X-Cart forums

[xim]

Hi xcarters,

I would like to share our plans regarding the 2016 roadmap. I will provide some details on the future of the X-Cart 5 platform today, and share the roadmap for v4 separately next week.

We believe that X-Cart 5 is the best platform to start a new business. Due to the flexible architecture, it allows you to keep up to date with the new releases and features. As a result this platform grows with your business.

There will be one more major update in the 5.2.x branch. Here is the roadmap: http://www.x-cart.com/roadmap.html. In addition, we are planning to close most of the existing bug reports in the bugtracker.

After that we are going to start working on the 5.3 version. Here are the key changes that we are planning to implement:

Integration with segment.com

Segment is the universal glue to integrate your website with almost any useful service on the web. You can see the list of their integrations here: https://segment.com/integrations

With Segment integration, our merchants will be able to import data into all these services in one click. This solution will provide our merchants with a set of very powerful tools to increase sales and conversion, analyse data, make reports, follow up with customers and partners, work with search engines and so on.

A new vertical checkout template

The 5.3 version will have a new modern and simple template featuring the new vertical checkout.

It will be an HTML5 template based on Bootstrap, LESS and CSS3.

Drag-n-Drop design tool

In 5.3 we will implement a tool to drag and drop the blocks and sections around the website and manage the visibility of these blocks and sections in the customer storefront.

Better Template Structure

We want to make the customization process for template designers as simple as possible. Thus, we will re-design the template structure to make it more straightforward for the new developers. We will also implement the ability to create a new theme by adding a new folder in the “skins” directory.

After that you will just need to choose the new design in the admin area, copy the template files you require from the original folder to the new one and customize the templates as you want.

So, it should be as simple as in XC4, but with better upgradability.

We also remember about our 3d party developers. We will provide them with the guidelines for the major changes in the template structure to adapt the existing functionality to the new version.

Replace Flexy with Twig

Flexy is a very clear and simple template framework, but it hasn’t been improved for several years. We need to move forward. We did some thorough investigation as to what we should choose - Twig or Smarty - and selected the first one. Twig is very similar to Smarty by way of syntax (hey, v4 platformers), but much more powerful and popular. At the same time, our development team has managed to tweak some Twig classes to make it work much faster.

We created an automatic converter to re-code over 2000 template files that we have now, and we will surely share this converter with the community to adapt the existing functionality for the new version.

So, more documentation, more examples, more functions, more flexibility for customization - http://twig.sensiolabs.org/

Editing of product groups and other UI/UX back end improvements

We collected a lot of requests to implement the product group modifying mode. This is a very powerful feature, but we want to give our merchants the best experience, so we cannot just copy the UI/UX from v4 to v5. Now we have a concept of how to make it easy and clear.

At the same time we will re-work some input fields and widgets in the admin area for better usability for merchants.

Also we want to implement an easy way to customize email notifications.

Code base improvements

We are going to increase the minimum required php version to 5.4. We would prefer to make it 5.5, but we understand that most of the hosting companies will continue using the older versions in the nearest future, so it is going to be just 5.4 this time. This change will allow us to update the Doctrine to the latest version which will make X-Cart faster.

There will also be some other improvements:

• The template executive functions will use absolute template paths instead of the “magic” logic when the system detects the “admin” and “customer” folders automatically. At the same time we will re-work the compiling process to reduce the number of created objects. This is a huge task that will affect most of the classes, but it will make X-Cart much faster.
• At the moment we have separate template sets for the admin and customer areas. Even if some files look the same and do the same thing, they still have to be separate files for each of the areas. We will add the ability to have common templates and widgets that can be used in both the areas, admin and customer, without the need to duplicate the respective files. At the same time we will provide the ability to apply a common template to just one of the areas by extending it with a file for that specific area.
• The structure of LC_Dependencies will be changed a little bit to improve the API for developers.
• The engine for the inclusion of static assets will be improved to redefine CSS classes more easily.
• The profiler will be improved to make it better suited for X-Cart classes. At the moment it shows generalized information. We want it to be more specific, which will be of value to developers as a tool for finding the store’s performance bottleneck.
• We will implement the Dependency Injection principle - https://en.wikipedia.org/wiki/Dependency_injection - to make the code more modular and mantainable by delegating object creation logic to the IoC container.

Surely, we will provide the community with the detailed explanation of what we have changed and the instructions they will need to follow to adapt their custom modifications and modules to the 5.3 version. For some changes (like flexy -> twig) we will provide an automatic script to convert the code.

These are our thoughts about the 5.3 version, and we want to discuss them with the community. Feel free to share your ideas and thoughts here.

[cflsystems]

Originally Posted by xim

Replace Flexy with Twig

When I just started to get a grip on Flexy...

All this sounds like a new cart system. I think you should give us at least 2-3 months with beta before official release.

What is the approximate release date you are thinking of for 5.3?

