The problem with OSS [long]

This message was sent from: Suggestions and Proposals
http://forums.oscommerce.com/viewtopic.php?p=190560#190560
----------------------------------------------------------------

I used to work for a company who spent years building a J2EE web app development platform with many of the features *nukes have, along with store functionality. They burnt through $5m of funding without selling much and went bust last year, and I've learned many lessons from this, so I'd like to put my point of view (free world et al.).

Take a couple of loosely relevant examples from Primesense's ten top Sales & Marketing steps to business oblivion:

* Try to ensure that nobody within the organisation ever actually talks to customers, particularly the management team or the marketing department. Encourage the belief that with the amount of knowledge and experience you've got on the team, asking real customers what they want is an unnecessary luxury. You may need to make an exception for the sales-force, or people will get suspicious, but this isn't a problem as sales people very rarely communicate anything useful back into the organisation anyway

* Don't segment your customer base. Treat all your customers exactly the same - never mind how large or small they are, what problems they have, or what their value is to your business. Make sure nobody tries to measure the profitability of different customer groups, or somebody might cotton on to the fact that most of your profit comes from a limited number of customers and suggest that you focus resources on retaining and developing them. This could have serious ramifications, prolonging the life of your company by several years.

* Move everybody out of marketing and into product development. Make sure nobody's responsible for finding out whether there's any demand for what you 're developing. If anybody questions this, just tell them that 'Great products make their own markets' and 'Nobody every developed breakthrough products by asking customers what they want'. If they really press the point, put a group of your product managers into a workshop for a day and ask them to produce a market demand forecast. That should prove conclusively that there is a sizable opportunity out there, just waiting to be seized.

Now with the above points in mind, take my position - I've been trying to decide which content management/portal/appdev platform to adopt for my clients existing and prospective as I don't want to write stuff from scratch, and I don't want to learn more than one CMS inside out, but when I read threads like these sometimes I wonder if I should bother at all with any of them.

For example, I have a client who is a reseller of digital content, mainly PDFs. They need supplier logins so suppliers can upload new products, alter pricing, etc. They also have a very complex pricing structure which depends on what subscription a user or company has bought, and in the latter case they want companies to be able to manage their own accounts - i.e. add users, see reports, etc.

I thought to myself 'OSCommerce, that looks great', but then it doesn't have the complex subscriber functionality I need, and as their site is also a community site I thought it would take too much customisation to be a viable proposition. Now I'm thinking more on the lines of integrating DreamAccount from [url=http://www.dreamcost.com]DreamCost[/url] with PostNuke, XOOPS, or Xaraya (as it has mods for these) - it deals with subscriptions and I can write the pricing module myself.

But then I look at PostNuke and find out the main developers have left for Xaraya, so I download that and man is it complicated - maybe I'm just thick, but it's not easy to use (I know it's still in Beta, but can't find any user docs anywhere). I've written a module for PostNuke before, and it's not pretty, which is why I looked at the OO designed XOOPS. XOOPS looks good, but then I read about Xaraya lead developer saying all the *nukes are flawed as it's hard to make them look totally different. Then I read PostNuke's got a new Smarty based template engine coming. Then I read about security and find PHPNuke has one of those security login mods with the non-machine readable number people have to type in when they login, which is a great feature my client would love - they can't afford to be hacked. PHPNuke I threw out ages ago because I didn't like the fact it seemed to be run by one guy, but it has a lot of support and mods - that's what's important to me when developing a c
 ommercial solution - I don't want to come back in a year and find my CMS of choice is discontinued.

My point is this: Why don't all CMS/Commerce developers take a leap forward and develop a common API / Interface? Take the who's logged in blocks for example, or forums. They all have 99% exact same functionality, so why do people have to create a mod, then convert them to X amount of CMS and commerce systems? Why move in loads of different directions when we could all move forward? 

I guess from the tone of this thread that OSCommerce will be developing their own CMS functionality for OSCommerce - if not then they will be losing out on a whole load of customer opportunities. Believe it or not, people /do/ want integrated CMS and Commerce - why should people have to spend many weeks integrating diverse systems? Integrated ones will always win, which is why *nukes are so popular. Open Source is becoming much more accepted in the business world, and with the possibility of selling a huge amount of support contracts it's worth listening to the market and taking stuff like this seriously.

I am sure my clients would gladly pay for such an integrated system - if you look at the commercial equivalent you're looking at tens of thousands of pounds - even if you sold business licences at $100 each you would make a hell of a lot.

Integrate, don't disintegrate.

Steve