Hi Guys,

I just thought I would start this thread to see if anyone else had some suggestions for how the usability (more manageability) of the e-commerce module could be improved in the future. I have been mulling over the idea of making some changes to the way the e-commerce module works to better suit the way I think it should work and I would love to hear other people ideas, opinions (good or bad) and just general musings on it.

Here is my idea. At the moment the e-commerce module becomes very unwieldily as soon as you have a large (100 plus) products. I believe this is purely because each product is set up as a new page in the system. I was considering creating a page (Product Group) which then uses the DataObject to attach products to this page rather than each product having a new page in the system. This way you would only have pages for the groups/categories in the site tree on the left of the admin panel.

If this is already available somewhere please let me know. I currently have a client who has about 800 products in about 50 different categories and I am worried that the current setup will become too difficult to maintain in the long run as she adds more and more products and categories.

If anyone has any thoughts or suggestions on how to improve the e-commerce module I would love to hear it.

Kind Regards,
Colin

@ccburns, I think this was a very thoughtful post and I agree. However, as to how to resolve the issue, I think the better solution is to have a module created based on another open-source piece of software, like Open Cart or PrestaShop. PrestaShop itself is modular, so, it may be better, however Open Cart is very nice too, out of the box. I think the problem with the E-commerce Module, as good as it is - and I am thankful for those who have worked on it and developed it - is that it will never become the robust module that we all desire. So, why re-invent the wheel? Why not just take another open source project and hack away at it to mold it into a native SilverStripe module? ...I don't profess to have this ability, to do this, but I think it is the only logical thing to do. ...Maybe we all could "chip in" for someone to get this going and finished?

Thanks @Yakiv, although perhaps being a bit of a traditionalist I am not a huge fan of that style of integration as in my experience it throws up too many other integration issues, it also means that the direction of a particular product is inextricably linked with another product (again this could be a good and a bad thing I guess).

Really what I think I would like to see is more use of the ModelAdmin to do this type of work. Were basically we have a list of categories and a list of products linking the products to the categories like you see in a traditional e-commerce module rather than what we currently see. I am about to start work on something like this although my time is pretty limited at the moment it'll probably take me a while to get something for others to see.

[REQUEST FOR HELP] When it comes to programming I am a bit of a hack, so once I have things up and running I would be really grateful if I could ask for some help to refactor the code so that it fits better with the overall theme of SilverStripe and methodologies that it uses.

Anyway, until I have something to show people I would be grateful if anyone had any suggestions of what they would like to see in an e-commerce module. What do you like about the current one and what don't you like.

BTW, if you are the original developer of this module please feel free to chime in (this thread isn't/wasn't meant to be offensive in anyway as it's awesome that there is something that we can all use now). I'd just like to make the addition of new products a little easier and less cumbersome when there are a large number of products.

BTW, if you think integration of another system is a better option please let me know and tell me why so I can understand your thoughts better.

Thanks,
Colin

Colin, you said, "it also means that the direction of a particular product is inextricably linked to another product." I don't agree. What I am saying is that you could use Open Cart or PrestaShop as your base software. Since it is open source, you are free to fork it into another project / product, specifically making it a native SilverStripe module. Then, this module could progress on it's own or take hints and upgrades from the original software's development, but I don't agree that it is inextricably linked to the other product.

If you have never used these other stand-alone products, you may not realize the power of them. Again, my point is, why re-invent the wheel. If someone has already produced a powerful piece of software, why plod along with a tiny, insufficient start-up module? It doesn't make any sense. ...This leads me to your call for idea requests. The Open Cart and PrestaShop pieces of software are already doing much of what you would be asking people to come up with ideas for. So, why not download these and see if you can hack away at them to make them a SilverStripe native module?

@Yakiv

The problem with forking another project to make it part of Silverstripe is that you will end up with frankensteins monster. The architectural differences between Silverstripe and any number of other eCart projects will be huge. At best you end up with a beast that might function well on a site but is horrible to maintain.

I also agree with Colin that there are some serious limitations in the architecture of the current module that have arisen by adopting the page per product model, but given that this module predates ModelAdmin this is understandable. Also order management is a huge problem.

We have deployed it once so far with another project about to kick off. Its acceptable for small scale projects below the scope and budget of Magento. If a client is serious about eCommerce then we would not consider this module as it does not offer the kind of management and feature sets that Magento does. That said Magento is a whole can of worms to deal with as well - not for the feint of heart!

Perhaps its time for an architectural overhaul and a new version for 2.4 - no pressure on the developers ;o)

I completely agree with you. The current architecture suits for small scale e-commerce sites but when they get larger it becomes an issue.

I won't be considering hacking in another e-commerce system for the exact same reasons @jasonstratford has mentioned.

My main reason for wanting to re-wrok this module is often I need a small/large store but also other functionality that is very easy to build with SilverStripe.

Anyway, it'll be a week before I get to start working on the system as I have to look after the kids this week while my wife is in hospital ;)

Cheers

"The architectural differences between Silverstripe and any number of other eCart projects will be huge." - not a good enough reason not to do it. Have you ever heard of Frog CMS? It's a PHP version of Radiant CMS (ruby based). So, I cannot agree with the reasons you are giving, although I can certainly appreciate that this would be a "monstrous" task. ;-)

I think that usually people give the counter reasons because A) they are not capable of doing it, or B) they know they would not be patient enough or dedicated enough for such a project.

...Be that as it may, I would be happy with even a solid bridge between the two at this point. One of the main issues is the users that will be generated from two side-by-side pieces of software. Keeping them synchronized and a seamless, single sign-on is important.

I just know for one I wouldn't have the patience or (probably the skills) required to integrate two disparate system, although I am sure there are people out there who do have these skills.

Cheers,
Colin