A Cloud for All Seasons
There has been a lot of buzz about the cloud, but up until recently it's mostly been about businesses using the cloud to smooth out fluctuating loads. This is possible because you only pay for what you use in the cloud. For instance, Green Man Gaming, where I work, is giving away three free games* this weekend, and our usage has gone up by several orders of magnitude, so we fired up a bunch more web servers to handle the extra load. That was on Friday. Tomorrow morning (Monday) we will kill them off, because the offer ends. We will only pay for the storage and processing we used during the three days. The alternative would have been to buy extra servers ourselves which would cost us capital up front, and result in them being completely idle except for a few days each year. It took a while for businesses to be convinced, and most applications need at least tweaking, if not some rewriting to run in the cloud, but businesses, especially startups, are starting to make fairly steady inroads into the cloud. The result of this is that the cloud vendors (who are all the usual suspects, Amazon, Google, and Co) are starting to look for new markets. The one they have identified is consumers with smart gadgets. As these gadgets gain more and more of their owners' vital information, why not have them backed up to storage in the cloud? And that's what all the latest razzmatazz is about. There are, though, two problems. One is the perennial problem of payment. The fact is that people don't like paying for mere services. I know we online service types are generally considered to be a dodgy bunch but it's not just that. People like something physical for there money. They will happily pay for a USB stick or an external hard drive to do back up, but wibble about paying just as much for online back up. Years ago, when I worked for the Commodore 64 UK network, Compunet, we were getting nowhere selling modems with a free subscription to our service. My then boss had a brilliant idea - sell the subscription and give away a free modem with every subscription. The cost was the same whichever way you did it. Now, however, we could not open the boxes of modems fast enough. People went away happy because they got a freebie they could get their hands round and fondle! The intangible 'subscription' had no meaning, a modem you could see and hold did. The other problem is a little more subtle. The whole premise of consumer continuous backup to the cloud relies on unlimited bandwidth. However, consumer unlimited bandwidth, especially uploading bandwidth, is exactly what the phone companies, cable carriers, and ISPs are currently fighting to eliminate, because it screws up their business model. How likely are people to use constant backup facilities when they have to pay significant amounts of hard cash to bandwidth to handle that back-up. My time with Compunet holds a lesson on this issue as well. When Compunet was running, back in the 1980s, phone calls in the UK (even local ones) were charged by the minute. I once worked out that for every penny the average user paid Compunet, they were paying the telco 5 pennies. The heavy users (mainly people using the chat and playing multi-player games) would have quarterly phone bills in excess of 600 UK pounds. Their quarterly subscription to Compunet was about 25 UK pounds. An extreme case, maybe, but it gives some indication of the problems faced by the cloud vendors. So can the vendors make money out of it? Not directly, I suspect, but they might be able to make the service fly as part of a more integrated system - the one to watch on this is Google (who else!). We shall see, only time will tell. http://pogue.blogs.nytimes.com/2011/06/16/excited-about-the-cloud-get-ready-for-capped-data-plans/
|
If you have any questions or comments about the articles on my web site, click here to send me email. |