Hosting uptime 99.9% – sounds good? Certainly in theory a 99.9% uptime guarantee means you should expect no more than about 9 hours a year downtime (or about three quarters of an hour in a month). Unfortunately the picture is more complicated. Firstly it’s clear that no-one can offer a cast-iron guarantee that their service will be available -  there will always be potential circumstances beyond their control. In reality the guarantee can at best mean financial compensation, and even then it is highly unlikely that compensation will be proportionate to the loss to your business.

When I asked a number of hosting providers about their guarantees they were very cagey, and usually referred to their 30 day money back guarantees or similar. Not a lot of use of course if you get a problem six months into a hosting contract. Even if they do offer compensation it is likely to be limited to a proportionate amount of the hosting cost, so if the service cost you £60 for a year the service could be down for a whole month and you may only be entitled to £5 back.  At best from a generous hosting company they may give you a full refund if they fail to meet their target (but does that mean they will carry on hosting the site for the remainder of the contract?). Of course this is at the basic end of the market where price competition is high. In the same way as Ryanair contests about compensation for air travellers when flights are cancelled, it may be unreasonable to expect large amounts of compensation for a service that only costs £5 a month.

Uptime means your website is available, yes? Well maybe it does and maybe it doesn’t. Some uptime guarantees are related to the hosting services network to the outside world rather than individual/shared web servers. Or the guarantee may be averaged across all their servers so if you happen to be hit hard you can only take comfort that others will have fared better.

As with most things in life you tend to get what you pay for, and higher rated uptime guarantees (e.g. 99.99%) are likely to cost you more for the hosting. Once hosting costs get into the hundreds of pounds a year range, it is worth tying down exactly what is meant by the guarantees and what happens if they are not met. A few questions to ask:

1. Is there planned downtime for the servers (maintenance time), if so how much is this, when does it happen, and is that included in the uptime guarantee?
2. Does the hosting company automatically compensate you if they fail to meet the guarantee, or do you have to notice and report the problem (the latter seems to be more prevalent)?
3. If a guarantee is not met, what actions will be taken to reduce/minimise the chance of future occurrences. For example will planned maintence time be reduced (and what are the potential risks of that)?
4. Is there a cap on the amount of compensation that will be paid if a guarantee is not met?
5. If a full refund is the compensation, will the hosting contract be terminated?

Perhaps the answer is some kind of insurance scheme that offers more flexibility in compensation. After all, a brochure website may be important to your business but is unlikely to lose as much money as a highly active e-commerce site, and yet the hosting costs/uptime guarantees could well be identical.

Share your views/experiences in the comments. I would love to hear of both negative and positive experiences with hosting companies, and also any suggestions you have for reducing the worry  of this aspect of running a business.

The short answer is probably yes, good design should be unobtrusive. The longer answer is that it depends on a number of things. New ideas or significant variations on existing ideas tend to be fairly obtrusive as they are unfamiliar and make us stop and think.

Examples

My favourite example of this in the world of web design is the tag cloud. The first time I saw a tag cloud it was a bit of an alien experience. I saw these different sized words and thought, well that’s interesting, is it art? Does it have a purpose? With a little bit of exploration I soon realised what it represented and how to use it, and it  became an obvious extension to more traditional forms of navigation including search. In some instances a tag cloud provides enough useful information just by itself, in terms of trending topics for example. It is very useful for a large site.

Breadcrumb navigation is another innovation that seems to have stuck. I heard about breadcrumb navigation before I saw it and I couldn’t quite get my head around it as described, but seeing it and using it makes it seem like common sense (although there are different ways of implementing breadcrumb navigation that can make a big difference to the user experience).

Unobtrusive then?

So when should design be unobtrusive? When it is an established convention or ‘norm’. As mentioned in the previous post there are a number of examples of this  and if things don’t quite work the way we expect them to then it jars us and makes have to think about what we are doing. As an analogy consider driving, we mostly do it without having to think concsiously about everything, but an unfamilair road sign would make us have to quickly process the information around us and decide a) whether we need to take action as a result and b) what that action is.

It can be a bit of a dilemma; as designers we want to push the boundaries, and innovate, and yet with a usability or accessibility hat on, we also want to make the experience as intuitive and automatic as possible. After all, a website that looks and works in exactly the same way as most other websites isn’t going to win any design awards. On the other hand, a website with lots of innovations or a completely different navigation system may be the bees knees but will it give the return on investment it needs to?

Please Comment

What do you think? Share some of your examples of good innovative design by commenting here.

Design by Committee

One of the often quoted rules of web design is that design by committee is a really bad idea. There are always a number of people involved in decisions around web design but problems arise when that number is more than a few. Projects stagnate through indecision, or web sites turn out to be bland and uninteresting because of the need to please everyone. For example a large committee given a choice between white or black will often end up going for a compromise grey.

From a web designers perspective, the ideal situation is where they are given free reign to make all the decisions themselves, but meanwhile, back in the real world, the best that can be achieved is that there is one or two people apart from the designer who can make all the decisions.

With a larger group, decisions inevitably take longer (either the group has to meet all together which depends on shared available time, or members have to communicate in other ways e.g. email). Group members all have their own perspectives and agenda for what they want from a web design project. Often this manifests itself as a requirement for every department to have their fair share of the home page real estate and a jockeying for position within that page.

From Committee to Community

Design by community is a different approach, where the decisions are based on the needs and opinions of hundreds or thousands of individuals. Although that sounds like a recipe for disaster, in practice it means that the normal means of reaching decisions has to be replaced by other ways, for example polling or focus groups. The process has to be more democratic to work, although inevitably there will be some individuals or groups who dominate to some degree.

For many projects design by community isn’t a practical option but it’s certainly used for many web applications and systems e.g.

• content management systems
• blogging platforms
• social media applications

What is also true is that aspects of design by community pop up in most web projects. Why do I say this? Because designers use established good practice all the time. There is innovation happening, but good design is generally based on existing principles that in some cases have been around for years. For example, it is very common for websites to display a logo in the top left corner of the page and for this to link back to the home page. There is no rule that says this is how things must happen, but it has developed over time and become an unwritten standard. It has developed by designers innovating and then the community feeding back. In this example the feedback has been generally positive which is why it has become the norm, however there is still debate over certain aspects e.g. whether on the home page the logo should not be a link as it would be self-referencing.

Some innovations disappear almost as quickly as they arrive, because they get the thumbs down from the community. They are either to difficult to use or implement, or inaccessible, or perhaps just ahead of their time.

Although the pace of change on the Web seems very fast, in many ways it is slowed down by this community action, which is probably a good thing. Development of the web needs both innovation and sanity checks. Corporate culture is generally quite conservative and a web design brief from a large organisation may have a long list of requirements which are effectively design decisions e.g. logo in the top left, search box top right, horizontal navigation bar. Things do change over time though and even the largest organisations are now happy to consider using things like breadcrumb navigation or tag clouds which were both radical innovations when they first appeared.

In Conclusion

So the message is, don’t ignore the community; use it to your advantage in establishing what are good practices for web design, in particular regarding usability and accessibility. And don’t be afraid to suggest this sort of approach for establishing requirements for large web projects, especially if the alternative is a twenty person committee who meet once a quarter to make the decisions.