If you want to try the SBS experience before you buy then you should do the following: get a server running Windows Server 2003 or whatever. Modify all of the normal administrative interfaces so that instead of doing what they normally do they completely screw the machine sideways from Wednesday to the point where you just sit there saying “I can’t believe how much that screwed the machine up”. Then write a pathetic joke of an administrative interface that doesn’t do anything that you want. Then get a bear trap, carefully lower your testicles into the thing until you hear the SNAP. Then weld the trap closed.

SBS server was designed with one goal in mind, to ensnare businesses and make escape almost impossible. Every possible escape route that doesn’t involve shelling out a pile of money to buy the server that should have been bought in the first place has been walled off.

DO NOT BUY THIS CRAP. If your requirements are so simple that you can’t administer your own AD server then you don’t need the functionality that SBS supposedly provides. If you do need these features then do yourself a favour and shell our the extra dough now instead of finding yourself completely pooched in a few years.

What do you think?

2.) I can add more features than Microsoft and also beat them on up-front cost, so my three advantages will be initial cost, no vendor lock-in and features. My one big disadvantage will be a lack of trust from small businesses.

So what concerns me is if I start hurting MS’s sales, will they just add more features and lower their up-front cost and thereby nail me because they are more trusted? They easily could.

What do you think?

I suppose I could gain trust by doing the integration on an established platform like CentOS or Ubuntu and then partnering with them, making my SBS an extension of CentOS/Ubuntu. What do you think? I would want to avoid partnering with RedHat because then we’re back to the initial cost disadvantage.

That said, it has inspired me to integrate and smooth out rough edges and create a SBS with more products than anyone else has used before. I’m talking firewall, router, monitoring systems, email, calendar, file server, database, directory services, terminal services, remote access, SSL browser-based VPN, a CRM, an ERP, etc. I’m talking dozens of apps. Scalability, security and ease of management would be top priorities.

I would create a TOTALLY FREE Small Business Server. I would sell support/consulting. This way, small IT shops such as Troy at his Christian radio station can manage dozens, even hundreds of end users for free. If he needs support, he knows he can just pick up a phone.

This won’t be easy, but I think I know how to do it. I’ll need to study a lot and learn project management and recruit some good programmers, but I think I can do it. I’m still thinking and praying about whether I should take this on (it’d be ambitious and I have a growing family), but I believe there’s a market in the next few years before SaaS really gets hot.

So, if I decide not to do this, someone else needs to take up the banner. Get inspired by looking at the usability and integration level of something like Untangle or OpenFiler or IPcop. Then start thinking bigger; add in many, many more services, as many as a small business would need. Glue it all together with a seamless web interface. Make sure it is secure and scalable. Hide many options from the admin, but still ensure that your glue scripts respect any manual changes made from the commandline. Hire tier 1 and 2 support staff. Test it with users. Start marketing it. You’ll need usability testing, programmers, infrastructure, technical support and other small business staff. Shoot, you could even run the whole business on your server, so that you A.) eat your own dog food and B.) work out any kinks that customers would see.

Integration, smoothing out the rough edges, a “kitchen sink” mentality, free to the community, more applications than any other solution. I think you’d have a winner. Linux has the potential to be a winner in this category.

I see The Var Guy’s point about Microsoft’s well thought out packages and price points. But when you stack this up against a solution that does the same thing – for free – and has none of the restrictions of the proprietary one, making it completely scalable to your company’s/situations needs, then (for me at least) it doesn’t make sense to lock myself into a vicious money-spending cycle that is eating away at my already slim margins.

This is why I feel it is important for Canonical to start getting together and marketing a ‘comparable’ package for SMB’s. All the great technology is there, it all works like a dream. But it needs to be marketed/delivered properly before a lot of proprietary minded people realise what it can do.

Not crippled?

“In fact they upgraded it from SQL 2005 Workgroup Edition to SQL 2008 Standard Edition in recognition of the fact that SMB clients need more features from their SQL.”

I had to chuckle – the proof is in the pudding.