Some thoughts about the admin

[kelvyn]

Well, now we have a "customising the admin" forum, I thought I'd unleash with some thoughts I've been having over the last couple of days of downtime.

Before anyone (hello Tony!) gets mad and has a mild heart attack and a kneejerk reaction, please, think about what I'm posting and why I'm posting it.

These new forums are a really good push forward. Maybe we can push things forward even more? All that follows is the combined three years of hearing the same thing over and over again from store owners.."the admin sucks", "the admin sucks" and "the admin sucks".

What Zen really really needs is a decent admin system.

The first person to write a javascript or ajax admin for Zen will be a rich man - I'd certainly pay for a copy!

I mean, the way Zen is, now, is geared towards tiny stores with a handful of products. Yes, you can Easypopulate thousands of products in, but then what?
There are some very simple tools to do mass updates of names and prices, but what if you want to move 20 products between categories? At least 6 clicks per product.

And attributes are a nightmare. Once you've worked out attributes (and I don't believe anyone fully has), you are then presented with a highly unintuitive system, where all the values for all the names are presented in a tiny, tiny box which needs masses of scrolling. OK, so I've made the box 100 high, but still, if you want to apply 15 colours to a product, why does it also present all the size options, when you can't use them? And why must those options be added one at a time? Why not allow multi-select?

I rather think that the team may have lost their way a bit with the development cycle. I mean, a major point release having a major feature of it being xhtml compliant?! NO-ONE, and I mean NO-ONE cares about that. Amazon isn't compliant, Google isn't compliant.
What the store manager cares about is getting the job done so he doesn't have to employ a string of students who all leave after a couple of days because the task of adding attributes and products is so tedious.
What the customer cares about is a quick response, not a system on a go-slow because each page takes close to a thousand queries.

If Ajax was ever made for something, it's this. The team need to sit down and look at the interface of, say, Wordpress. And even if they can't bring themselves to Ajaxify it, a matrix of tickboxes wouldn't go amiss...just so I could tick 11 boxes and say "move these to this category" or "delete".
The phpmyadmin interface is a good example of what I'm talking about.
Also, take a leaf from the Wordpress plugin system, where a plugin must follow a standard in which it identifies itself, its version, and its homepage. The world of Zen plugins is "drop me an email and I'll send you a copy, mate". No uninstall routines.
There are a couple of Zen contribs authors (Tim K etc) who write contribs to a very high standard, but the rest will leave you with over-written code and a database full of crap when you remove them.

I know they do it for "free", I know they are way ahead of Oscommerce, but their development system is closed source, and by that I mean "closed to suggestions" too. Gallery have a "features vote" system whereby someone adds a wanted feature, others vote on it, and the top votes get priority. That way, things the developers didn't even know were wanted, like bulk watermarking, hidden galleries and flickr style tagging, have been incorporated. But the same basic functionality which would make Zen a winner has been asked for for 3 years now, with barely a hint of a "perhaps in the future" response.

Anyway, rant over. Thoughts?

[Merlinpa1969]

Kelvyn, no heart attak,
no hollaring,
how about you listen to what the devs have been telling you
The admin is due to be redone, however the part that the customer sees needed done first.

that is the part that makes you the money...

[Svanis]

Kelvyn, no heart attak,
no hollaring,
how about you listen to what the devs have been telling you
The admin is due to be redone, however the part that the customer sees needed done first.

that is the part that makes you the money...

As far as I can see customers been placeing orders since i started with 1.2.5 or whatever.
I do very much agree with kelvyn.
What difference does it make if customers can place orders if there aren't any products available to place in the cart?

Thank God for Easypopulate! Where's Easyattributes?

[tlyczko]

The very last paragraph of your commentary is excellent.
The developers, if they are really interested in improving ZC *for* the community, should implement something like this.
Perhaps the first two items up for voting should be:
1. Standardizing how contributions work
2. Improving how to set attributes
ZC has a lot of potential and a small development team.
I have in the past commented about Macromedia's extension model -- the code is closed-source, the method of extending is open, consistent, standardized, and quite open!!
Such a method of standardizing contributions would pique the interest of advanced programmers who enjoy coding and store owners who require some specific functionality sooner rather than later.
Hopefully 2006 will bring improvements in these areas.

Thank you, Tom

P.S. The forum font size is WAY too small, please increase it!! Not everyone on this list has 20/10 eyesight like whoever developed this forum, smile!!

[Festus]

What the store manager cares about is getting the job done so he doesn't have to employ a string of students who all leave after a couple of days because the task of adding attributes and products is so tedious.

