Recently we’ve been discussing the implementation of a new VLE for a major Government client. One of the issues raised at a recent meeting was to do with password fatigue and so I suggested using OpenID.

OpenID solves the problem of password fatigue by automatically logging the user in through their Google/Yahoo/Facebook/etc account. My experiences of OpenID as a user have mainly been through Stackoverflow where I know many people have been put off by their lack of familiarity with OpenID. They fear they are giving vital usernames and passwords to the owners of the site. For example, to log into the new VLE we are putting in place, a learner may need to enter their gMail details and to the lay-user this is fundamentally counter-intuitive. I am looking for good routes around this. My thoughts so far are:

1. Should we send the user off to eg. Google to log-in there then come back to our site.
2. Should we create a Facebook app, where users seem to be happy to share their information with all-comers!
3. Should we put a plain old OpenID login on the VLE with a little explanation to reassure the user (in my experience the user will not understand or care)
4. Should we look into using Windows Cardspace

Perhaps we should implement all these options and any more we come up with, then those users who understand OpenID (and are probably more likely to have password fatigue) can use that, and other, more casual users can log-in a more standard way.

I’ll let you know how things progress and what eventual solutions we come up with and of course I welcome the view(s) of our zero to many readers.

Just a list of the current ‘vaguely’ technical podcasts I am listening to at the moment. I’m interested in podcasts on eLearning, software design, Internet technology and general computing, Project Management, Finance and Economics and Business. Really the struggle is to find canonical voices in each of these areas.

Stackoverflow: (http://blog.stackoverflow.com/?feed=podcast) was always a favourite of mine, but they don’t seem to have broadcast since April. I haven’t looked into why yet. Stackoverflow was Jeff Atwood and Joel Spolsky spouting forth about the process of developing a successful web application and bringing it to market.

eLearning Stuff: (http://feeds.feedburner.com/elearningstuff) - I’ve just started listening to this and it’s providing me with a great feel for the current state of the art with eLearning systems and politics without having to spend extra time on it (ie. Listening whilst driving/programming etc). I shall keep listening and would recommend it to you if you are in the sector.

Money Talk: (http://feeds.feedburner.com/MoneyTalkPodcast) - Is a podcast by David Kuo. David Kuo is the financial expert from Suntalk with Jon Gaunt and he has a very entertaining manner. He’s strong on financial restraint and monetary discipline.

PCPro Podcast: (http://feeds.feedburner.com/pcpropodcast) - It can be a bit silly, but it does keep me up-to-date with the latest in consumer tech which very often overlaps with my role here.

This Week In Google: (http://leo.am/podcasts/twig/) - Very helpful and really keeps me up-to-date with the politics going on between Google, Microsoft and Apple etc as well as providing loads of interesting tips on new technologies and how they can be applied.

I’d welcome any of your comments (all zero of you who ever read this blog!) on any podcasts you’ve found useful in these sectors.

We are looking to include a quick reporting tool in our WebALDAS data analysis system. What we would like is for a user to be able to select an area of interest such as numbers attending a course, then select a differentiator such as Ethnicity. The tool will then realise that the best way of displaying this data is, for example, a pie-chart, which will be immediately shown to the user. The user would ideally then be able to interact with the pie-chart in some way, using mouse-overs for more detailed information, or exploding segments in order to highlight a point, etc. I would like them also to be able to print out a formatted report ready for meetings or brochures etc. If the user wanted to look at something different such as retention rates on courses differentiated by age group then a pie-chart would not necessarily be the best display mechanism and a stacked column chart might be more appropriate. The system should recognise this and display appropriately. Then the user might want to force a different charting tool to the one automatically selected by the system.

There are some excellent web-based reporting tools available, the best of which we intend to incorporate into WebALDAS as we try to increase its penetration into Government statistics analysis using our unique blend of separating data-analysis from data-management and highly tailored and personalised reports designed specifically for high level users who are short on time.

