Information week reports on some figures from ComScore which show huge growth in US mobile access to the web.
Mobile Web Usage Doubles — Cell Phone — InformationWeek.
Anyone creating or managing a web project without considering mobile users is missing something important.
How much television do you watch each day? How much time do you spend using computers, mobile phones, and all the other devices which allow you to be in control?
The traditional broadcast TV business is in decline, even if the TV companies like to pretend that it is not. The passive, one-way, synchronous nature of television is being usurped both by active involvement and communication on one hand, and asynchronous time-shifting on the other.
If I want to feel involved with other people, I don’t need to talk about last night’s TV – I can join in with active interaction on everything from facebook and twitter to simple SMS. If I want to watch a particular program or movie I don’t wait for it to come around on a local channel and re-arrange my life to fit the TV schedules – I get a DVD, grab it from bittorrent, or watch it on youtube.
Paul Graham writes
Now would be a good time to start any company that competes with TV networks. That’s what a lot of Internet startups are, though they may not have had this as an explicit goal. People only have so many leisure hours a day, and TV is premised on such long sessions (unlike Google, which prides itself on sending users on their way quickly) that anything that takes up their time is competing with it. But in addition to such indirect competitors, I think TV companies will increasingly face direct ones.
and
opyright owners tend to focus on the aspect they see of piracy, which is the lost revenue. They therefore think what drives users to do it is the desire to get something for free. But iTunes shows that people will pay for stuff online, if you make it easy. A significant component of piracy is simply that it offers a better user experience.
A neat term for an important but under-mentioned concept.
We are all unique individuals and every situation too – at work, at home, and in the community – is unlike any other; as unique as every snowflake. Yet despite this fundamental truth which has arguably been true for all time, the world seems to be designed for the opposite: sameness.
image by CaptPiper
Diversity is all around us. Everyone has different abilities, preferences, skills, and knowledge, just as everyone has different access to technology and devices. Yet most systems are built on limiting assumptions about users and/or devices. Software designers expect people who use the software will be happy to adapt to the the way the software works. Device manufacturers try to “lock in” customers and prevent them using other devices.
The challenge in a world where a vision of convergence is near enough to reach out and touch, is to celebrate and empower this diversity, allowing every person to interact with each system in whichever way which suits. Even more, though, the real effectiveness comes as this interaction changes over time and mood, and situation.
Breakthrough software solutions, the ones which set the baseline for the future, will be the ones which adapt so well to user preference, history, and context that each interaction is as different as a snowflake.
Read more at The Snowflake Effect: The Future of Mashups and Learning (Wayne Hodgins) 2009 and the associated Word document about the snowflake effect in learning.
Following from my previous post about an agile approach to business, there are a lot of practical things which can be done for almost no cost and which have a very positive impact on the agility of a team.
A recent article from InfoQ looks at some good ways to set up a working environment to get the best from an agile approach.
In the world of software development, the hot term is “agile“. Agile is about making what’s really needed, even when things change, to ensure that the business gets the best value for the work put in.
It’s always interesting to see how this approach translates to other types of business. Shane Hastie has recently written an article for InfoQ examining how an agile approach suits the habits of successful restauranteurs.
Recently, Google has contacted site publishers to inform them of a change in the way google ads are targeted. User activity will now be recorded and categorized, and used to select and prioritize which ads are displayed.
The interesting part of the announcement for me is not that Google are doing this, it’s a natural step, but that they seem to be using some sort of fixed ontology to classify ads and characterize users. General wisdom is that this is considerably less flexible than a looser “folksonomy” approach.
Either way, it will be interesting to see how this pans out, and whether Google can manage to grow its already huge advertising income even more.
Read more at Interest-based advertising – AdSense Help.
When planning work, it’s easy to get tangled in detail and lose sight of the real goals of the organization.
A product rarely sells itself. What sells a product is the advantage it brings and the benefits it delivers to the customer. It is the benefit of the product that sells rather than the product itself. What is the advantage of the requirement you are stating, and what is the benefit it will bring the customer?
Marc McNeill has some interesting observations on how to translate this approach to gathering and prioritising requirements for software systems.
Thinking about value in terms of advantage and benefit | dancingmango.
Sumeet Moghe has put together a good list of ten aspects of leadership. Well worth reflecting on if you are in any sort of leadership role.
Guy Kawasaki says, “A players hire A players, B players hire C players.” This leads to the inexorable slide to Z players which finally leads to a Bozo explosion. What you want to do is avoid that slide by hiring A players and as Kawasaki rightly says, the A players actually hire A+ players. This in my opinion is the first step to building a high performance work culture where people want to learn from the colleagues they hire.
You can’t hire the best people and not trust them. That’s like buying a BMW, but pushing it to work everyday — you can never enjoy the benefits. Sometimes, people may frustrate you, disappoint you but as Randy Pausch said famously in his speech – “Wait long enough and people will surprise and impress you”. He said, when you are pissed off at somebody, and you’re angry at them, you just haven’t given them enough time. Just give them a little more time — and they’ll almost always impress you.