LOL that is funny because I hired my nephew in high school to do this... but he hasn't gone away yet. I pay too much I guess.

[kuroi]

The forum font size is WAY too small, please increase it!! Not everyone on this list has 20/10 eyesight like whoever developed this forum, smile!!

If you use Firefox you can increase it yourself. Just press CTRL+. This used to be a problem on the IPB-based forum because of the fixed width, but with the new liquid layout, you are limited only by your monitor size. (For some reason you can't adjust the text size for this forum in IE, another good reason to switch to FF)

[seethrou]

For the font size in Firefox, another option may be more convenient.

In => Tools => Options => Content => Fonts & Colors => Advanced
And inside there, you can set the font you like.

In addition, of course you can set the Minimum Font Size of you like.

[kuroi]

The developers, if they are really interested in improving ZC *for* the community, should implement something like this.
Perhaps the first two items up for voting should be:
1. Standardizing how contributions work
2. Improving how to set attributes
ZC has a lot of potential and a small development team.
I have in the past commented about Macromedia's extension model -- the code is closed-source, the method of extending is open, consistent, standardized, and quite open!!
Such a method of standardizing contributions would pique the interest of advanced programmers who enjoy coding and store owners who require some specific functionality sooner rather than later.

Here's a contributor's take on what you say. I agree with the premise that the Admin is in dire need of an overhaul. Especially for those site developers who have clients needing to be let loose to maintain their own stores. I doubt the Development team disagree with this as much of the Admin seems to be leftover from OS Commerce.

One area that I felt was lacking was user access control. Although there was a contribution in this area, it was very basic and seemed to be no longer supported. So I picked it up, enhanced it, and released Admin Profiles. Now this contribution necessarily has to interact with all other contributions that a user has installed, so that should naturally make me inclined to agree with your suggestion that contributions should be standardised. But I don't.

It's not that I don't believe in standards, I do. It's just that I believe more in the energy and initiative that open source software unleases, and fear that once you start to apply the disciplines and structures that accompany professional software development cycles, you risk stiffling them. I can say without hesitation that if my contribution needed to be assessed for conformance to standards, I just wouldn't have done it. On the other hand I am happy for the Zen Cart community to decide freely for themselves whether it's a good contribution and worth using.

On the Macromedia Extension model i.e. closed source, open method ... well Admin Profiles was built on and earlier contribution, Admin Levels, and reading through the threads for other contributions, it would seem that it's not the only one to have a history like this, i.e. somebody has an idea, implements a basic contribution, gets bored wanders off; somebody else picks it up and takes it to a new level. My initial build of Admin Profiles used the core code from Admin levels, and although much of that has now been re-written, if it weren't for the open source model, Admin Profiles, and I suspect many other free contributions, would simply not exist.

[Woodymon]

Would not this post been better placed somewhere in the new "Configuring the admin" forum area?

[JollyJim]

What Zen really really needs is a decent admin system.

.....
And attributes are a nightmare. Once you've worked out attributes (and I don't believe anyone fully has), you are then presented with a highly unintuitive system, where all the values for all the names are presented in a tiny, tiny box which needs masses of scrolling. OK, so I've made the box 100 high, but still, if you want to apply 15 colours to a product, why does it also present all the size options, when you can't use them? And why must those options be added one at a time?....

IMHO Zencart already has a decent admin system. It communicates with the core code and does what it was designed to do. Sure, it is not intuitive but that would require cosmetic changes not core code changes.

The features you want regarding attributes need an external system not part of the admin code. This external system should communicate directly with the database and not the admin. This means an external database application for offline usage, which dumps the data back into the online database when ready.

I have developed such a system and it has not been easy. I now understand why Zen Dev have not bothered doing one. Quite simply, in order to cater for every store type, attribute type, admin function and plugin connectivity, the offline application development is enourmous. I really mean it - we are talking immense development time. In effect it requires a seperate development team dedicated to working just on this application so as zencart evolves, the update system for managing store inventory evolves too. I doubt Zen Dev could spare the time, so cosmetic changes and minor tweaks is what the admin system will receive. Quite right too because the front end is what sells the products (as Tony pointed out).

Now curiously, when I asked for help mapping the system I created and improving it, nobody came forward. Only one person donated to the admin help system I was developing to run concurrently alongside the offline database management system. So again, I don't blame Zen Dev for hanging back on developing the admin section.

For the record, I can upload 20,000+ items into the database with attributes and get them online within the hour. Sure there are issues and the system is not perfect, but if I tried I could probably load up 300,000 items. However, it is not ready for public release and at the rate of current development probably never will be.