Tool 1: amCharts

This tool looks excellent and could be just what we need. It has a reasonable range of charting tools and some excellent special features which take advantage of the inherent interactive nature of computer based reporting. Features such as scrollers and the facility to zoom in on complex data, or zoom out for a better overview, all in a very fluid and responsive environment are particularly appealing. The big downside is the requirement for Adobe Flash which may make deployment a little more complex, or more worryingly may make things tricky when trying to provide a service to tightly controlled Government systems. However, this solution does look to be the best bet in most scenarios. The pricing is very good at €85 for a single licence. None of our existing Government clients have a ban on Adobe Flash on the machines we would expect to use this on so at this stage amCharts is looking like a good option.

Tool 2: LogiXML

This tool is extremely impressive. It is very powerful and provides state-of-the-art charting and reporting features. It seems it can create any report you can think of, and many you can’t. It can also deploy its reports using AJAX based HTML providing a rich, responsive experience without the addition complexity of relying on Flash. I am very much for taking away complexity from the user, and by not relying on Flash that is one less thing that can go wrong at the user end. The big problem with LogiXML is the price. It is phenominal and we’re talking in terms of 5 figures here. Also, it’s a little bit like Microsoft Corporate Licensing in that it’s virtually impossible for a mere mortal to get their head around it. For our requirements in this situation I currently think LogiXML is too comprehensive and too expensive. If we had a specific contract willing to pay to have a significant proportion of the power offered by Logi then that would be a different matter.

Another advantage of LogiXML is that it offers both .NET and Java based versions. As we deploy most of our web applications using Apache on Unix machines the Java option means we don’t have to fire up a separate .NET server.

Tool 3: ActiveReports

I have been a keen user of ActiveReports since its very first incarnation as an alternate to the truly dreadful Crystal Reports that was bundled with Microsoft’s Visual Studio. Activereports has always been very powerful and flexible and allows you to create whatever report you need. The charting tools are now pretty comprehensive and they have a Flash based viewer as well as a Web-report viewer. I have not seen any demos of how interactive the web-based viewers are and whether they have the same kind of responsiveness as amCharts. At this stage there is no need for me to look further into Activereports for 3 reasons; (i) they are .NET based and I would rather avoid adding an additional platform to WebALDAS purely to show reports when other options are available, (ii) Flash/Web Report Viewer is required (same as amCharts) so that is an additional negative and (iii)  it costs more (the professional edition which has the web report viewer is £1057).

Tool 4: Crystal Reports

I have been through the learning curve of developing with Crystal several times. Each time I have been forced to return to it, because of some legacy app, I always have to go through the learning curve again such is its esoteric nature and unintuitive paradigm. It’s horrible and messy I’m afraid. Whenever we refer to it here we always say the name ‘Crystal’ as if we had a speech impediment such is this tool’s lack of ability to communicate. I really wonder how Crystal got the penetration it did. It’s as if Microsoft was the big kid it hung around with and developed its customer base by telling everyone it was ‘mates with Microsoft’.

Tool 5: Xcelsius

Xcelsius now called SAP Crystal Dashboard Design I think (I can never work out quite what Crystal and its derivitives are going on about) provides a powerful, user generated report designer whereby the user can create their own report from with a web browser. This would be very useful if we needed to give the user report creation facilities. However I think it is fairly well tied into the Business Objects environment and hence the horrible world of Crystal and, having spent some time researching all this, I am not convinced by the concept of giving users the ability to create their own reports as most users want just to see the reports, not to become developers themselves. The learning curve is such that they may as well pay us developers to use our years of experience and specialism to create the reports for them and integrate them into the systems ready to use.

Conclusion

I think we will try out amCharts because of its price and impressive output. There shouldn’t be too much effort involved in investigating how well we can integrate it and asking a few questions will find out if its reliance on Flash will be too much of a problem. LogiXML could be really useful, but at this stage would be overkill.

Let us know if you have any other reporting tools you would like to recommend that would fit the bill in this situation. I suppose our ideal would be amCharts using AJAX-rich HTML.