[xim]

Originally Posted by cflsystems

All this sounds like a new cart system. I think you should give us at least 2-3 months with beta before official release.

What is the approximate release date you are thinking of for 5.3?

It is the good question. We are planning to implement all the codebase improvements first and release 5.3 alpfa/beta. After that will start working on the feature set. I believe we will show 5.3 prototype in March-April with the official release in May-June.

We definitely need at least 2 months for beta-testing.

[hartwellj]

The problem with all of these changes, is that you're not building. You are re-creating. You guys are making things completely impossible for existing XC 5.x.x users to update. All of the custom CSS will be lost, and the mods break.

You keep changing the core so much and it's just becoming overwhelmingly expensive to upgrade, seeing a person has to keep paying to adapt their modules.

This is why I'm seriously considering moving down to XC4.

[xim]

Originally Posted by hartwellj

The problem with all of these changes, is that you're not building. You are re-creating. You guys are making things completely impossible for existing XC 5.x.x users to update. All of the custom CSS will be lost, and the mods break.

You keep changing the core so much and it's just becoming overwhelmingly expensive to upgrade, seeing a person has to keep paying to adapt their modules.

One of the main important point in the current requirements for 5.3 - is to save the upgradability for existing merchants. We understand that the changes are huge, but we are going to make the adaptation process very simple.

We already have the beta of the Flexy->Twig script convertor to re-code all the files automatically, we will share it with the community after QA phase. So we will invest a lot of resources to save the upgradability and help our merchants and developers to reduce the adaptation costs.

[hartwellj]

I'm hopeful, but just doesn't seem practical. I've talked to a ton of other XC5 owners and the bugs are endless. It would seem the focus should be more on bugs.

The problem with the new versions, is that there is no way to know if it breaks the logic of a custom mod or other mods. So every upgrade becomes a new site all over. Just when you tackled the bugs, you start all over with a new version. There is no way for the script to pre-predict breaking the logic of a module and in a production environment, merchants loose money.

The upgrades between versions in XC4 were much more seamless and with less headache. I'm not quite sure you guys keep updating the core so often. I would trade stability for features any day of the week right now, because the goal of a merchant is to make sales.

Originally Posted by xim

One of the main important point in the current requirements for 5.3 - is to save the upgradability for existing merchants. We understand that the changes are huge, but we are going to make the adaptation process very simple.

We already have the beta of the Flexy->Twig script convertor to re-code all the files automatically, we will share it with the community after QA phase. So we will invest a lot of resources to save the upgradability and help our merchants and developers to reduce the adaptation costs.

[totaltec]

Originally Posted by hartwellj

I'm not quite sure you guys keep updating the core so often. I would trade stability for features any day of the week right now, because the goal of a merchant is to make sales.

I think everyone would appreciate less frequent updates. I was early into the mix developing modules for X-Cart 5. However when I realized how often I would need to update my modules in order to keep pace, I lost interest. I think releases should be 1 year apart, or at least 6 months. But you guys already know I feel this way.

All of the improvements you have outlined above sound great! Looks like a solid plan Max!

[kevinrm]

From my experience, the bugs have been getting less and less as the versions progress. There were still quite a lot of bugs up to 5.2.10. There are still some but they seem to be manageable.

In our case the upgrades haven't really broken any of my custom mods except for one, and that was fixed rather expediently by the mod developer.

I've been upgrading since 5.0 but I only went live with it at 5.2.4, mistakenly assuming the cart was ready to be used for live commerce. It probably wasn't really ready until 5.2.10.

For sure - I won't be upgrading to 5.3 anything until I'm really sure it will be a smooth transition.

[cflsystems]

Originally Posted by totaltec

I think everyone would appreciate less frequent updates. I was early into the mix developing modules for X-Cart 5. However when I realized how often I would need to update my modules in order to keep pace, I lost interest. I think releases should be 1 year apart, or at least 6 months. But you guys already know I feel this way.

Glad to see you are coming around now

Especially with the forced 1-year subscription model and no other way for 3rd party devs having to change modules every 2-3 months is not good.
XC5 needs to step away from the XC4 model when upgrades are released very often and mixed. It has been said many times - DO NOT release mixed upgrades. Or at least allow users to install bug fixes only. Bug fixes usually will not break store or non-XC modules. New feature may do this.
Forcing user to install new features which may or may not work just to get one simple bug fixed is not a good practice.

Also - have 2 upgrades per year max. No one cares about the version as long as it works. Like it was mentioned earlier in the thread - "I would trade stability for features any day of the week". Having stable cart system that can actually produce good ROI is way more important than having a non-working system full of features - they cannot be used anyway....

[RichieRich]

If the software needs to move to a different type of code that can be converted with a script then Im sure that is best for the software if that decision has been taken, especially if Flexy is not updated. The question is how seamless will the converter be. Im sure it sounds easier than it will be.

Just want to add, I think right now X-cart support and with the bug tracker is the best I have ever seen it, and that is using since 2004.