Web Services - Thoughts

[Richard Conyard]

I am thinking about expending some time in the next revision of our software working on splitting certain functionality out of the system and making it available as web services (because of how we have written the software this is relatively straight forward), but before I go too far down the line I'm wondering if there is any interest and at what price threshold.

Basically in non-technical terms we've got a lot of modules for our content management system, some of these are:
News Articles
Events
Jobs
Newsletters
Polls
Categorised Links
Product Directory
Shopping Basket
Transaction Logging
Auction & Promotional Codes
Courses
Document management
Forums
And what I'm proposing is allowing people to drop these directly onto their site, regardless of where it's hosted or what programming technologies are available on their server (as long as there are some). The admin systems and databases would all be hosted by us.

Would you be interested, or do you think people would be interested? What would you be willing to pay either monthly or annually to have these on your site? Or for web developers out there would you be interested in resell, if so at what cost with what margin?

Technically we're very close to be being able to do this anyway, the next upgrade will take us closer; but I'm wondering whether it's worth taking the extra step and expending the energies with marketing etc.

Any feedback appreciated.

- Edit -
Not interested, don't know what the heck you're going on about are valid responses as well!

[TWD-Tony]

I'm not 100% sure how you are thinking of implementing this - maybe something like a RSS feed from your site into the clients site? With the admin done on your server and output fed to the client website...????

A few problems spring to mind... Customisation is one - if a client has the ability to FULLY customise the content to suit then okay, otherwise the "feed" would look odd. Security is another... IF a client was running a site that was running SSL - the fact that some of the site content was coming from outside the domain would throw up security warnings - not good :(

I'm not trying to put a dampner on your idea - I'm just curious on how this would work and how you could get around certain issues :lol:

Tony

[Richard Conyard]

Actually would be pretty straight forward.

We run standard XML formats for all data returns at present anyway, so we'd wrap these in a SOAP service and provide a piece of code that could be placed on your site to bring back the data.

Since the data is XML we also have corresponding XSL files which would be make available if you wanted to change the structure, or the XSL uses CSS for styling which could be changed.

So if you change nothing you get the standard look and feel which depending on how "designer" your site was may look a little out of place.

In regards to SSL we're talking about doing this server side.

[JoyDivision]

I have good experience with XML Soap web services with C#.NET but none with PHP (yet). However its somthing I need to get into and I am always on the look out for web services I could cosnume to add benefit to my sites.

I am assuming the concept with PHP4 is pretty much the same, I just hope the lack of real object oriented support in PHP4 dosn't make it too hard. ASP.NET is not really a viable option for me at the moment due to hosting costs.

I am not sure quite how any of those would benefit a web developers website or a computing network installation companies.

What does the document management web service do exactly?

[Jayne]

:lol: too techy

Jayne

[creospace]

They will probably work, I have some similar products although im sure not to the same exacting standards as yours :)

In theory i'm sure it will be great it's just how you market them etc.

Gary

[Richard Conyard]

Technically I know they will work. There will be a few fun and games with things like baskets etc. in regards of where to hold the basket data (probably on our servers), but these can be overcome.

It is the marketing which is the worrying part though. I know this is our weakest area. If the concept doesn't sell itself then it makes it tougher.

[Richard Conyard]

Thinking about a non-techy way of putting this... here's what I've come up with:
News Articles
Events
Jobs
Newsletters
Polls
Categorised Links

On your site, updatable by you £30 / month - no need to change hosts or get a re-design.
--
Obiviously pricing isn't set, and the wording isn't great - but it does put it in a non-techy framework.

[JoyDivision]

The problem is at the moment its just a product, you need to convince us why it would benefit it us :D*

*I have been reading too many marketing books :p:

[Richard Conyard]

No at present it's vapourware and needs to be written properly.

What I'm wondering is whether it's worth taking the extra step of development, and the extra mile of marketing to turn it into a product.

[DuaneJackson]

As a web developer, if you were able to provide a complete set of templates in various languages for taking the information and displaying then this could be of interest. It'd be effectively reselling the system but keeping the code and data under your own control.

The main selling point that springs to mind for the end-user is what you've already identified - they can use thinkcolony features without the hassle of changing hosts.

It's attractive, certainly. But I wouldn't like the job of having to market it to non-technical end users.

I've got to look into creating an API/Web service next year for KashFlow. Not looking forward to it!

[Richard Conyard]

Well that would be the golden goose - if we could bring in the entire Colony feature set without people having to move hosts that would be fantastic.

To be honest the front end is either light enough, or modular enough to handle most of it.. hmmm interesting.

Duane, on a completely non-commercial level if you want to have a chat, send over some ideas or have a second seasoned developer eye over your ideas for moving into more web services then PM me as and when, I'm more than happy to take half hour out to come back with ideas / suggestions.

[Enigma121]

Not wanting to burst your bubble here Richard, but most of the services you are offering here are already available to your target audience (site developers) in GPL libraries within Sourceforge.

I can't see a huge market for it, as there are free alternatives available. Web services aren't exactly "noddy" HTML, so if you find a client who knows enough to use them (or indeed what they are), chances are they will also know enough to use the GPL code too.

The financial model for these web services isn't clear to me yet. In a way you are hosting, so presumably some of the fee covers this. I can see the model working well for say phone number or address lookup services, but you are talking about less well scoped / bounded functionality.

To sell any module on it's own you'll need a well documented API for other site developers to make use of.

BTW if I'm totally wrong and things do take off, we could assist with a java port / JSP integration option.

[Richard Conyard]

No you're right, there is a lot of this functionality over at source forge.

I guess the one part of it is that very rarely are they all batched together, which would be a USP in part. As I said originally these are just thoughts and we're talking very early doors.

At present I've enough to keep my devs and designers busy dealing with clients that are going for the current cms until spring next year, but it's always good to think about new ideas.

[MichaelG]

Hi Richard.

We developed a web services module for our CMS about 18 months ago - wonderful concept. The commercial reason behind it was to ...... umm should I be giving this away to a competitor? What do u think Richard?

[Richard Conyard]

We already have a web service module for our CMS for consuming services, and we're split out on the downstream to deliver web services pretty much ( http://www.thinkcolony.com/whatiscolony/technical_details.asp ).

As for what do I think, well I'm more than happy for you to share ;-)