Scared away from osCommerce but wonder about same issues here?

[CaptainStarbuck]

I'm not new to software at all but just testing the waters with web-based eCommerce. For our first FLOSS eCommerce package I installed osCommerce. It seemed a bit "rigid" so I started looking at community contributions (modules). After a while it was apparent that most mods for osC are old, unmaintained, and there's no telling if any given mod will work with the current version of the core.

So here I packed my bags and came here - and many of the mods and themes here haven't been updated in a year or more either. eek

Why can't we get a list of mods and templates that are compliant with the current core release v1.3.8? How can we tell without looking at each individual mod whether they will work with 1.3.8 even though they were written over 1.3.7 or maybe 1.3.0? Why doesn't anyone (with any of these FLOSS packages) think to add to the mod index pages a column for "confirmed for v1.3.8"?

I understand that there is always going to be a lag from one point release to another. But I tend to look at the attrition rate of modules as a measure of how stable the software is. Do we really want to get involved with software where we load neat modules now and they can't work with the next release a year later?

Here is some info to support what I'm talking about. The Other Modules list of add-ons has 143 items as of today. Of those, only 60 of those were updated this year. 60 add-ons claim v1.3.8 as the ZC version. (The '60' number is pure coincidence since these aren't all the same as those updated this year, but of course the numbers should be close.) 51 add-ons claim v1.3.7 support.

That means there are 51 add-ons in this group, about 1/3 that may or may not work with the current release, and considering this release has been available all year, chances are slim that these modules will get pulled forward. The remaining 42 (again just a sample group) represent about another 1/3 which seem to have been completely forsaken.

So from this very unscientific experiment one might conclude that each 8 months or so we flush about 1/3 of the modules that we've added to a ZC installation. Some of these will probably get rolled into the core, but it still means there is almost a complete turnover of modules every 2-3 years. How stable is that? During this period every site owner will go through the same soul searching: "will my module work with this release?" "what can I get to replace it?" "can someone help to update add-on?"

I guess you guys are pretty happy with ZC, but I think an attrition rate of 1/3 of the software every 8 months or so is a bit brutal.

Ahem - so whaddya think? Thanks for your time.

[stevesh]

It's not my place to speak for the Zencart team about the issues you raise, but I have a few thoughts:

When I look at the listing for 'Other Modules', there is, in fact, a column that lists the Zencart version the mod works with.

Obviously, the mods are written by people in the Zencart community, and not the Zencart developers, so updating mods for new Zencart versions would be the responsibility of the author. I'm not sure it would make sense to remove a mod from Downloads just because its listing didn't mention if it was compatible with the latest version. Of course the authors should keep their contributions current. Not everything that should get done does, unfortunately.
Also, some of the confusion can, I think, be a result of simple error. There's a mod that shows as being for 1.3.5, even though the author updated the file in March of 2008. My guess is he just forgot to update the version number.

Your numbers (flush 1/3 of the modules every 8 months) assume that, just because a mod isn't certified for the current version, it won't work. Obviously that isn't true. Also, the number of mods which are listed as being for older versions doesn't necessarily have anything to do with the stability of Zencart. It may be that there isn't much interest in a lot of mods, so they're abandoned by their authors. I would argue that they should remain in Downloads in case someone does want to try them, or adapt them to the current version, as long as the version information is avaliable.

When you've worked with Zencart a while, it isn't that hard to look at the files that the mod contains, and decide if it will work with the current version. Installation on a test site and thorough testing works, too.

I don't disagree with a lot of what you've said, but I think that you're describing a 'perfect world' kind of situation which, in my experience isn't approached by any OSS.

[Website Rob]

Hi and Welcome to the Forum.

When it comes to Add-on Modules in the Download section, one must remember that someone took the time to create them and then offered them for free; so you get what you paid for. Although the number of free Mods available is a testament to the popularity of script, it is also understood that one installs them with no respect to operational condition. The more popular Mods will usually have threads for Support but that is about the extent one can expect, unless you contact the Mod creator or write code yourself for the Mod.

One thing to always remember is that it is prudent to first test any Module within a test site. This is a proper developing method for a site and gives one the opportunity to review changes before applying to a live site. Most people don't do that though or even worse, don't do regular backups of their database, and end up in a bad situation.

[Merlinpa1969]

there are also alot of modules that DONT need to be updated, they work just fine. IH2 is a good example....

just because they didnt get a new name dosnt mean that they dont work........


best way is to find one you want and do a quick search

[swguy]

When it comes to Add-on Modules in the Download section, one must remember that someone took the time to create them and then offered them for free; so you get what you paid for.

... and if you want more, you can pay more. There are all sorts of people who inhabit the Commercial Help Wanted forum, offering services in various aspects of cart maintenance. Or, if you have the skills and time, you can DIY. It's your choice.

[Kim]

Hhhhmmm, interesting point of view, basing your opinion of a piece of software on the contribs.

The Zen Cart project is very different than the other shopping cart in many ways. First and foremost, you can not even run a basic shop with osC unless you add dozens of modifications just have the most rudimentary functionality. Zen Cart can be used right out of the box for a simple shop without having to add anything except a template. (Which btw - you can't just drop into the other cart without adding a mod beforehand)

You will also find that because the code is very stable and extensible that lots of mods don't have to be updated from one release to another as has already been stated.

