dtcnet.co.uk

I posted way back at the end of June on my use of Magento over Cubecart. Well it’s been seven months since then and I’ve had a few people asking how things are going …so here are a few thoughts.

My first Magento store is now open and doing pretty well. There were a few …well, lots of problems along the way, but most have been ironed out. I think a lot of it is learning how to use Magento properly – and I’m still learning.

I had to sit down with my client for a good few hours at the beginning so we could figure out how to do product options. She’s now pretty proficient at it, so despite being amazingly complicated, it appears to be manageable.

One of the major problems I had was with uploading images. Each time we attempted to upload an image everything would seem to upload ok but then Magento would act like you didn’t upload anything and then log you out on the next reload. After extensive trawling through the Magento forums I discovered that two lines in a php.ini file were all that was needed to fix the problem (extension=pdo.so and extension=pdo_mysql.so). I think that the fact that Magento was unable to prevent this or figure it out for itself shows its age.

Another feature which shows Magento’s age, and actually makes it feel like it’s still in beta, is table rate shipping. It sounds perfect in concept, and the usage instructions on the website are pretty comprehensive, but it simply doesn’t work. Not just for me either, so many people were posting comments about its problems

I think my biggest problem I’ve had, actually a humongous massive problem was with updating Magento. I dread updates and avoid new addons. On the surface the process seems very automated and straight forward, but I’ve tried several times now and none have been unproblematic to the extent that the Magento installation usually ends up completely broken. I was only able to fix the problems by doing fresh installs. Magento is so obese that the length of time it takes to back it up (especially once its populated with products) renders updating and extending completely impractical in my opinion. Furthermore, customising a skin is incredibly complicated to start with, but maintaining one between Magento updates is simply painful.

Don’t get me wrong, I’ve decided that Magento’s rich feature set completely kicks Cubcart’s arse. Product options, associated products, and discount codes are amazing and the one page checkout is pretty good too, but many features still seem unpolished, and user-unfriendly. However before Magento can assume dominance in the online commerce market, I think it needs to transform from a online shopping cart with very limited CMS functionality into a full CMS with complete commerce functionality.

Finally, and most importantly, a stable and reliable method of updating which does not require impractical backups and does not break skins is paramount for any progression.

Just 66 days after the release of version 1.2.2 , dTabs 1.3 is now available for download. 1.3 fixes several bugs and adds few new features including:

• New “between” argument for dtab_list_tabs for adding content between tabs.
• New “fadetime” argument for dtab_list_tabs to enable control over the length of time it takes to fade.
• Provisional support for javascript free css menus.

Kubrick Tabs has also been updated to support dTabs 1.3.

Since discovering Magento, I’ve stopped developing with Cubecart. The first thing that attracted me to Magento was that it was free and it didn’t require each page to have “I’m a tight-ass using the free version of some cheapo shopping cart” in it’s title …and it was template driven.

After looking into it more deeply I’ve found that Magento is much more sophicated than Cubecart, supporting product options, grouped and related products, multiple stores, discount codes, “seo friendly urls”, and customer whishlists, all “out of the zip archive”. However after working with it for a few weeks now, it’s becoming more and more obvious that Magento is still only a baby. While it’s seemingly bursting with amazing features, some of them seem incomplete and most seem to have been designed from the viewpoint of the programmer rather than the end user, resulting in Magento being more complicated than sophisticated.

Here’s a few problems I’ve had with Magento already:

• There’s no way of viewing all the products in the store by manufacturer
• The urls aren’t all seo friendly, in fact filtering/layered-navigation makes very seo unfriendly urls
• New administrators are given no role by default and so can’t log in to the admin panel
• The CMS system is bollocks. There’s no parent-child relatinships for a start
• Themes/skins are far too over complicated to make and edit
• The documentation is nowhere near complete, for end users or developers

A couple of final thoughts:
There is definitely something to be said for the simplicity of Cubecart.

I think part of the major problem with these open source but commercial web applications is exactly that:- while they are open source so the community can edit and modify the source code to meet their needs, they are commercial. I think this results in much less contribution from the community and most of the contributions that are made are commercial.

The dTabs Wordpress plugin has been updated to support that major database schema changes introduced in WP 2.3, and the Kubrick Tabs Wordpress theme has been updated to the latest release of the Default theme (1.6) shipped with WP 2.3 so it now supports the new tagging features and Widgets. In addition it now supports the deactivation of dTabs, should the need ever arise

dTabs 1.0.4
Kubrick Tabs 1.4

So it’s been a while since I did any work on it. Development on the whole project was put on hold as the client decided she didn’t want an online store just yet, so the mod lay unfinished on my hard drive for a few months. This week I’ve started working for the client again and so the mod is coming along nicely. In fact, I think it’s just about finished. It needs some testing before actual release, but it’s pretty much there.

So since the last update:

• I’ve finished the stock check facility for before an item is placed in the basket, if the item is out of stock, instead of being placed in the basket the quantity is set to 0. In addition a notice is displayed upon selected an option combination which is out of stock (before the item is placed into the basket).
• I’ve done some work on the admin user interface, changed the colours and layout a bit. I’m still not completely happy with the colours but it’s hard to get enough contrast to show the different levels and sublevels (eg products, option combos, options, values).
• Tidied up the mod’s deletion behaviour so that all the necessary records are deleted when they should, eg when deleting a product off the products admin panel all that products options and combos need deleting too.
• Improved security over and above the level of the original options admin panel – so that users can really only do as their permissions allow.
• Finished the installation script and included it into the new options admin panel so you don’t have to mess around with sql files or statements to update the database – just click a button.
• Written installation instructions – simple (over-right old files) and advanced (step by step changes in code).
• Various other fixes and improvements

So all in all, the mod is looking pretty finished and almost professional (on the exterior, the code isn’t pretty in parts!). Once I’ve done some more testing, got the client using it (scheduled for a weeks time), and set up a Cubecart store to sell it with, well that’s when it will be on sale.

On the subject of updates. If I bought a mod, I would hope to get bug fixes for free, but at present I can’t see an easy way of controlling this except for maybe just making bug fixes publicly accessible, and putting any updates which include new features for sale. If you have any ideas or know of a good mod that will allow customers to download updates for free then I’d be interested in your opinion.