Jakob Nielsen’s latest article is a must read. It focuses on usability from a scientific and advocacy approach. On the scientific side, he aptly puts it that usability is not about formulas but rather recommendations. It is scientific that all recommendations need to be put to the test to see if it works – hence end user testing.
Nothing infuriates me more than hearing ‘Lack of Knowledge Management Stories from Hell’, and a few prominent businesses feature alot in here – their employees don’t want me to mention names.
Now I’m going to take a guess that these ‘managers’ know f*ckall of knowledge management, because frankly if they did, these problems/issues would not arise in their working environments.
I guess they won’t know how to spell “knowledgement management”, so I’ll have to give these damn fools a few lessons. To make it easy, we’ll show them the mirror so they can see how disastrous they really are:
Lesson one is a bit difficult because… um.. these people refuse to be taught. They are ‘in charge’. They ‘make the rules’.
Rules: this is the stupid thing about ‘their rules’. So long as you do a job and the work gets done, there are no rules, no policies, no guidelines, no roles, no privileges. Rules, thats a privileged word only ‘managers’ can use.
Managers: I’m reluctant to question their suitability, because as we well know, the record would be pathetically depressing. Little wonder, we have the brainless competition – worst candidate gets the job. If they don’t get their way by abusing their authority they resort to trantrums – shouting and the lot.
Tasks: This is what we do everyday, don’t ask why we do it, don’t suggest ways to improve it. Don’t be surprised if the task doesn’t make sense or is counter-productive. If there’s a deadline, they’ll let you know on the last minute, because if they did so earlier, it would be a productive decision.
And if a ‘manager’ ‘abuses his/her authority and gives a counter productive task, just do it, don’t question it, they cannot be taught. Instead they’ll give you a ‘rule’.
If you recognize all four of the above listed items in your workplace, please see a psychologist. Its pathetic to see this as the norm in workplaces, particularly corporations. It is atrocious that in 2005, these companies know absolutely nothing about knowledge management.
Knowledge management brings two benefits to organisations. Firstly, there is the documented processes and procedures that supports the growth and development of the organisation AND its staff which results in a competitive edge for them.
Secondly, it is the sanity that it brings, the pleasant work environment, and this also helps productivity and the competitive edge. People feel valued, the contributions are appreciated, their input is welcomed. There is ‘team leaders’ not ‘managers’.
Whats your favourite e-commerce store? Amazon, Barnes and Noble, or perhaps Kalahari.net? Mine is Loot.co.za. Today I got my copy of Don’t Make Me Think by Steve Krug, and I’m pretty impressed by the book and the experience.
Loot.co.za does many things right. The adding to basket, good search results, security, lots of payment options and lots of delivery options. I tend to pay by bank deposit (prefer this method – and I don’t have a credit card).
In the past I ordered items by Kalahari.net, but doubt I will in the future – for one reason only. KALAHARI.NET – YOUR PRICES ARE ATROCIOUS AND RIDICULOUS!!! I’m hoping to do the PHP Zend Examination. Loot.co.za wants R260 for the study guide. Kalahari.net wants R400!!!. It is absurd to contemplate how they determine their prices – a real f*cking rip-off!!
Where can Loot.co.za improve? Two areas:
Firstly, if you pay via bank deposit, they tell you that you have two days to pay the money in the bank. I think this is a bit imposing, even though probably not enforced. The only time possible for me to pay is over weekends – so a bit consideration here please.
Secondly, some of the estimated delivery dates seem to be too long. E.g. some books from SitePoint are set as backorders even though they were recently published. Others take three weeks. Then again I might be biased here – always looking for cheap prices for SitePoint books.
Nonetheless, I’ll recommend Loot.co.za anyday. Perhaps you’d care to visit to just view my wishlist.
This week we had some end-user / usability testing on the KEWL.Nextgen E-learning system. And frankly, it was much better organised and prepared for than the previous.
Our testing plan was to give users a series of tasks to complete over three sessions. The tasks were structured to give a flow to them, and there were positive results.
We only had five users out of forty invited, but this is this recommended amount (lucky us).
The test was a roaring success in terms of the feedback it gave. The feedback was not positive most of the time, but it showed flaws in the interfaces we didn’t consider.
The users appreciated the test plan. It was focused, they had tasks to complete. Sometimes testing is meaningless as users lack desire or enthusiasm. Here they were given a map, and we were observing how them implemented it.
Another positive experience is that the group were required to test the instant messaging and email as part of the test plan. This ended up being used for more than that. A result of having a group testing, not just one or two people.
The major problems could be listed as follow:
Titles - pages that didn’t have titles created confusion as to the purpose of the page.
Labels - this is related to information architecture, but wrong labels gave false impressions. Particularly, users jumped steps because the labels gave the impression of more than one task.
Validation - particularly on the web, this is important, else lots of useless records are added to the database.
Confirmation - when confirmation of actions were poor on non-existent, users were left in confusion on what to do next.
There are certain misconceptions, which quite frankly, are incomprehensible to have with regards to end-user testing. They are not only meaningless but also counter-productive.
End-user testing is about ‘playing’ with modules. On the contrary, end-user testing has to test tasks and activities. If the testing is not focussed, it is a waste. Users feel they don’t have direction.
End-user testing is about getting user feedback. This is true but “it depends” (quoting Jared Spool). The best feedback is the observation of users, not what they say or put on paper.
“… The matter is certainly serious enough to warrant investigation by the US Securities and Exchange Commission and maybe even the de-listing of Telkom’s shares from the NYSE. It also has potentially very serious (and seriously embarrassing) implications for SBC Communications in the US …”
And yet none of the flippin newspapers in South Africa are prepared to cover it!
