— IT Services@Bristol

Please be aware that on 31 October 2014 the IT Services blog will be closed. You can keep up to date with IT Services activity and reports via our Facebook, Twitter and Google+ accounts/

Read More


This week has seen the annual BETT show being held at the ExCel Centre in London. BETT is a trade show that brings together suppliers of technology for education under one roof. I was there mainly for 2 reasons:

  • To meet with Google, along with the University of Sheffield and the University of York – to talk about their plans for 2014. Unfortunately, everything we talked about was confidential so we can’t share it yet – but there are some big features/products coming out in 2014 – keep watching!
  • To seek out some potential suppliers for the new University lecture capture project. UoB is looking to being recording lectures and making them available online and accessible. The project is at the very very early stage, but funding has been secured and I think you’ll begin to hear more about this in 2014/15.

The scale of BETT is enormous. Speaking to some American colleagues today, they think it is by far the biggest exhibition of it’s type that they go to. From the smallest one man companies selling branded paperweights, right up to the big multinationals such as Google and Microsoft – everyone is represented here and looking for your business.


BETT is quite often aimed more towards the younger side of education, and across the entire exhibiton floor were those aiming very brightly coloured and easy to hold products. From an IT Services view, it’s good to see that companies are trying to get computer skills in earlier and earlier – I saw one company that was selling programming courses for 5 year olds! The next generation of IT Services! Aiming a little higher with their Technics programmable product is Lego (yep, the very same people who make plastic blocks) – they had one of the most popular stands that I walked past:


If you’ve seen “Educating Yorkshire” on the TV, you’ll not have failed to notice that education methods and tools have completely changed since we were at school. In particular, gone are the blackboards/chalk to be replaced by electronic whiteboards in all manners of shapes/sizes/functionality. There was small ones, massive 100inch+ ones, durable ones and one designed for teenagers to kick a football at! One of the biggest names in electronic whiteboards is Smart Technologies and they had a big and very popular stand:


The University has quite a few different examples of whiteboards around the campus – get in touch with Service Desk if you’d like to see one working and how it can work in your teaching.

Also making a big splash this year is coding apps. Be it for Android, iOS, Windows Phone, Smart TV, Vtech devices, Xbox, Playstation – anything and everything seemed to have a helper app or learning environment for people to learn how to code for the various platforms. On the Microsoft stand, they were really pushing coding for the Windows platform – pushing the ease of the platform, how it can be used on the desktop etc – and, making a little joke at the same time:


To finish off, BETT has many seminars going on in various teaching ‘theatres’ around the site. Unfortunately this year I didn’t have time to go to any of them, but the subjects are quite varied and sometimes quite out there. If you can get there in 2014, I’d suggest you take an entire day as it’s a big exhibition with a big agenda – you need the day!

Mally Mclane
Communication and Collaboration Services Manager
University of Bristol

Read More

Nick Skelton, Mark Elley & Chris Mayo from IT Services attended the Future of Technology in Education event on 11th October. FOTE is a free one day conference in London organised by ULCC, University of London Computer Centre. FOTE brings together academics and IT staff to talk about IT in education, and the lineup of speakers this year was the best yet.

An element of the programme was an informal panel discussion on the subject of “how to successfully implement change”. The discussion was chaired by Yousuf Khan, Chief Information Officer for Hult International Business School. Yousuf put questions to a panel of a panel of CEOs and CIOs:

  • Adrian Ellison, Director of IT, University West London
  • Heidi Fraser-Krauss, Head of IT, University of York
  • Richard Maccabee, Director ULCC
  • Cathy Walsh, Principal and CEO of Barking & Dagenham College

Having chatted to the panelists informally over dinner the night before about their successes (and their failures!) Yousuf was well informed. Some of the panelists didn’t expect the info they’d already shared to be brought up in public on stage! So it was a frank and interesting discussion, not too cosy.

Adrian and Heidi had both led projects to introduce major changes to email in their universities – Adrian to Office 365, Heidi to Google. So I listened with great interest to compare their experiences with ours in moving to Google at Bristol.