Anyway, welcome to Zen Cart. Hope you find what you are looking for and we wish you well.

[bigbadboy]

2 things to add.

I've said this before but a 'latest mods' section (or better RSS feed) would be v.helpful.

Also, you'll find this place a thousand times more helpful than osc forums where if you dare to ask a simple noob question you'll be hounded like you've just committed some violent crime.

[albertsavage]

Captain, I'll reaffirm what others have said here. Many of the contributions DONT need to be updated simply because the development was done so well. All of the contributions I downloaded (about 20 in total) worked regardless of the version posted. They just worked because the out of the box functionality is stable.

I ran several shops on OSC until recently. For open source e-commerce, there really is no comparison to Zen. Really.

Contributions are top notch, and only one of mine required a coding change. It's a real time saver.

Try to remember, though that Open source is a community effort, and as soon as you download OS software, you are part of that community. That includes for testing and improvement. You "pay" for the software with your time and effort, which, hopefully, is very little.

From my unscientific research, the "pay" software has far fewer features.

[CaptainStarbuck]

Thank you all for your very civil and encouraging notes.

I understand the FLOSS model entirely and the fact that we pay for our free (beer) software with some personal initiative to improving and updating the free (libre) code ourselves. In case there is confusion I'm not expecting 100% production quality code from individuals who generously provide contributions for everyone's benefit.

There are at least three kinds of community support that we see for software like this.
1) Offerings that have one extremely dedicated author providing everything - and a demanding community that asks that person to shoulder the burden for all changes. Projects like these tend to die as soon as the author burns out. It's a shame to see the FLOSS model abused like this.
2) Some packages have a lot of contributions by individuals, but once the add-ons are available they aren't maintained very well by the individuals. The number of add-ons is impressive but the fact that they're mostly one-offs means the code is tough to maintain and/or that the community in general is really more about "take" than "give".
3) Some packages have add-ons that are maintained by small groups of people who are motivated to see them improve and progress on into the next release.

I see osC and ZC more in the 2nd category, and CMS packages like Drupal in the 3rd. That's what bothers me. This community is obviously large and very cordial, but aside from the developers of the core package it seems to be more oriented toward individuals getting what they want for their personal sites rather than real collaboration in the spirit of open-source.

To respond to another comment - old add-ons shouldn't be removed for many reasons. I'm sure there are people still using the older releases and having the older code online is helpful. Keeping everything on the site gives people the opportunity to download, enhance, and perhaps re-upload updates. My point there was that for someone who just wants to get a site running as fast as possible, it's difficult to wade through a sea of add-ons that may not apply to the code package that was just downloaded. If people here know that a given add-on won't work with the current release then that add-on should be on a secondary page, and only plug'n'play modules should be available for people who can't code but just want a shopping cart. I believe there are many more of this type ("I can't code the internet, I just want to sell my stuff") than people who both sell widgets and they're familiar with how to navigate LAMP.

Similarly, as far as DIY, of course this is the OSS model and developers like those of us here know how to play the game - I just choose which game I play very carefully. Many people are going to be coming here because they want to sell widgets and not because they want to participate (now anyway) in some philanthropic project. They want (for example) a way to show more pictures of their widgets - they do not want to have to learn PHP or CSS as part of their business expense (time=money) in order to update some module so that they can display their images. Sure, there will be some people inclined to DIY. What disturbs me is that with all of the people who seem to like ZC that there aren't enough people "doing it themselves" and contributing back to the addon section. Everyone can DIY for their own little purposes, but that doesn't help the project, and we're making a heavy investment in the project when we install it online and invite people to visit our sites. I'd like to see less Do It Yourself in isolation and more Do It For Everyone.

As far as testing - full agreement - we have subdomains setup with test environments for different releases of packages that we use. I try to beat up the software in this environment before giving it the ol' QA stamp of personal approval and rolling it into production. I don't mind testing. What I do mind is that it looks like I'll wind up personally maintaining 1/3 to 2/3 of the modules I download and become dependent on, sometime within a year of going production with them. Customers love when new features are added but they get very upset when features are removed. Sure I can test old modules with new ZC core releases but for my purposes now, I need to sell widgets, not shift my average work week to maintain ZC add-ons.

One of the common comments about ZC is that it's difficult to learn and not well documented. That only increases the burden of DIY testing and code maintenance. But thanks for the note that older mods tend to integrate well with newer releases. The effort to understand the package isn't any less but the pain from module maintenance may not be as bad as I thought either.

Honestly the same comments can be applied to a lot of FLOSS out there including WordPress and various forum packages. The only reason I'm focused on it here is that we're supposed to use ZC as our customer facing store front - it's a very important piece of software, much more so than a blog or forum, so I'm being particularly careful in this decision process.

Thanks again.

[Merlinpa1969]

One of the common comments about ZC is that it's difficult to learn and not well documented

forgive me for being blunt here, but the people that are saying the above have either NOT actually tried ZC. are paid by Harald or probably should actually have a developer....
My father who is just this side of a three thumbed blind recess monkey was able to setup a ZC...

Thats who I use to test new stuff... IF he can figure it out then anyone can