In this month’s issue of al-Qalam (June 2005), Naeem Jeenah hits the nail on its head with an article on Muslim complacency with crime. He highlights how the Schabir Shaikh and Nelson Mandela art scandal should be an embarassment for Islam and the community, yet it seems we are hiding it as an isolated incident of one of our ‘brothers’.
Naeem goes on to further deplore this mentality when it comes to spouse-abuse, ill-treating workers, tax-invasion, corrupt business practices. We know who these people are, yet we proceed as far as regularly embracing them, as though nothing is happening.
He goes on that it is high time we follow an isolation policy. This last point is already controversial, but lets see how the moulanas and imams react (to protect their personal interests). They are going to proclaim such an attitude haraam (forbidden), but won’t provide an alternative. `Ird-ul-Muslim – violating the honour of a believer is forbidden.
Yet it is ironic and hypocritical that these ‘wannabe victims’ are first-class perpetrators of violations.
One of the great frustrations for the PAGAD movement was that families of drug merchants were given protection to enter Masaajid. These people knew their ‘brothers’ were doing wrong, destroying lives, yet enjoyed the ‘privilege’ of the Imam’s protection to perform Ibadah (worship) – again a hypocrisy.
I’m going to stop my vent for a few reasons, but to conclude that I agree with Naeem on his points. We should not be blinded by criminals and their religious bodyguards. Our eyes need to be opened by Islam to a life of righteousness and morality, encouraging the good, and forbidding the bad.
“Do not get angry. Do not get angry. Do not get angry.” – Prophet Muhammad (PBUH)
“Help your brother whether he is oppressed or an oppressor. Help your oppressing brother by stopping him from oppressing others.” – Prophet Muhammad (PBUH)
* It also makes me appreciate why in Islamic law, if a person has to pay blood-money in Islam, the burden falls on his entire family, not just the criminal. Clearly we are suffering when Islamic law is not implemented.
After weeks of trying, I’ve finally managed to complete the audio for Wahied Kannemeyer’s website (www.wahied.com). The month of May was kinda busy for me, but its a nice feeling to see projects being rounded of.
Wahied’s website represents a success for various reasons, and re-inforces much of the ideas I have on web development:
Content first. Probably much to their annoyance at first, I chose to work on the content. But once this was completed, the rest was a breeze. Furthermore, we very perculiar to have free flowing content, works carefully chosen, and no marketese.
CSS layout. The website uses CSS layouts, not tables. Laid out so the content loads first.
CSS Effects. I loved the position=fixed effect, and used this here. For Internet Explorer which doesn’t support it, it uses absolute positioning, so it degrades well. Here again, we had to plan the CSS carefully.
Templates. First time I used it properly, and it works well. The client can change text, without worrying about affecting the layout.
Web audio. The web needs to be taken to its potential, and this just adds to the value of the website.
Search Engine: We haven’t done much here, but Google lists the site as #1 for a search on ‘Wahied’. Perhaps, its the relation between the domain name, good content and layout that puts content first?
Some of the ‘ other lessons’:
The client insisted on transliteration which affects search engine optimization, or does it? In any case, we tried to compensate by having the page titles without transliteration.
Internet Explorer is a nightmare. The CSS and PNG problem is deeply annoying.
PNG whilst better, does sometimes have a big file size – worth it?
Usability in free open source software (FOSS) sucks!! This is a generalisation – but more often true than not. And in trying to understand why, people have come up with the term ‘GUI wrapper mentality’:
Ask any Linux enthusiast what makes Linux great: They’ll tell you it’s the kernel, the engine, the framework.
There is way too much emphasis on the ‘engine’, that development does not focus on the user. It focusses so much on the ‘engine’ that the resulting interfaces are extremely shoddy to say the least. It seems that the engine just had to get some type of interface put together to give it a ‘nice’ look – hence the term ‘GUI wrapper mentality’.
Following a usability approach focuses on the user instead. Its the user interaction that directs development, not the engine.
Some see this as a case of what came first, chicken or egg? Who can one think of development, they ask, if you don’t have an engine or framework to consider first.
The answer is easy!
Planning a software should be done regardless of the platform or language. This isn’t always easy, as sometimes we already have an idea of what script, etc. we tend to use. Knowing the limitations of languages is important, but more important should be considering the limitations for user interfaces.
Furthermore, in many programs, particularly the object-oriented, e.g. Visual Basic and Java, you first program the interface, and then add the action listeners.
Finally, to quote Linux guru Eric Raymond:
The biggest obstacle between open-source software and world domination is not Microsoft, it’s our own endemic cluelessness about how to design software that won’t make nontechnical users run screaming… A glitzy GUI interface fails to compensate for some astonishing blunders. There are lessons here for other projects.
There is a neat attribute first introduced with Internet Explorer to programmatically turn off auto completion (complete the entry as you type) for web-based forms. Here’s a screenshot of auto-completion which comes in quite handy at times.
However, there are times when you would like this feature to be turned off. An example is when you are busy with user administration. You usually have:
<input name=”username” type=”text”>
This will often try to autocomplete usernames you have already added, or even your own username. To turn this off, simply add an additional attribute to the form.
This is browser specific and works for Internet Explorer and Firefox but worth the inclusion as a W3C standard.
The above is input specific, but you could make the whole form as such by adding it to the form tag.
Firefox, however, does have an irritating bug. If you have that item saved as a username / password match it will default the values to these username and password values WITHOUT auto completion. Highly irritating.
The alternative is to rename the name of your input box.
Mozilla also has plans to identify text close to it, even if the name of the text input box isn’t the same.