I've not really touched web development for a good few years now but have been keeping my toes dipped into various fields of interest as far as computing's concerned for a while. Having left a computer science course (for now) after the 1st year I'm looking to go down the self-taught route to gain some experience and have what I believe to be a viable concept for a website that I'd love to build but don't have the complete knowledge or experience to do so. The basic concept is to create a 'wish-list' style website from which users can manage a list of every material object they long after or wish to be given. The idea came out of a widespread hatred for Christmas shopping due not to the process of buying gifts but merely having to write a list, send it out to a dozen people then realise I missed out on 1/2 the things I intended to put on it in the first place. Likewise, I like many people also find the concept of shopping for a dozen or so people and the inevitability of buying something they've already received a daunting and tiresome experience. So the solution, as explained, is to simply create a central, dynamic location for your wishlists that anyone with permission can access, comment on and 'tick-off' in real time as well as allowing yourself to update it along the way. The principle aim, however, is to make the experience quicker and simpler, and not to create some overblown social networking site for presents. My personal wish for a site such as this is so I can in fact lessen the focus on materialiasm when it comes round to birthdays and holidays, not glorify or over-complicate it. Thus I'd like to think I could just glance at someone's list, know if anyone's bought a particular gift then make a decision there and then and check it off the list without having to let everyone else involved know. Although I've only drafted the structure for the site in my head thus far, I'd like to think the following ideas would at least give some sense of how the site would work: Users Possibly using the OpenID API to avoid making users sign-up Users select any holidays/anniversaries/birthday they wish to receive gifts for which will then tie in with a reminder-email system for their contact list Lists Each user has a central list that stores all the items on their wishlist(s) Sublists are created for whatever purpose the user wants- e.g. a birthday wishlist, wedding list etc... Sublists also allow for comments and 'checking off' purchases- features which are only viewable if you're on that particular person's contact list. Items When creating an item, the user should simply be able to add a name, direct link to a product page and rough price for the item. There may also be a link to a pricematch search for that product automatically created by the site (e.g. Froogle) As mentioned before, people on the user's contact list may comment on or check-off items. When item's are checked-off as purchased, they still remain on the list (but ghosted or highlighted) until the list creator removes them. Because only permitted contacts can comment on or check-off items, the list creator cannot see whether someone has purchased them a gift on their list. One of the main issues with the possibility of using the OpenID system for accounts is that finding existing users or adding current ones would become very difficult unless you were to manually search by each contact's name to add them (seeing as you wouldn't know which account they are using as a login)- a process which defies the concept of the website being quick and simple. So what would I like from you guys? Firstly, I would simply like advice on how you all think the site should work and what features it should and shouldn't include. I'd using a very open time-frame to build the site as I've got very little experience of modern web standards and will be picking a lot up as I go along. To that end, my second request would be for forum members who are enthused by the concept and able to help on a technical level to let me know! It'll be a good few weeks before I even start building the foundations of the site so I've got plenty of time to debate the appropriate standards/languages to use. So really, any help or advice is appreciated!
I like the linking option, and the entire idea seems like a very good one, possibly something I would use.
I wouldn't bother with OpenID (look into Facebook Connect if anything), but other than that it looks good to me. However, Amazon recently did basically the same thing with their Universal Wish List system - they give you a little JS bookmarklet and it handles the rest for you. It's not exactly the prettiest thing ever, but it did the job pretty well for me this year. I'd be happy to give some advice as-needed; I don't really have the time to write tons of code but can help out here and there. The basic concept isn't especially difficult to execute on the back-end, really just two DB tables and a couple of different queries to update them. Interface could be a bit trickier, and I'm no designer but can give some tips on creating the html/css/ajax/js/whatever once you have a design in mind.
Cool, thanks for the advice and offer! I was hoping the backend wouldn't be too difficult so it's good to be reassured. I hadn't heard of Amazon's break into the market until now- I'd checked around on the web before hand to see if something similar existed but it looked like a gap in the market. Thankfuly Amazon's service is US only for now so I might get a few users regardless- I had originally intended for it to allow you to import wishlists from other sites (such as Amazon's) but there didn't seem to be any option for devopers to do so. Style-wise I like to think I've got a reasonably good eye for design and interfaces, it's just a matter of getting the coding spot on as far as the CSS and HTML is concerned. Obviously the site doesn't need to look too flashy but it can't stay looking bare if the main content is just going to be a list. As far as the server's concerned, should I be developing the local copies using Apache and phpMyAdmin for now or are there more widely used options that I should be getting to grips with instead?
I develop all of my stuff locally on a MAMP install; I use Navicat Lite instead of PMA most of the times since the interface is better and it supports some of the new stuff in MySQL5 that phpMyAdmin doesn't (stored procedures, and much better support for views) but you probably won't need to worry about that. The only time you'll ever really come into compatibility issues is very occasionally developing under linux and then sending to a Windows server (which you'd almost never do), and issues with PHP5 vs 4, especially some servers that have both installed and do weird forwarding. Basically, you'll probably just want three tables: users, wishlists, products. Users table would have all of the typical stuff. Wishlists will have name, user_id, and description (maybe). Products would be the name, url, price, descritpion, notes, and wishlist_id. The whole thing is pretty simple - if you've got a good amount of experience coding, you could probably get it in some sort of working state in a weekend. Dunno what languages and stuff you're planning to use, I'll assume PHP and MySQL, but just make sure you know or have read up a bit on security for both. I picked up this book quite a while ago and still flip through it for tips. It doesn't cover all of the latest and greatest stuff that you can do with php5+mysql5, but nothing ever does. (As a disclaimer, I'm currently contracted by O'Reilly to do a tech review for an upcoming book, but I recommended that same book to the author of the one I'm reviewing). Dunno if you've thought about the revenue model for it - I'm sure you'll at least want to cover bandwidth/server I think adsense would cover you quite well when showing a list of products, or PriceGrabber/Froogle as you mentioned earlier.
Inserting commission links into the product links would probably work as a half decent revenue stream with enough users.
Amazon's affiliate API would work great too. It may take a bit of work to pull it off effectively, but the API is damned easy to work with and their payouts are fantastic for how much work is involved. Certainly enough that my little clipboard url to is.gd shortened link Applescript tacks on affiliate tags to Amazon links. I think that dealing with their product search API could work really well. For a cut I'll help out quite a bit with implementation
Ta very much for the offer- I'd definitely be looking to have someone else on board and am more than happy to share whatever profit the site might make. The Amazon affiliate idea sounds good- I'd like to keep the site fairly independent from tying itself down with one specific retailer but always considered product links to be the main revenue opportunity. Have added you on Live Messenger if you want to chat more (not sure if your gmail addy is compatible though).
