First of all, Rinalds - You are the MAN! You did beautiful work, don't stop working on this project.

Here are few ideas what can be improved:

Frontend
1. ability to choose how many items per catalog
2. Filtering by attribute color
3. Ajax slider price filter
4. 360 degrees product view (http://jquery.vostrel.cz/reel)

Backend
1. product list grid with pictures (on rollover shows bigger image)
2. manufacturer's keywords and description (which will be seen when
visitor opens manufacturer's page)
3. filtering by color (when adding attribute, possibility to select color for it)
4. Icon for attribute
5. business rules depending on attributes (example: discount only for women dresses)
5. business rules - create mass different coupon codes with same attributes (maybe separate them by commas when entering coupon code)
6. set how long products are 'new' (example one week after upload, they have icon 'new'
7. SMS to admin when purchase is made (or sms to client, when his product is shipped)
8. Social network integration (facebook, twitter)
9. Mobile version

Also system uses a lot server's CPU... This is major bug...
Because when our shop has 20 visitors at the same time - engine works very very slow...

Another thing which would be very nice is to track project's progress, see it's updates. Sorry if i posted few same ideas from your roadmap, maybe i didn't see them

P.S. I'm a web designer, if You need any help with graphics - don't hesitate to write me. I think this cms can become LIVE and very popular :) don't stop developing it ;)

The biggest reason for doubt in Livecart is that the communication between the developer and the customer is lowsy! - Users never know whether this cart is a dinosaur or whether development is still happening - also you would do well to realize that most of us are 'holding out' with this cart in the hope that the developer is prepared to 'develop' - put simply it is not very fair to expect business to wait forever on an update and honesty is required - simple updates however small would eliminate a lot of this doubt - realize that most of us using this cart have invested heavily in time - but time is either well spent or its not - I for one am questining whether the time I put into this could be better spent with another cart

There is no need to make another non-working, or half-working, release.

Make sure the current code base is as bug free as possible. Not only in the most simple use cases, but in real environments with multiple advanced features combined.

Until new users are able to download, install and use LC without too much hazzle there is very little need for now "half-ready" features.

As you pint out in the roadmap document "interesting things" can happen with prices/taxes/discounts ... make sure it works! .... and disable any features that does not work.

Having a cart that sets random prices and/or makes customers fall asleep due to long execution is really not someting a merchant wants in his cart.

You should handle money as a safety critical resource. Imagine if this was a medical device, and it was behaving randomly ... a lost customer or a killed patient is really not much different.

LC already have among the best and most user friendly customization - albeit it works best in FireFox. But I see no real benefit in spending time and resources in developing this further sot it also works on other browsers, until other issues are fixed.

Most users have very limited technical/programming skills, so when they are to chage CSS or a template it is way above their skill level. They try something because they got a hint from someone, but they are in no way in control of what they are doing - and I have helped more than one user to restore their frontend after getting lost in frustration.

Offering the same tools on IE, Chrome, Safari ... does not really help.

I would suggest going another way. Keep those tools as is, but make some 20-30 templates available that are more modern. Take a good look at other leading carts - no I did not include OS Commerce ... ;-) ... and make similar templates targeted for different regions and industries.

Try to get a template community running.

After having worked with LC for about two years, survived a totally vanished developer during the end of a paid customization project, paid for non-existing express support, I am still here. It has been a time consuming and expensive project for us, but also very educational in many ways.

#1 should be to restore confidence among existing paying customers

I have also published a few commits, but there seems to be zero interest from R, ... is there any interest for community driven development? If there is no interest, I will simply fork off and go my own way.

Our working shop can be seen at www.electrokit.com. It is not a standard livecart due to numerous problems. :( ... but we got it working eventually.

We have lot's of bugfixes we could commit, but will not do until there is shown some interest and gratitude from R.