Read more at Free as in Freedom: My “Ten”ets of Leadership.
The measurement and comparison of web site and web application popularity is vital for any business model which depends on advertising for some or all of its income. Every advertiser wants to get the best value for money spent. In general, higher popularity implies higher value for advertising, although this can be offset by knowing more about visitors, and selecting appropriate advertising.
The world wide web has been around for decades. It would be nice to assume that measuring and comparing popularity of web sites is a solved problem. It most assuredly is not. With the growth of convergent applications and the spread of application usage across an increasingly diverse range of devices the situation can only become more complex.
There are several major ways to try and gain insight into web popularity. The oldest way is to examine the server logs of the systems hosting the application. Servers typically log every page request from every user. In theory, all that is needed is to determine which requests originated from which users, and the count of distinct users will give the popularity.
Unfortunately, there are several problems with this approach. The first problem is how to determine which logged requests belong to which users. Some applications use “cookies” (small nuggets of information stored by a web browser and passed back to the server on every request). However, at best a cookie only identifies a particular web browser on a particular machine – a user who visits a site from home and from work will likely have several distinct cookies. Other problems with cookies include some visitors deliberately switching off cookies, or clearing stored cookies and appearing like a new visitor. In a multiple device world, may devices simply do not support cookies.
Other approaches to correlating log entries and unique visitors, for example using IP addresses or URL parameters are even less reliable. It is also important to realise that many log entries will not be caused by real human visitors at all, but rather by “robots” such as the system Google uses to index web pages for its search engine.
The single biggest problem with relying on server logs for measuring popularity is that it is almost impossible to compare different systems. Not only will the information logged often be different, but server log files are private data. Each organization will probably be able to compare the popularity of its own web sites and applications with each other, but probably not be able to compare them with competitors or other established web sites.
As we have seen, there are significant problems with server logs for popularity measurement. This has led to a range of alternate techniques and services, each of which has its own advantages and disadvantages.
A familiar approach to advertisers is the one taken by Nielsen. Just as for TV ratings, a select group of people are issued with devices to track their internet usage, and the results are treated as statistically similar to the broader population. This has the advantage of being similar to the traditional way of rating TV, but it also has the same disadvantages of group bias and generalising from a small group to a much larger one. This approach is unique among web traffic analysis in that the company knows more than just the web activity of its sample group. Nielsen collects demographic and personal data about its test panel, and also interviews members to obtain more subjective information.
Nielsen’s reputation as a source of web rankings took a significant blow in 2007, when they arbitrarily decided to change the way of calculating overall popularity. There is also a strong argument that the statistical problems with using a small test group are magnified in the case of the web. Unlike TV where viewers choose from at most a few hundred channels, on the web there is an almost limitless number of sites to visit. For this reason, Nielsen web figures probably only make sense for the most popular few web destinations.
In an attempt to take a similar approach but to gather much broader statistics, Alexa have been busy for years encouraging web users to install monitoring software in the form of a “toolbar”. Alexa rankings suffer from some of the same problems as the Neilsen approach. Alexa generalises from a relatively small set of users, suffers potential bias toward the kind of people who will install their toolbar and, significantly, lacks data from anyone prevented from installing the Alexa software by company policy.
A third approach to tracking site usage and popularity is the one exemplified by Google Analytics. With this technique the site operator is required to place some special code in every page of the site. Whenever a visitor views a page, Google is notified, and compiles both specific and overall statistics. In general, this is considered a more accurate way of tracking and measuring overall site usage than the statistical approach taken by Nielsen and Alexa. However, it faces its own problems. Google’s embedded code relies on a specific browser feature (JavaScript) to be present and enabled and for the web browser to be able to effectively communicate with Google’s logging servers. If either of these are not working, then no activity is recorded from that user. Most significantly, though. Google Analytics shares a problem with the traditional analysis of server logs; the data gathered for any particular site is private, and unsuitable for comparison with competitors.
So, for now, there is no single best way to measure and compare web popularity. Attempts are continuing to try and solve this overall problem, but I don’t hold out hope for an answer any time soon.
As for other devices and communication methods, there is next to nothing. Some web-based techniques happen to work with a small number of mobile devices, and some TV-based techniques might work with some set-top box interfaces, but for now user and popularity tracking for convergent applications is still back in the “stone age” of reading log files.
I predict that there is a major business opportunity for anyone who can crack this problem.
For years Frank Carver has been paying attention to the strange world of convergent technology. During that time he has discussed and researched broad subject areas, come to some surprising conclusions, produced and distributed digital media, scattered ideas and opinions like sparks from a firework, and above all consulted for businesses both large and small to help develop and deploy successful systems, services, and products in this highly complex arena.