Speaking with Chris and Mark after the event, I was pleasantly struck by how much the three of us thought we thought “we already know a lot of this stuff”. I have personally learnt a huge amount about change from the successful Google project at Bristol, and continue to draw lessons from it every week. But there’s a difference between knowing something and practising it, especially when we are all inevitably busy. So here are some of Adrian, Heidi, Richard and Cathy’s top lessons on change:

Engagement & communications is essential

Get engagement right across the board: from top to every user in the organisation. Communication is critical right the way through the change, not just at the start. Need to win the hearts & minds of the senior leadership of the institution

Engagement is critical. Where do things go wrong? Projects where you think you have the mandate to make a change, but when there are wobbles and things start to go wrong, you realise you haven’t done enough engagement up front. You need to go back and do it all again.

Everything must be easy to use

Everything we provide has to be easy. This should be the top point in the strategy. Think from the users point of view. We overestimate how interested people are in tech. People just want to do the job. They don’t particularly want to change the tools, don’t think there’s anything particularly wrong with the current ones. In the early days technology was too complicated for the average user, now it isn’t. Our thinking as IT departments hasn’t caught up.

Above all people want tools to be easy to use, and thats as true of academic researchers as anyone.

See from users point of view

The way IT see the world doesn’t necessarily match how our users see the world. The trick to success is that we need to see from their point of view.
Make sure what we offer solves the users problem. Don’t assume you know “here’s the solution, what’s the problem”. Don’t just listen to one group of stakeholders.

How to overcome resistance to change

People are resistant to change due to fear of the unknown, or complacency with how things are now, but they absolutely want to do a good job. To lead people in change you need some element of dissatisfaction with the current setup, a compelling vision of where we will get to, and an early success you can point at for reassurance.

Dissatisfaction with status quo + Vision for where to get to + Achievable first step = Overcoming the resistance to change.

Keep your head up high when things go wrong

Get the balance right in how you project yourself: be humble but be brave, be singleminded but flexible. Above all, don’t get into a bunker mentality when things start to go wrong. Keep your head up high. When there are wobbles go back and engage with people. Do everything you need to do in the good times, only more so.
You can watch recordings of this years FOTE event, including the panel discussion at Mediasite.

Read More

iStock_000008183124SmallOn 14th October we held the third and final Futures Cafe in our year long season of events exploring the future of IT. Each cafe has focused on the role of IT in university strategic objectives, this one on promoting Innovation & Enterprise. We had four speakers and a good audience of mainly IT Services staff. These are my own personal notes from a thought provoking event.

John Manley

John Manley, member of University Council and recently retired as director of HP Labs set the scene and chaired the event. John asked us to consider: What is the role of IT Services? How does IT enable innovation? And how does IT innovate itself?

Harvard Business Review published a paper “How Bell Labs Creates star Performers” (Robert Kelly & Janet Kaplan, HBR July-August 1993) which is a particular favourite of John’s. Bell labs is an IT/engineering-centric organisation – very relevant to us. The star performers weren’t brighter than the rest. The paper is available via the university library’s ejournals if you want to read it, but it comes down to two points. The people rated as star performers, by their peers and managers, had the willingness & ability to take the initiative, and were good at networking. They got on with stuff off their own bat, and they built their networks of contacts well, such that when they needed help colleagues responded quickly.

These aren’t innate skills – they can be learnt. We can all get better at them. So two questions we might ask: what can IT Services do to improve networking, and to encourage people to take initiative?

Liz Holcome

Liz is a geographer by background but now a lecturer in civil engineering. She works on better drainage & landslide prevention in the developing world, using computer models. Liz is at early stage in her career and growing her research group.

Liz made the point that if you have an intellectual advantage, you need to get innovations out there straight away, before another research group overtakes you. There are many problems in academics maintaining code for such models, eg loss of expertise due to staff turnover. Sowhat can IT Services do to help researchers get things done quicker?

Liz described a Random Hacks Of Kindness Hackathon she took part in – Liz pitched her requirement for visualisation software, 20 hours of intense teamwork from 5 people produced something 80% complete and usable. The team who took up her pitch won the event!

Liz mooted idea of something similar internally. Academics pitch their software development requirement at a public forum, IT experts sign up to develop it. Or even the Google 20% time idea – a big ask, but something in it?

Phil Langton

Phil covered the teaching and learning perspective. He’s now a teacher, not a researcher.

