All Change

Yes … It’s all change at the FlyingGod! I have resigned my position with my company and am now moving on to a competitor which has caused a bit of a stir. My new role will be less about SOA than this one was and more about getting involved with the business people and satisfying their requirements whilst understanding their business.

When I think about it, that is what SOA should really be about. The situation currently is that I believe that most SOA initiatives are driven by the IT department and that doesn’t seem right somehow. I tend to feel that the IT department is not the best place for this type of initiative to be controlled. If the business take hold of the idea of Service Orienting their business then the whole project will become much more successful.

I may have blogged this before but in my mind there is a need to start talking about a “Business Oriented Architecture”. This would be the thing that really drives forward the whole concept of the layered architecture that is served from the IT Infrastructure right the way up to the Business layer of the architecture. The whole essence of what SOA is trying to do is to serve the business and provide that level of agility that they need to compete successfully in the marketplace. If our IT architects (Enterprise Architects) were really that, then they would be involved in the business discussions with the IT customer – the business.

Looking at frameworks such as TOGAF and Zachman it is clear that an Enterprise Architecture should naturally take into account the business drivers and requirements. In fact, that is what I particularly like about TOGAF and I have recently been reviewing the TOGAF 9 material. The Business architecture feeds into the TOGAF cycle and then the requirements are at the hub of that cycle. It naturally follows an iterative approach to the architecture but is driven by the business needs and controlled by the business requirements.

I expect that my new role will begin to pull together the business needs and requirements and then make good use of that information to determine where an xOA will benefit the customer. I use the term xOA where “x” could mean anything such as;

• Business
• Data
• Service
• Application
• Infrastructure

My five layered approach then is driven by the requirements of the business throughout the whole architecture stack and we could even start talking about a ROA – Requirements Oriented Architecture.

Hmmm … Now there’s a thought!

Will SOA Last?

I had an interesting thought the other day but before I get to it I think I should explain that I have been around the block more times than I care to remember and I have had my share of IT innovation. SOA doesn’t feel like anything particularly new but I think the speed of change of business is.

I also have a pet hate – I seem to work in an industry that is rapidly making my choices more and more limited. I did some consulting work for a large retail chain in the United Kingdom some time back and I worked on their forecasting tool. The way the algorithm worked, the days sales were taken and analysed and re-stock levels were broadcast from Germany to their warehouses across Europe. These then triggered automatic stock replenishment for every store. Sounds really good doesn’t it – unless, like me, you are outside of the bell curve when it comes to your size! The forecasting technique ensures that whenever I went into their stores they would definitely not have my size in stock. Their forecasting was based on past sales and therefore it rapidly becomes a self fulfilling prophecy. We don’t sell any trousers that size therefore we won’t stock any. Guess what – if you don’t stock any you won’t sell any … and on it goes.

Henry Ford told his customers that they could have any colour they liked as long as it was black – and that is the premise of industrialisation. Today we can have a variety of colours but, let’s face it, the choice is limited and I struggled when I bought my last car to get anything the colour I really wanted. I had a Ford Mustang when I lived in the USA (the 2001 model) and the colour was stunning (to me). It was Laser Red and if you looked closely it had a slight tinge of blue in it that gave it real depth. I like the colour of my current car because of the depth that it has despite being got at for the fact that it is a sky blue!

People like choice – businesses like choice and SOA is touted as the vehicle of “choice”. Is that really correct and is that what people/businesses want?

Two examples;

1. Take mobile phones or iPODs for example. Did anyone think that there would ever be such a market in items that you could so readily customise by buying different fascias for them? Did they think, at the time, that personalisation would be such a big thing?
2. Take Mobile phones again … The phone is becoming a multifunction device, a bit like my Printer/Scanner/Photocopier/Fax machine. In this case people want a single device that does everything for them.

Personalisation and Multifunctionality – there are some words for your Scrabble board.

Back to where I started – I was thinking about Stereo systems and how they have changed over time such that they used to be “separates” but now they tend to be all in one units. The Audiophile will inevitably want the separates and the ability to choose which components they put together whereas Joe Public will probably be quite happy with a composite, multifunctional  system. The Audiophile will want to customise the solution for their own environment whereas Joe Public doesn’t really know the difference between Left and Right, Treble and Bass – just as long as it has a Bass Boost button.