We’ve had an interesting issue with Moodle 1.6 this week which was very hard to pin down with much searching and head scratching.

One of the courses set up by a client is very, very, very long. All their resources are held in one huge course. This was the way they started doing things, and have maintained that ethos, despite being shown other ways, merely because it is now their habit. Fair enough, the customer is always right etc etc.

However, we’ve just moved that particular Moodle over to a new ‘Cloud’ server in order to take advantage of the extra power and flexibility. Unfortunately this course slowed down to an unacceptable degree. Instead of taking a few seconds to load up it was now taking 15 minutes or so. The server was much more powerful and was not reaching anything like its capacity.

We power this particular Moodle with a PostGreSQL database and the database process was maxing out. It turns out that Moodle’s text-cache in filters in the admin section was set to 1 minute. The text-cache speeds up the Moodle when there are many thousands of users and you’re using filters, however in this situation it was drastically slowing it down when there were only a few users operating huge courses.

All good fun. Apart from that, the applications moved over to the new server have transferred extremely smoothly.

In this age of information inundation you must have a dashboard of some sort communicating ‘at-a-glance’ information about your daily work. There has been a lot of hype about software dashboards and all manner of tasteless devices have been implemented, often so they end up highlighting trivial information and hiding the vital statistics.

A software dashboard for your area of business is a critical top-level view on the health of your responsibilities providing key decision-making information and giving a context for passing information up to managers or clients. A well designed system will give you an edge, an X-Factor, over the competition, which will rapidly boost your reputation, effectiveness and productivity and may be something you have been missing up until now.

Much software is developed around an ever increasing feature list and whilst developers are working all-hours to satisfy this demand, very often the research necessary to make software truly effective goes amiss. This results in self-loathing users who moan bitterly about their computer illiteracy, when really the fault is in the bad design of the software.

Over the last decade or so, countless terabytes of data have been collected as industries began to understand the importance of accurate and reliable data. But much of this data, though correct and robust, is only accessible to those with the more esoteric tools of database access and programming techniques. But many Executives, who do not have the time nor inclination to learn these skills, are consequently in thrall to the right person who can mine the information from the data warehouse.

It is worth investing to ensure data is easily available to any stakeholder who needs it; in an effective, intuitive and memorable way.

We’ve installed Windows 7 on four machines here now. One on a big Opteron server and the rest on very desperate Toshiba Portege tablet PCs. By very desperate I mean they were barely usable. From new they ran so slowly and in such a cumbersome fashion that you could have forgiven the operators for taking a big hammer to them rather than using the more standard touch-screen interface.

I have heard some commentators alleging that the project managers for Windows Vista insisted that everything should be testable automatically rather than using people to test the programming. As a result, though Vista does not even pass the ’sniff’ test (i.e. as soon as you use it you can tell it’s off) it was released as if it were a great step forward for the industry. And in some ways it was, particularly in security and robustness. But you could tell it wasn’t built with human users in mind.

Things have changed philosophically with Windows 7. The name; nothing pretentious or romantic, just a minimalist number; and that, in some ways, tells us much about the new OS. They’ve tidied it up. Made it put the user first. Cleared out that sense of bloat so prevalent in Vista. Reduced the amount of resources it uses. The tablet PCs we installed it on now have new life, they are fast, responsive and are a pleasure to use. The Opteron server goes like a rocket.

After 10 years of amazing changes in hardware and software Microsoft have finally delivered for Windows; a sound OS which has been crafted with care.

At Terrabase we now offer PRINCE2:2009 Project Management in isolation of software development, meaning we will offer management of anything from house builds to the development of an international software solution.

In our newsletter, in order to give a brief example of a PRINCE2 principle, we gave this overview of Product Based Planning. Let us know if you have any comments:

1. Identify the products your project is going to make.

For example, if you are building a house, you may want some solar panels however you could split the ‘Product’ into three smaller products (e.g. List of Solar Panel suppliers, Delivered Solar Panels, Solar Panels fitted to house) which makes it easier to plan.

