Google - not so foolish after all ?

There has been a rash of articles in the last few days about the announcement that Google is testing a new email service (Gmail). Google put out a press release about it on April the 1st - most of the news media reported it as a straight news item but most of the 'digerati' were convinced that it was an April Fool joke. The consensus now appears to be that it is for real. Assuming for the moment that it is true, it seems to me that it's email Jim but not as we know it.

According to the information in the link above, Google plans to launch Gmail as a free service with 1Gb (yes, gigabyte, not a typo) of storage and a different way of looking at how people access and use their email. Not too surprisingly, the idea is based on the use of search technology - rather than sorting your email into folders it all stays in one folder and you use a search interface to find the message you want to read.

For me this is a significant change, based on habits accumulated over many years, but I have known people who operate their email in this way. Rather than filing things away in folders they just leave everything in their inbox and either sort by name or date when they want to find something or use the search function in the email client to track it down. With most email clients (like Outlook as used in most corporate environments) this just leads to things getting slower and slower as they weren't designed to operate this way. When they start up they try to read all the message headers in the folder and when you have a few thousand in there it starts to struggle. If your email is stored on a server (like an Exchange server) it gives that a bit of a hammering as well.

Presumably Gmail is designed from the outset to work this way so it should be a bit more efficient and with the Google indexing and searching engine behind it it should be quick. With 1Gb of storage (to most intents and purposes effectively unlimited), as well as not filing anything, presumably you would never delete anything either.

Now, the question is why ? Google is supposed to be readying itself for an IPO, so why would they want to launch a service for which there is no charge but which will cost them a lot to deliver, even although disk space and processor power is cheaper than ever ? Well, as one commentator noted, search is a vital component of most heavy Internet users' toolkit but email still remains the killer application on the Internet.

Google has been phenomenally successful as a result of their idea of using PageRank to order search results. Without going into all the details behind it (which are available in a paper by the founders of Google for anyone who is interested) this uses the democratic nature of the web to rank the pages it finds, on the basis that if lots of sites point to a page it is probably quite important, particularly if those sites themselves are important by the same definition. Thus a link to your site from the BBC web site is worth rather a lot more than one from the Parish Council web site. So far so good, but since the PageRank algorithm is well known, people have been trying for a long time to subvert it to get higher placement. Indeed whole companies exist for this very purpose and Google themselves think that its value is decreasing over time. It is also reported that Microsoft is going after Google in an attempt to unseat it as the leader in search technology. Whatever you think of Microsoft, when they come after your market it is wise to be concerned !

One problem for search engines is that they are not very 'sticky'. People use them because they give the best results but if another one comes along which seems to be better then there is little to prevent people from switching to it and the multiplier effect is such that these changes can happen quite quickly once they get going. Google's own rise to prominence at the expense of AltaVista, Yahoo, etc. is a testament to this. Another problem is that users of search engines are anonymous, so that although Google probably has one of the largest user bases in the world it knows almost nothing about them.

I think that Gmail may in part be driven by a desire to overcome these problems. To sign up for an email account, you need to hand over some information - at the very least they will have a valid email address for you ! How much information people will be willing to hand over will depend on how useful or valuable they perceive the service to be. This helps overcome the lack of information about the users. Also, once people start using an email address it is much more of an inconvenience for them to change it and let all their contacts know about it and update their address lists, etc. So, it is much stickier than search - once captured people are less inclined to leave. For people who are interested and can be bothered, it is very easy to register your own domain name and have an email address that doesn't change as you move service providers but the proportion of people who will do this is very small. So, Google gets some information about you and can make it less likely that you will stop using their services. I don't suspect any particularly devious motive for this - I think it is just reasonable commercial self-interest.

So, will I use it - probably not, but I recognise that I am not typical of the majority of people. Many people use Hotmail, Yahoo Mail, etc. and for these people the Google offering will probably be quite attractive. If they can make the user interface as simple as the one on the search engine (although they just changed that, to less than wild acclaim) it could be an appealing option for a lot of people.