Phil gave us various anecdotes, often evidence based. He started by mentioning that two of his biggest research grants in the past had been started by chance tea room conversations: reading the tea leaves. Serendipity, interdisciplinarity are important. In teaching Phil gave us evidence from a study that showed if you cut back on the content of a lecture people will remember more. Phil used good teaching practice in his talk, getting us to actually do something and think about a problem, rather than just listening. Phil mentioned the importance of language – how often he realises that before he can have an engaging conversation, you have to establish that what each party means by a word is the same thing. Misunderstandings are easy and frequent – underlined by an anecdote about a A grade biology student, who thought blood vessels were like little ships – vessels…

Requirements for innovation start with a practical need, an itch to scratch. Two of Phil’s we might help with:
How do we annotate on digital documents in layers that we can show and hide?
How do we capture where the laser pointer on screen is pointing, when we capture and record the lecture, so we know what the lecturer was drawing attention to? There’s a great TED talk showing innovative use of cheap Kinect technology to do this.

Finally a quote:
Thinking is the hardest work there is, which is probably the reason why so few engage in it – Henry Ford

Paul Walk

Paul was our external speaker, from EDINA, a JISC funded project. Paul provoked us with a quote from another, unattributed institution: “I’d outsource my granny if I could!” But Paul wanted to ask us: what effect does outsourcing your granny have on your ability to innovate?

One of Paul’s observations was that the closer a developer is to the person using the service, the easier it is to have steady, incremental improvements. This can be a challenge in the new environment of Software as a Service. But APIs are becoming more and more important in this environment. Paul is a fan of APIs. He makes the point that APIs are interfaces for developers.

In a SaaS environment there are various models as to where the technical expertise and developers sit which Paul talked through. But his favourite has the concept of the Strategic Technical Developer in prime position within the organisation using the service. These are experienced people who have been around for a while, have excellent understanding of the local organisation and a good network. Just because you outsource, doesn’t mean you can do without developers!

Paul would like us to evaluate our capacity for local, technical innovation, bring engineers into an organisation at all levels, and never forget the talent and source of innovation in our student cohort.

Nick Sturge

Nick Sturge of RED is responsible for SETSquared, the University’s high tech incubator centre. He’s recently established The Engine Shed incubator space at Temple Meads. The job of these spaces is to help nurture and grow small innovative enterprises.

Setsquared is all about the people. In evaluating new businesses, Nick concentrates on the people, not the idea. He asks the founders three questions: how did you get to where you are? What have you got? Where do you want to go?

You have to be realistic when pursuing opportunities. About 40% of companies hosted by Setsquared are helped to shut down after 12 months. an interesting idea – how many university initiatives are deliberately shut down, and how many carry on indefinitely when perhaps they shouldn’t?

Speed is essential in innovation. Too much process getting in the way can be a problem. Instead build as you go, bringing your customers with you in the design. Get to the minimum viable product, 75% done, and go to market.

Innovation is both thinking up something new and getting it done. The idea on its own is no good. Nick dreamed up an idea for a portable MP3 player a long time ago in an idle moment in a lab, but it was Apple who actually built the iPod. Successful enterpreneurs cross the boundaries from research, to development, to market. This involves taking risks.

Nick pondered what are the stumbling blocks to innovation within a large organisation like the University of Bristol? One of them is reputational risk – both to the individual and to the organisation. Innovation involves risk taking, and being able to fail without taking too much of a hit. We need to be able to accept failure gracefully. Nick also pondered what role does gut instinct play? And how do you decide who to trust – so and sos gut instincts are good, but not someone elses. Part of it is building up reputational capital in advance.

We finished with a good discussion, with contributions from everyone in the room. John asked us each for something that provoked, surprised us, sparked our interest, or something we disagreed with. Many good ideas for future actions were aired, and captured during the debate – many worth a whole blog post in themselves…

Read More

National Student Survey

This week I thought I’d take a look at the National Student Survey, the results of this year’s having just appeared. Neil Davey has written an excellent report, and I’ve cut and pasted a lot of his words. The interpretation is all mine, however.

Bristol’s result for the 2013 NSS was an improvement from 2012, overall satisfaction rose by 1% to 87%, compared to an average HEI score of 85%. So, that’s all good then.

In IT Services we are interested in question 17 – ‘I have been able to access general IT resources when I needed to’. This forms part of the Learning Resources section, which also includes a question about access to library services and resources and another about access to specialist equipment, facilities or rooms.

