module for GO has to purchase an extjs license for $4,475,-. This effectively means GO will no longer be open source...

Now we are in the part of the discussion what will influence the base of GO. I think/believe the power of GO is the open source modules and as addition the extra (closed source) features. That is also the base we develop our extra stuff.
If Sencha does change licensing (and why wouldn't they, its not the first time) then it can be interesting to take a look at other JS frameworks (Like Angular JS) and put effort in that framework. So maybe we should break up the discussion ExtJS or HTML5 and start a new poll with several JS Frameworks with their pros and cons.

[rasos]

Most of our users sync their smartphone with caldav/carddav or ActiveSync. But some do not want to have "business data" on their mobile and they would like to access all functions through the mobile web browser. But the current GUI does not allow e.g. to navigate through E-Mails on a small interface properly.

The new challenge are tablet users! They definitely need a responsive interface, so our vote goes for HTML5.

[mcharvat]

You never know who will change the license and why. People who didn't buy Sencha license before Sencha changed license policy simply had to replace frameworks - mostly with Ionic, pure Angular or React.

If you say power of GO is an open source - what do you mean with that? UI? Functionality? Community? If community how many people commit anything? Maybe community are testers who will use (read "test") webmail (core) for free, I don't know. You can simply compare Roundcube and GO. Which one is better and why ? Both are simple webmails with functionality like notes, tasks, address book etc. So whats the power of GO? IMHO that's UI what makes GO unique.

From poll - how many people who voted are paying customer? I don't know how it works in other companies but here people mostly hate changes - and UI is big change. When Intermesh changed Projects some companies stayed with old version of GO because it was really big change. And now you want change whole UI? Some people love changes other simply not. I would like to see poll for paying customer only because it could show different numbers.

As somebody already wrote you dont need mobile version - just support tablets. Problem with Ext 6 is price. License is pretty expensibe for small projects. I dont know how many licenses Intermesh sold. Currently it's 6 years since Sencha released ExtJS 3. When you stay with ExtJS 6 for 6 years without extending support you simply have to sell 4 licenses (199 EUR per license) per year to cover UI license. You will get Sencha's support for 1 year which is enough (release date of ExtJS 5 was 1st June 2014 and last 5.1.1 release 7th May 2015 - there is also nigthly 5.1.2 but nobody knows if there will be any official release),

Question for Intermesh - when you create different UI do you plan to continue with development and support old UI (ExtJS3) too?

[mschering]

The price is really not the problem. 4500 euro is not much for what you get. If you have to develop all of that it will cost much more. But the GPL version of extjs is fake because they only release buggy versions that we can't use for GO.

Group-Office has always been open source and it should stay open source. So this new policy of Sencha makes the decision easier for us. We can't use Extjs anymore. (see also https://www.sencha.com/forum/showthread ... ost1115793)

But I think we can build a better UI without extjs too!

@mcharvat Whatever we'll choose we will support the current 6.1 with Extjs 3 version as long as our customers need it.

[itdep@vpa-expert.ru]

I vote for ExtJS6
We specially disable and remove any graphical themes from GO, thas are looks as popular social networks (bootstrap-based). The working instrument must looks as desktop applicatios, this is stimulate to working processes, but not to other.

[mcharvat]

@Merijn - yes you are right you can always create similar layout like ExtJS Triton theme - for example http://chandraadmin.com/admin

You can always create custom override to fix the issue. If you think GA is much better than GPL version you are wrong. I'm stuck on ExtJS 5 nigthly because latest GA version had new bugs. With complex application like GO you cant live without these overrides

However the biggest problem is GPL license. With GPL you can publish app to Apple store. But nobody say you cant sell GO for 1 EUR with unlimited updates. Probably in that case you cant use source forge.

There are other frameworks which are under MIT license so you have always choice to select the best. Of course there should be always companies who have problem with different framework because they need overwrite theirs modules and learn new framework. I remain loyal to ExtJS but let me know if you need any help.

[Glatzmatz]

Let me share with you this post.
I do not think the extjs remains an option for Group-Office.
https://www.sencha.com/forum/showthread ... nouncement

[mcharvat]

Let me share with you this post.
I do not think the extjs remains an option for Group-Office.
https://www.sencha.com/forum/showthread ... nouncement

Thats what Merijn already wrote.

Change license to commercial and people will pay (donate) project for (with) symbolic price or use different framework. There is no other choice.