The
Ultimate
Business Machine

Technology, business
and innovation.

And, not least, about
the Mac.

Weblog Archive Cutedge

by: Bernard Teo








Creative Commons License

Copyright © 2003-2012
Bernard Teo
Some Rights Reserved.

Fri 04 Nov 2005

The Spare Room Tycoon meets the Corporate Orifice

Category : Commentary/spareRoomTycoonMeetsOrifice.txt

I'm taking a break as I try to figure out why this route I've taken is so hard. My interest is in helping people use technology to kill their competition (kill in the metaphorical sense, not as in violence, but it does reflect an intense longing for vindication).

So, why not work with corporate IT departments when they specify their "allowable platforms" - Windows, .Net, Visual Basic, or even Java? But this pre-supposes that I hadn't already spent a good deal of thought on that same choice of technologies. The difference, I believe, is that I have a context to situate that choice.

I choose tools that will allow me to build in four what I quote five months to build. That extra month is my profit or my buffer against the vagaries of my trade - the users' mis-specification of what they need or simply an increased appetite for features once they see what can be done. It would be a saintly user that will offer to pay more than what they have secured for their end of the deal and we know there are no saints in business.

In any case, I would be committing suicide if I choose a tool that will cause me to take anything more than the five months I've promised to build. But corporate IT departments don't seem to have that urgency. They seem to keep one eye on their employability in the job market rather than to do right by their company. Never mind if they themselves couldn't build anything useful in two years - that was the case in my earlier parable of the Corporate Orifice.

But why do IT departments have such a voice, even in the face of such abject failure? Because technology is complicated and the IT departments have no incentive to make it less so.

For example, insurance is a very complex subject - you insure a ship and the ship travels to many ports, they hold cargo of many items, and they do transhipments, moving the items between ships and across ports. You have to accumulate the risks and check that it doesn't bust a limit, handle items being insured in one currency, billed in another, settled in a third and consolidated as an accounting balance using a fourth. You have to calculate agents' commissions and all the discounts being offered across all the cargo items. You have to be able to do all these complex calculations and yet be able to project that same system to a freight forwarder's office, to capture as much of the data as possible at source.

The right system to use is a judicious mix of technologies - one that will give the users at the back office a friendly GUI-based application that will provide all the computational assistance they need to set up the insurance arrangements, tabbing from one field to the next. And yet offer a restricted subset of the controls to be accessed by the freight forwarders and other intermediaries over the Internet via a less rich browser-based interface.

To build everything that is required via a browser-based interface is just not possible, given the current state of the technology. We've built something that does what is required at the two ends of the spectrum using 4th Dimension - we can offer a rich client-server based application for the back office, and a browser-based interface for the intermediaries, with both sharing the same computational core in terms of the insurance-based logic. But when the IT department pronounces that it has to be Java, VB, or .Net, pronouncing the magic words with such awe and abandon, who are mere mortals to raise any questions?

So why not take the job? Because I've already had Hai Hwee do a prototype in Java and JavaScript to understand the limitations of the technology, and I know, from having built one working system, that you can't make the same thing work with those other technologies, in two years or even five. That's why there are developers who are looking to break that technical impasse and improve the speed of development, with efforts like Ajax and Ruby on Rails. The jury is still out on whether these will help, but technology keeps mutating, and it's stupid to be so dogmatic about Java, VB, .Net and anything Microsoft, even if they are huge today.

So why not take the job, still, even if we know that it won't produce a thing but we'll get two years of work and get to play with VB and .Net all this while? Because I can't see how the IT departments can continue to face the users while producing nothing. And we're confident we can find other things to do to avoid being part of that scam. We already have a system that works today, but for the IT department.

It's going to be a long war to change people's mindset, because it will take time to educate users that you can try to understand the technical issues even if you don't know the details, so long as you try to understand the concepts. If you're willing to understand issues and avoid learning things by rote, you should insist that IT people talk along those terms, instead of hiding behind the buzzwords. Make them explain what all these means, because they're often all sound and fury, signifying precisely nothing.

That's the loneliness of the Spare Room Tycoon, finding the balance in life between profit and ideals and keeping the sanity. But the journey is reward. Or is it?

Posted at 8:19AM UTC | permalink

Mac@Work
Put your Mac to Work

Sivasothi.com? Now how would you do something like that?

Weblogs. Download and start a weblog of your own.

A Mac Business Toolbox
A survey of the possibilities

A Business Scenario
How we could use Macs in businesses

VPN Enabler for Mavericks

MailServe for Mavericks

DNS Enabler for Mavericks

DNS Agent for Mavericks

WebMon for Mavericks

Luca for Mavericks

Liya for Mountain Lion & Mavericks

Postfix Enabler for Tiger and Panther

Sendmail Enabler for Jaguar

Services running on this server, a Mac Mini running Mac OS X 10.9.2 Mavericks:

  • Apache 2 Web Server
  • Postfix Mail Server
  • Dovecot IMAP Server
  • Fetchmail
  • SpamBayes Spam Filter
  • Procmail
  • BIND DNS Server
  • DNS Agent
  • WebDAV Server
  • VPN Server
  • PHP-based weblog
  • MySQL database
  • PostgreSQL database

all set up using MailServe, WebMon, DNS Enabler, DNS Agent, VPN Enabler, Liya and our SQL installers, all on Mavericks.