The percentage agreeing with the statement ‘I have been able to access general IT resources when I needed to’ increased from 88% in 2012 to 89% in 2013 (hurrah!), compared to a national average of 87%. This is composed of those who are definitely satisfied (gave a ‘5’ out of 5) and those who mostly agree (‘4’). Bristol’s mean score was 4.3, the same as in the last two years.

But, what does the question mean to the average student in the street? When I was an undergraduate it would have been straightforward to answer. I could nearly always find a card punch to use, and the tray in which you placed your card deck was always readily accessible, but there could be a bit of a scrum when the operator eventually brought out the pile of print outs, so I guess I would have given it a 4 out of 5. However, in the seventies students weren’t asked their opinions and the University had only two computers. Today it’s a lot more complicated. Does ‘general IT resources’ mean the portal, email, filestore, or printing, or does it mean the PCs in study areas or teaching labs? Does it mean access from home via ResNet or broadband, or the use of one’s own laptop, tablet, or phone? If Blackboard has a funny turn on the day the survey is filled in, does the disgruntled student change their ’5′ to a ’2′ .

Interpretation is a big issue, and that makes it very difficult when we are asked to plan how to increase our score next year, because the question is so vague and wide-ranging we just don’t know what to change. Fortunately, we can look at the results of our own student surveys which give us a pretty good idea of what services students value but think we could do better. In the past, provision of computers in public areas and printing facilities came in for some stick, so we have worked on improving these.

We can also look at the breakdown of the NSS score by department.Departments scoring highest for question 17 include Computer Science (98%), Chemistry (97%), German (97%), Geography (97%) and Medicine (96%). At the other end of the scale Theology (76%) and Classics (75%) scored lowest, with mean scores of 4.0 and 3.9 respectively. Both of these departments scored very low for library resources and specialist facilities, though scored well for teaching so there may be general resourcing issues here which need to be addressed. This doesn’t appear to be simply a facilities matter as departments based in the same part of the precinct scored far better.

However, all these numbers need to be treated with some scepticism and caution. In 2012, Electrical and Electronic Engineering posted a 100% satisfaction figure (we love ‘em). But this year, they had dropped to 93% and we have been asked to comment on why this should be so. Ignoring the fact that 100% was a pretty remarkable result, and that 93% is still stonkingly good, there are some snippets hidden in the tables which show you have to be careful about dealing with such bald numbers.

ee-nssAll the results come with a confidence interval which indicates how reliable they are and is expressed in terms of a 95% chance that the real value is within the confidence interval. The graph shows the reported score in green and the confidence interval as a blue line. The interval for the Electrical and Electronic Engineering score this year was from 81% to 97%, i.e. there is a 95% chance that it lies between 81% and 97%. This is a big range! The confidence interval for last year’s 100% score was 91% to 100% (unlike The Apprentice candidates, they cannot give more than 100%). It is possible (though a low probability) that last year’s actual rating was as low as 91% and this year’s was as high as 97%, so instead of lamenting a 7% decrease we perhaps should be celebrating a 6% improvement!

In such circumstances, trying to predict or explain the NSS score is like pinning a tail on the donkey and it would be tempting to try to ignore it, but that would be an error. Although single year results may have more noise than signal, the long-term trend is something we do have to watch, especially against our competitors. It’s also important because other people, especially prospective students, think it is. Continuing to try to improve the student IT experience remains a high priority even if the NSS itself doesn’t tell us anything about what to do. All we can do is to try to spot patterns, and continue to supplement the data with our own survey results.

Read More

Things are really getting moving now on the replacement of XP and the introduction of ‘managed systems’. I think we have reached a key moment: the number of managed systems in SCCM is greater than the number of non-managed Windows 7, as you can see in the graph.


So, we now have 42% managed systems, 38% non-managed Windows 7, and just 20% XP (down from 25% XP at the end of June). From that, you might conclude that getting rid of the last 1700 or so XP systems is doable by April next year when XP goes end-of-life. Of course, it’s not that simple. As I’ve pointed out before, this only shows the systems that SCCM knows about. Estimates suggest that there could be another 2,000 XP systems out there. The trouble is we can’t be sure because they are invisible to us. My guess is that come April we will have very few XP systems recorded in SCCM, but we’ll still have around a thousand to mop up.