I am concerned, I suppose, that SOA will become the domain of experts and will be perceived as being the expensive option when in fact it should be the cheaper option. The trick is to get that perception across sooner rather than later with tools that will allow Rich Internet Application (RIA) development. Yes … I’m on my “end-user” hobby horse again, but it’s true. If SOA is always “under the covers” then the business will never get an understanding of what it is and does for them. They need to see clear evidence that they get a personalised, multi-functional solution to a business problem that provides them with a competitive edge whilst at the same time giving them rapid adaptability. If they can get that type of result, SOA will be a “no brainer”.

The Final Night Throttle

The last 1h20m – logbook
I had a little more time to make up than I had hoped I suppose. No real reason why that should be a problem, it was just that I still had that extra 20 minutes to do before I did the solo circuits at night. Having said that, Howard, my instructor, was happy to take me out again and we decided to do an extended left-hand circuit to the south of the airfield. It also gave us the opportunity to walk through the extended circuit just in case I needed to pad out the time a little.

We had pretty much decided that once we were under way, I would then keep an eye on the time and try to adjust my circuits so that I was back on the ground having done my five take-offs and landings in just over the hour.

I got to Southend (EGMC) at around 19:15 and was chatting with Howard and Paul (a 747 pilot) for a while before doing the usual walk round the aircraft and pre-flight checks. As usual it looked like the aircraft needed refuelling and so I headed off to the pumps to sort that out. Southend has two Avgas pumps and one of them was U/S (unservicaeable) and that just happens to be my favourite – not really sure why. I was surprised at how noisy the radar was! It is the first time that I have been to the pumps when it was reasonably quiet and the noise from the radar was quite remarkable – hence my remarks .

Steve was the Air Traffic controller this evening and is a man of few words – it’s not a problem but it means that all communication is incredibly brief!

Once refuelled it was back to the “Aviators Bar” (you know, that really doesn’t sound right!) for a final prep before we set off. The sunset was beautiful and the sun was casting lovely red, orange and pink hues all over the remains of the clouds. It was going to be a superb evening.

We did the final power checks – no wind to speak of so Howard reminded me to still point the aircraft away from the centreline. Normally you face into the wind but what happens when there is none. In fact, you need to point away from the centreline so that if the brakes fail you will head off the tarmac and not onto an active runway – good airmanship .

We powered off down Runway 24 with a left had turn clearance. At 600′ I started to turn but – silly me – we were going to do an extended circuit, so I needed to turn back on heading 240 until we were over Canvey Island. At that point we turned towards Sittingbourne in Kent and there are a couple of masts that are lit up on that side of the Thames. Using one of them as a reference point we headed off towards it and then turned another 90 degrees to head towards Shoebury Ness (further East of Southend). The danger areas (D138, D138A and D138B) were not active so we had a clear run until we turned on a heading that had Maldon on the nose. The runway beacon was still in sight and I could still see some of the runway lights even at that distance.

I had tuned in the ILS on NAV1 while I was on the ground and this time the localiser seemed to be working just fine. It was giving me both height and track indications so we used that to get an idea of where we should be heading.

The landing went fine – it is very different at night and so much quieter – much less traffic and we backtracked to the Alpha taxiway and I dropped Howard off to go and spend a very spooky hour in the clubhouse alone. I say spooky, because when I got back he was commenting on all the scratchings and scrapings he had heard underneath and in the ceiling of the clubhouse. Guesses as to what those are on a postcard – or comment – to me please .

So … Now I was off on my own. I had an hour to fill, it was 20:50 (local) so Howard and I reckoned that I needed to be back and brakes on for 21:40L. Off I went- “Lights, Camera, Action” (Lights, Give a good Lookout, Transponder) and I was off. The air was calm but once above the airfield there was clearly a wind of some description acting to push me eastwards. I had been cleared to takoff and report finals (I told you he was a man of few words ) so I didn’t even have to bother with a downwind call.

First landing was not so bad and not so good, but then maybe I am a perfectionist. The nosewheel was juddering like crazy – that’s always unnerving when it happens. Then a short backtrack to line up again. I wasn’t sure about that as I was trying to burn time, not save it!

