Slashdot Mirror
Forbes Ventures Bold Predictions For IT, Linux
LinuxThis writes "Everyone's favorite, Daniel Lyons and other Forbes journalists have made some bold predictions about IT in 2004. Interesting quotes include 'Microsoft warms up to open source, and tries to make a buck off it', and the best, from our main man Daniel Lyons himself: 'The end of 'free'. Free didn't work for dotcom pet food stores, yet much of the rhetoric around technologies like Linux and voiceover-IP still involves this crazy notion that companies can make money by giving things away. They can't.' Even better, he suggests: 'SCO Group will settle its lawsuit against IBM. Both sides will declare victory. The Linux community will turn on IBM.' This is interesting considering his previous observations about OSS.."
Another ridiculously bad prediction in his article (unrelated to OSS but I'm sure willl be fascinating to slashdotters) is his final bold claim of "To repeat last year's prediction: "In 2004, Nintendo will have followed Sega's lead by exiting the console business." In 2004, it will." This guy obviously doesn't pay much attention to any sales numbers outside the US where Nintendo is well ahead of Microsoft in terms of console sales worldwide. In fact Nintendo was close to getting caught up with MS in North America but a shortage of Zelda bundle Game Cube's towards the late stages of the holiday season caused a slight drop in sales. He's just trying to make a lot of US corporate friendly predictions get people talking, he either doesn't believe what he is saying or is simply looking for attention. A lot of it is pretty ridiculous.
The method of forecasting by asking a few people what they think is going to happen is called the Delphi Method It is, in my opinion one of the overall weakest methods by far, and especially if the views are collected the way Forbes has done. In normal practice the initial and raw opinions are improved by feedback to the group for more refinement, which obviously has not happened in the Forbes article - hence, the almost idiotic "predictions."
And as you rightly said, these people don't have the faintest clue as to what is happening. Their job is to get paychecks by telling their clients what they want to hear ... and they will keep on telling it ... Henry Blodgett anyone ?
There are tons of other methods to do Technological Forecasting, (an article that I wrote many years ago) and I wish some more work that has more solid basis is presented for Tech Forecasting at /. We deserve better "predictions" than this ....
To see a world in a grain of sand, and then to step back and see the beach where the sand lies
I see five conditions under which the free model can work.
1) Price insensitive customers: "Free" can work as long paying customers are tolerant of paying higher prices to support the cost of providing some level of free product or service. For every "free' customer (e.g., who does not pay for the bandwidth & IT to provide the download) there must be a paying customer who is willing to make up the difference. If two companies are equal in quality and features of the product and service offered but one company is "giving it away for free", then the free company will have extra costs from offering that free product/service and have to charge higher prices. If customers are very price sensitive, they will eschew the company that offers "free" wares and pay lower prices at the non-free company.
2) Low total cost of "free": The unreimbursed cost of the free product or service must be low relative to the revenues generated by paying customers. This occurs under a combination of two subconditions. First, the marginal cost of the free product or service might be low (e.g., the modest cost of bandwidth). Second, the fraction of freeloading customers might be low (e.g., something prevents everyone for taking advantage of the free offer). The lower the marginal cost, the higher the tolerable percentage of freeloaders.
3) Customers who contribute services: The viability of free is enhanced by service contributions from customers. Thus the definition of a paying customer goes beyond money -- some customers provide valuable services in the form of code contributions, beta testing reports, and support on discussion forums. These contibuting customers provide a voluntary service in exchange for the "free" product. Although such customers do not help pay the bills, they do reduce the organization's costs (eliminating salaried programmers and helpdesk personnel) and they increase the value of the organization's offerings (thus justifying the payment of subsidies by paying customers).
4) Nonconfident customers: If customers are not confident of their choices, they may prefer the "free" model as a way of try before you buy. At some level many proprietary software companies do this by offering "free" trial versions of their software. The companies give away a time or function-limited version of the product and get paid for the full/unlimited version of the product.
5) Obligatory follow-on purchases: "Free" can also work if acceptance of the free product obligates the customer to buy additional products or services down the road. Giving away the printer in order to gain an ink cartridge customer is a good example of this. The challenge, for the provider of the "free" item, is to segment the customer population to ensure that only heavy users of ink cartridges, for example, will accept the offer of a free item (maintain a low percentage of freeloaders who take the printer but don't use it much).
These are neither mutually required, nor mutually exclusive conditions. Some combination of all 5 can ensure the viability, even the superiority, of the free model over more pay-for-what-you -get business models. I'm sure others here can think of other conditions that enhance the viability of the free model.
Two wrongs don't make a right, but three lefts do.