The idea is that managed systems will be easier to support. So, can we see an improvement as a result of having more managed systems? Well, the answer is yes and no. I looked at the tickets with subcategories ‘Install application’ and ‘Issue with current application’. Those subcategories account for around 25% of all tickets. If we compare the last three months with the same quarter last year, then there is a 14% reduction in the number of tickets. If we look across the zones, then the biggest reductions are (with one exception) in the zones with the most managed systems (D, E, F), as shown below:


The exception is Zone A, with the biggest percentage of managed systems but just a modest 2% reduction in tickets in these subcategories. I’ve no idea why, but one theory is that Zone A had a fairly well sorted out build before managed systems arrived so will have seen least improvement. Time will tell, but those reductions in D, E, and F are well worth having.


Read More

In this technical age, email messaging, telephone, online chat, wikis and other communication channels have become a fact in our daily lives.

At the University of Bristol email is still the most common channel of communication and more widely evidence suggests this is still a dominant communications channel, rather than the new social media options now available. Wired reported in January this year that in the area of online selling, “Email is Crushing Twitter, Facebook.”

Does that mean we can safely ignore social media, video collaboration and other cutting-edge forms of Communications? Not at all. The best strategy going forward is a multi-channel one, that allows you to wield the right tool at the right time, rather than stick with an outdated one during the wrong time. The University also has the challenge of communicating with a diverse audience with a variety of preferences on how they consume information.

How well such channels are used, or managed, at the University is another matter. Using the appropriate communications channel, targetting the message to relevant people and ensuring the message is clear are the critical factors.  The channel of communications is selected based on how well it matches the purpose and audience for that communication.  A case study of HMRC explains the value of various forms of communication:

  • Face-to-face conversations and oral communications for more detailed discussions to clarify issues and engage effectively.
  • Written and email communications provide clear statements that can be recorded and filed.
  • Online communications have revolutionised ways of working by providing fast, cheap and efficient ways of interacting.

In IT Services we have a range of commuication channels identified for various purposes along with a Communications policy that defines communication responsibilities and when certain channels should be used.

A challenge the University faces, and services such as IT Services, is to ensure a coordinated approach to communicating with staff and students, ensuring channels of communications are in place that suit the audience and that communications are delivered in a professional manner.

The infographic includes some facts and figures on some of the the most popular communication channels. Click to view the full-size image:


Inforgraphic: EmailTray – a Smart Email Client

Read More

portal new design screenshot

The aim of the team behind MyBristol has been to provide the best possible service to staff and students at the University of Bristol.

Unlike many systems within the University, MyBristol was built and is managed by IT Services and has a team of developers working continuously behind the scenes to develop, enhance and integrate new UoB services.

While we’ve not consciously defined this approach it does sit well within the concept of Agile Design.

Agile Design  is a reactive rather than predictive development method. The manifesto states four main values:

1. Individuals and interactions over processes and tools: While the agile method does employ various techniques, e.g. scrums, processes and tools are a means for creating more frequent interaction between team and users, not an end in themselves.

2. Working software over comprehensive documentation: While agile does require some planning, there is more of a “just do it” mentality in order to ship a Minimum Viable Product(MVP). This reduces the amount of documentation from planning to post-ship analysis.

3. Customer collaboration over contract negotiation: With agile, the goal is to get things done quickly. For that, users need to be engaged in frequent communication with the team, not negotiating terms.

4. Responding to change over following a plan: There are few “straight paths” in agile. Work occurs in a series of iterations so that the team, sponsors, and beta users can test various elements of the MVP as it’s developed, and change course if necessary.

On 30 July we will launch an upgraded version of MyBristol. Wipnone MyBristolhile not quite revolutionary the changes are significant in terms of look and layout and also brings the notable benefit of a mobile friendly interface for smartphones.

During the upgrade project we gathered a lot of information from staff and students about what they wanted and the launch on 30 July reflects this. However, we had a load of information (and work!) that we simply didn’t have time to implement. With many projects this would be seen as a failing but not for MyBristol.  Our ethos to continually improve and enhance services means we just have more to work with, which is great.

And it doesn’t stop there. We are using the launch to generate greater engagement with users of MyBristol and will be running some testing over the first few weeks of the launch to capture more feedback and engage staff and students in future improvements and enhancements of MyBristol.

