Monday, 5 May 2014

Data Meme

Really great meme here that all programmers and data geeks will appreciate.

Friday, 18 April 2014

Ubuntu 14.04 desktop review

I have been running Linux on and off for a long time now (albeit with a brief dalliance with Apple and an even briefer one with Windows 8). I find Linux fast, stable and really easy to use. 

In the past, at various times I have used Manjaro, Zipslack, Wary Puppy, Crunchbang, Debian, Mageia, Fedora and Ubuntu. I have a preference towards the Debian forks, and since the Unity bar came out, Ubuntu has been my favourite.

Now before I get crucified by the uber-nerd brigade getting hairy chested about Arch, I like the unity bar. It's very productive. It suits my way of working. I like the way it looks, too. It inspires me. End of argument!

So when the latest version of Ubuntu came out yesterday (Trusty Tahr), I decided to upgrade right away. First impression is that very little has changed. Canonical have ambitions to deliver smart phones and greater integration of Ubuntu across all platforms. So I have lowered my short term expectations, on the expectation of much more later on :-)

The first thing you really notice is they have upped the design quality. Everything is brighter, sharper and a little - dare I say it - slicker. Unity appears to be delivering search results quicker, and the controversial Amazon search is now an opt-in function, rather than automatically on. Here are some more features I have noticed.
  • The lock screen now has the same look and feel as the rest of the unity interface, which is a welcome addition that looks great.
  • Sound volume can be boosted much higher - a useful feature if you are plugging your laptop into different sound systems.
  • Application menus can be put in the window, on the dock or in the HUD menu in unity.
  • Windows can be resized in real time without that annoying box.
  • There is increased graphic support, including NVIDIA, Optimus and HiDP. 
  •  The unity icons can be made much smaller than before.
  • Windows do not have any borders.
  • The guest session gives you a warning about being a temporary session on opening.

"Cut to the chase, Rich. Is it worth upgrading?"

My mother's laptop is still quite happily running Ubuntu 12.04, which will be supported until 2017. There are no big changes to functionality with the 'Trusty Tahr'... no 'must-have' features. Ubuntu has chosen a path of evolution, rather than revolution. Certainly, it is an improvement from the previous releases. But I'm enthusiastic about 15.04, which I expect big things from. I would count 14.04 as being a bit like Apple's change from Leopard to Snow Leopard. It's a small change, waiting on the anticipation of big things to come.

Saturday, 12 April 2014

How I gave Windows 8 a chance

As my Apple Mac recently gave up the ghost (a combination of some very heavy use on my part, and diminishing returns on hardware performance), have bought a new desktop. It's a very nice Acer, that I know I'm going to be happy with!!!

Interestingly enough, it came packaged with the latest version of windows - the much maligned Windows 8.

Now I'm a fair man. I'll give anyone a chance. I got annoyed with Windows XP due to virus and reliability issues, so I moved to Apple, and I was happy for a while. I then kicked out IOS Lion and replaced it with the Linux-based operating system called Ubuntu. Since XP, Microsoft has released Windows 7, which I currently use at work, and I'm quite happy with that. So I decided to give Windows 8 the benefit of the doubt.

So here are my observations:

The whole experience of running windows has been significantly simplified. To me, windows 8 is a reaction to Apple. They are trying to achieve the same level of gloss and slickness that the guys at Cupertino are so good at. And generally, they achieve it. But there is still quite a lag. It's smooth, but not fast.

The metro bar has clearly been designed for the new generation of tablet computers. But I'm afraid it's just not good enough for power users who want to be really slick between applications on a desktop. There are too many times where the UI just doesn't give you the opportunity to see how you can do something straight away. I became quickly frustrated with it, and within 45 minutes had decided to get the latest linux distro and wipe windows 8 altogether. 

I'm sure windows 8 is excellent for the up and coming generation of tablet computers, but I find it clumsy and resource hungry on a desktop computer. It made a brand new computer run slower than my 8 year old Apple Mac ever did. And that's just not good enough for an experienced user who wants a dense and powerful information experience.

So I'm back to Linux, which positively flies on this computer with plenty of power to spare. As the support for XP has been withdrawn, why not look into Linux as a replacement for XP? Get your requirements and head to the link below to see if you can find a distribution that is good for you. My suggestions are either Zorin, Ubuntu or Mint.

Find a list of all linux distributions here.

Saturday, 22 March 2014

6 tips for great project management

Project managers are made, not born. No-one looks at their children and says, "She's going to make a great project manager when she leaves school." 

In my time, I have worked with some really good project managers (and some bad ones, too). I have also managed the occasional project, myself. So here are my casual observations about what makes a good PM.

1. Practice your ability to assess and re-assess priorities
I have watched as huge projects veer totally off course, due to poor reaction to changing circumstances and managers blindly sticking to obsolete priorities despite spiralling costs and no progress. Learn a methodology for prioritisation, and learn to spot the tell-tale signs that the goalposts are changing. If it's a large project, build milestones in your project plan to check  prioritisation.