2. Work out the order you need to make your products.

In our example we would create ‘List of Solar Panel Suppliers’ first, then ‘Delivered Solar Panels’ then ‘Fitted Solar Panels’

3. Then you work out the activites required to create each product.

To make the ‘List of Solar Panel Suppliers’ you may do a web search and make some phone calls.

Using this approach of ‘Product Based Planning’ you can quickly work out what you need to make, how to split larger tasks into smaller ones, what order things need to be done in and what needs to be done to complete your project.

Included in your plans you should end up with two diagrams; a Product Breakdown Structure, and a Product Flow Diagram. Search for these terms online and you’ll find plenty more about them.

To an outside observer it must look like we all do pretty much the same job; sit in-front of screens with keyboard and mouse and move bits of data around.

This probably makes us all experts in some way for knowing what we want our computers to do, and what we don’t want them to do. The trouble is we don’t necessarily have a clear way of expressing that and making it a reality.

If you’re convinced you’ve got the perfect computer solution for you, your organisation or the world you can now quickly and easily bring that idea into reality, and the first step might be to follow our suggestion of using Balsamiq Mock-ups which allows even the most technophobic person to create the key beginnings of a software solution.

It’s not our software, nor do we make any money from plugging it, but we do hope that once you’ve made draft designs for your solution you’ll give us a chance to put in a bid for building it for you. Or if it’s a fantastic concept, perhaps we can come to an arrangement for how to partner up.

You can watch a video of Balsamiq Mock-ups in action here

Balsamiq is incredibly simple to use, like a very easy and prescriptive drawing program. You’ll be up and running designing interfaces almost immediately with the difficult work being creative rather than technical.

I was listening to some podcasts recently where the contributors were talking about recruitment and how so many of us have got it wrong. The gist of the argument was that, in his company, everyone was paid the same amount for the same job. He did not allow people to negotiate better rates, or employ people merely because they sold themselves better at interview.

He was saying that so much of the interview process meant companies hired good negotiators rather than people who were good for the job being considered.

This particularly applies in software. The best programmers, for lots of reasons, seem to find human relationships far more trying than the relationships they hold through their computers, and an employer who takes someone on because they sell themselves well at interview, may not get the best programmer.

This would apply to other professionals and certainly anything technical, yet we’ve all experienced those technical companies full of smart looking young people, who can talk a good patter and use all the right sales terms, but cannot make the systems work. Whereas some truly creative individuals may be left in their bedrooms, hacking away merrily, to subsequently come up with the next Google, Facebook or Microsoft. That geek could have been yours.

As it says in the good book (well, on a T-shirt anyway); The Geek shall inherit the Earth

You’re passionate about your work! So are all your colleagues! And all your staff!

Well crafted software can take lessons from the games industry to bring out the passion lurking within you for the more mundane parts of your day. Games are designed to be instantly usable and to bring the player up to speed automatically. They give feedback to let you know you’re doing well and that you’re learning as a competent and capable human being might. They give you trophies to reassure you that you’re the best, that you do indeed ‘kick *ss’.

Well designed business software can learn many lessons from the gaming industry to enhance its productivity. Some key aspects of excellent software include:

* it understands your level of expertise and gives you feedback to let you know when you’ve improved
* as soon as you use it you’re able to take advantage of its main features and not get caught up in configuration minutiae
* it has a simple, practical and considerate process for getting novice users up to speed
* rewards (badges, credibility points, global score tables) to keep reminding people that one day everyone will realise just how good they are
* a responsive, fun-to-use and familiar user interface

An example: Stackoverflow, which we have mentioned in previous articles, use a combination of credibility points and ‘badges’ which you earn by interacting with the software. Obviously, you never need motivating(!?) but your staff, or colleagues will find this kind of feature a powerful driver.

Software systems such as these are becoming less exclusive and available for the more discerning organisation, and when you consider the amount of expensive wasted hours they can reduce, the pay-back time can be very small indeed.