It’s important for MyBristol to have a clear sense of purpose, a strategy to help guide these developments, otherwise it could be a bit of a mess.  Those principles revolve around:

  • Convenient access to UoB services and information

  • Personalised content

  • ‘Thin’ gateway

  • Present information for action (provide enough information so you know it’s worth going to the full application)

  • Added value, through integrated announcements, alerts and work flow

Using this as our guide we will use some techniques to inform developments.  These include:

Frequent iterations – using the goals identified and feedback by users  - we will look to implement solutions in small chunks.  We’ve also looked at developing personas that can inform the ad hoc design experience.


Constant engagement with users is essential to test and re-prioritize, allowing us to gain a much better sense of what staff and students are looking for.  We encourage continual feedback with users and will be regularly contacting a selection of users to test changes and feedback on requirements.

We hope our staff and students find the new look of MyBristol better and useful,  but if you’re not satisfied we’re listening.  But keep in mind that if something is a little rough around the edges, it’s work in progress!

Read More

XP and Managed Systems

Yet another month flies past. Most of the students have disappeared, so we will try to push on with all those things that could not be progressed until the summer. One of the key things is to get rid of as many XP systems as possible and to push on with managed systems. On both of these there is already some good news from the last quarter and Sys Ops and the zones have made great progress.

XP and MS


Three months ago, 35% of our Windows PCs were running Windows XP (according to SCCM). Today it’s down to 25%! In absolute terms we’ve gone from 3571 to 2341, a reduction of more than a third.

The percentage that count as managed systems has risen from 20% to 33%. Another way of looking at it is that 30% of our Windows 7 systems were managed systems back then, and that is now 40%.

The improvements occur across all zones, but B and C deserve a special mention for the speed with which they are catching up with the others on managed systems. Zones A, C and D did especially well at reducing the number of XPs.

There’s still a long way to go and the last XPs are likely to be the ones that are hardest to eradicate, but things are looking more rosy than they did in April.



Read More

Several people have been asking me recently about mobile devices. Chris Mayo & I gave a talk at a UCISA Bring Your Own Device event, and I gave an internal Friday morning talk for IT Service staff (with toasted crumpets – you had to be there :-)

Here’s a blog version with just some of the points from those talks. Sadly without the crumpets. But we do have a nice picture of a toaster:

Smartphones and tablets are not the same as PCs. We have to think about them differently to understand the opportunities they present.

A laptop or desktop PC is a general purpose computer. It can run any application. Tablets and smartphones supplement a PC or laptop, they do not replace it. They run many apps, but their form factor makes them more specialised devices. They are optimised for communication & information consumption – good for phone calls, short emails, reading documents. They are not good for producing long documents. So mobile devices will not replace PCs entirely.

Mobile devices are appliances (hence the toaster!) They are designed to be locked down and ‘just work’, not to require endless configuration. Apple or Google act as gatekeepers. The only applications that can run on the device are those available through an app store. The devices are designed to be secure and to be easy for consumers to use. Nothing is perfectly secure or easy, but compared to PCs they are a huge step forward.

In future the number and form factor of mobile devices will proliferate. There will be different form factors for different purposes, eg ebook readers, smartwatches, portable pico projectors, etc, and people will own multiple mobile devices. Convertible devices, eg laptops that turn into a tablet, have not been a success in the market, as they offer the worst of both worlds, not the best.

Mobile devices are personal, intimate. They are designed to be used by one person, not designed to be shared. They can be fashion items, and are marketed to consumers, not enterprise. Three quarters of adults in the UK now own a smartphone. The University can’t afford to provide uni-owned smartphones for all its staff, but that’s ok as our staff have their own already. So we embrace the trend for Bring Your Own Device.

Mobile devices are causing a generational shift in the IT industry. Since 2011 more mobile devices have been sold than PCs and laptops and this is trend is accelerating. In future (and for some users already) mobile devices will be the primary way to access IT. Unlike desktops and laptops, where Microsoft held sway for a generation, there is no sign that any one supplier will be dominant in mobile. Google and Apple are both big players and this looks set to continue, with others inc Blackberry and Microsoft fighting to be a viable third player. Lack of a single dominant player means that organisations must support more than one set of devices, standardising higher up the stack (eg with enterprise architecture).

Change like this offers exciting opportunities, and of course some risks. More in future blog posts.

(Photo of a toaster by Roger Lancefield, flickr)

Read More