2. Questioning skills
Learn to ask questions in an open, constructive and non-judgemental way. Then shut up and listen. If you don't believe what they are saying, do not let that show in your behaviours. Building trust is important, and holding your own council will help you to achieve that.

3. Use information honestly
You have heard the phrase, "knowledge is power." Be open and honest with your sharing of information in all of your dealings. Knowledge is not a weapon. It is not a way to tell everyone how clever you are, either. People are smart enough to know when this is happening, and will not want to help you. Use plain English and use technical terms and acronyms only when necessary.

4. Plan and schedule your communication appropriately
During complex projects, people want to make sure that everything is communicated. They may spend some time planning what they are going to say, and how they are going to present to you. So be consistent. Set a schedule for communications and stick to it. Don't keep badgering people for unscheduled updates. 

5. Keep an eye on progress
If you can find a way of checking actual progress yourself, it is useful to do so. If a document needs to be sent, make sure you are copied into it. Look for evidence that tasks are being carried out. The buck stops with you!

6. Be confident, mature and friendly
Everyone on the project looks to you for inspiration. So inspire a sense of confidence in people. Re frame the work as an exciting challenge, and problems as being interesting. Allow people to share their concerns, and give appropriate emotional support. You will get much more out of people when you are on their side.

Saturday, 22 February 2014

How to justify your business intelligence spend

In these recent times of austerity, finding funding for analytical projects is a thankless and often ignored cause. The people who control the purse-strings can suddenly get very short-sighted if you are asking for a new server, or the latest software. You can talk about impending obsolescence, increased efficiency, improved customer service and insight until you are blue in the face. You could also build a cast-iron risk assessment, highlighting the certainty of impending doom unless you upgrade. It will be ignored. Most rational arguments will not win you any backing at all. 

Here is the truth.... bottom line is.....

Your solution MUST improve sales. 

Regardless of any other consideration, if you can convince your colleagues that the new solution will improve the selling prospects of your organisation, then you are more likely to get funding for your pet project. 

So before you pitch your idea to the board, go to your sales and marketing departments first. Find out what their problems are and make sure your solution can support what they are trying to do. I guarantee that your likelihood of project approval will at least double with approval and backing by sales and marketing.

Saturday, 8 February 2014

The data knowledge gap

In most organisations, there are many people with general and specialised knowledge about the business. They will have good product knowledge, good knowledge of how the processes work; good customer, financial and regulatory knowledge.

Go into the technical part of the organisation, and they will have specialist information knowledge. They may know how the processes work. They will know know how the fix them if they go wrong. They will also be technically adept at developing new systems and implementing change.

But there is a yawning gap in many organisations - it's knowledge of the data itself. Those who manufacture the data (usually sales or customer services), rarely have to deal with it. Those who read the reports that are derived from the data will not necessarily know about the data itself.

But when projects go wrong, very often the data that is at fault or unsuitable for purpose. Because there is no specialist knowledge of the data itself, the people who produce the reports get unnecessary criticism for getting things wrong. Knowing where you have data problems is a major step for any organisation to take. For once you know you have problems, you can stop laying into your reporters and make organisational, system and process change to keep everyone happy.

Saturday, 11 January 2014

5 tips for running a technical team

I've worked with a lot of B.I. managers, and have also managed a business intelligence team myself. I have seen many varied leadership styles that have worked well. But there are some cardinal rules that could make your life much easier - and here they are:

1. Don't micro manage
Although your team is heavily analytical and technical, there is a very large part of the development process that is truly creative. If you stifle them with endless detail, they will feel heavily restricted. Set high level rules, and let your team take care of the detail.

2. Give them the right tools to do the job.
It is sometimes hard for laymen to understand the complex variety of tools that are available. When your technical team want a new tool, they usually have a clear reason why it is useful. But a non technical person, can fall into the trap of narrow thinking (i.e. if your only tool is a hammer, then every problem starts looking like a nail.) Beg, borrow, cajole, steal... to get the latest technical equipment. Your team will be grateful.

3. Protect them from office politics
Analytical people tend to be intelligent, rational introverts. An introvert does not shy away from social interaction, rather they expel a lot of personal energy - much more than extroverts, who get their energy from such interactions. So office politics can be particularly draining for them. If you can't shield them from negative work influences, they can find things difficult.

4. Teamwork, teamwork, teamwork
Many may well give off "leave me alone" vibes, but they will also want to feel valued as part of a cohesive a team. Don't let them get silo'd. Give them opportunities to work together and make the experience as positive as possible.

5. Listen without interrupting or being judgemental
Right out of management 101... make sure you listen to them - particularly if they are having problems developing a solution. Perhaps there isn't a solution to the problem. Just listen. A problem shared... Be that strong shoulder.