Second take-off and back into the circuit. Everything was going smoothly with a bit of turbulence until I got to about 800′ then it all calmed down again. Someone else was coming into the airfield and was now on long finals. Aircraft are so much easier to see at night! I had him visual in no time, and then Steve (Air Traffic Control) said that I could do an orbit if I wished. If I wished?! I thought he was supposed to tell me! Now I was confused. What should I do? I didn’t want to do an orbit so I carried on flying straight and level. I know what he was trying to do – we needed some separation as the other aircraft was going to be in my way. I decided to extend downwind but realised that I had better let Steve know what was going on. After the call, he was happy and said I could turn base at my discretion because the other aircraft was now abeam me.

The second landing was still not up to my usual standard and I felt that I was trying too hard and needed to relax.

A short backtrack; don’t forget the flaps; don’t forget the mixture; no really, don’t forget the mixture; check the DI (Direction Indicator); centre the trim and I was off again.

Steve gave me the same clearances every time; “Cleared take-off runway 24 Right Hand circuit wind etc. report finals you’re number 1″. I’m number 1 and I haven’t even taken off yet!

The rest of the circuits passed without drama except that time was now running away with me and so I started tightening them up. I did relax and start watching the ILS – that’s cool!

Finally – back on the ground, backtrack to Alpha Taxiway and then back to the club.

I’m sure most pilots know the elation of having achieved another milestone in their experience. Howard already had most of the paperwork filled out and so we sorted out my logbook, he signed it off and now all I have to do is pay for the aircraft hire, pay the CAA and I will have another piece of paper for my licence book. I think this time I am going to go to Gatwick in person to do this transaction. I worry about posting my licence off to anyone!

So … 5h10m later, I have my Night Rating. I’ll let you know how the trip to Gatwick goes.

Night Time Navigation Exercise

Flight Plan

Log Book

The third stage in my night rating is the Night Time Navigation exercise which involved flying from London Southend Airport (EGMC), north to Bury St. Edmunds, across to Felixstowe and then back to Southend. As you can see from the flight plan (which is all Zulu time) the nights are getting longer here in the UK and so it is getting increasingly harder to get everything done before 10:00pm local time. Why is that a constraint – I hear you ask – because landing fees double after 10:00pm!

The C172 that I fly has an interesting problem with the magnetos and the nose wheel. The magnetos tend to foul up while you are taxiing around with the mixture rich so it is important to lean off the mixture a little while preparing for departure. Tonight the problem presented itself right from the start so we were delayed a little getting that sorted out. The nose wheel problem is more alarming, especially if you are unaware of it! At speed (ie: the last few yards of the take off run and the touchdown on landing) the nose wheel judders so you have to ease back on the stick to release some of the weight.

It still seemed quite light by the time we had done our checks so we dawdled a little before making the final “Ready for departure” call although officially it was night time by then. We lined up on runway 24 and then some final checks on the instruments (Direction Indicator lined up with Magnetic Compass) and we were away.

I haven’t really done much serious flying “to the north” of the airfield so although I had chosen South Woodham Ferrers as the start of track I wasn’t really sure what altitude we would be at by the time we got there. In actual fact it worked out beautifully as we just touched 3300′ when we got there. We levelled off and trimmed the aircraft and the air was so calm that for much of the rest of the flight I was able to leave everything pretty much alone.

The main concern was by a town called “Witham” (pronounced wit-am). Just to the west of the town was Class A airspace which started at 2500′ upwards and with our cruising height of 3300′ we weren’t allowed to enter it. My flight plan took me through the centre of Witham but clearly there were other forces at work that meant that we passed to the east of the town. That meant that we were comfortably clear of the restricted airspace.

We tuned in the Clapton (CLN) VOR and because everything was so calm I started playing with a second VOR (Barkway BKY). The second radio that I have in the aircraft had a problem with the frequency display. The digit after the decimal point wasn’t showing at all so I had to cycle through until I got a response. This demonstrated the importance of the I in “SID” – Select, Identify, Direction. Once I got a response from a station, I could then listen to the morse code identifier of that station and determine if I had selected the correct station. In this case I was relieved to hear B-K-Y being broadcast (you can see from my flight plan that I had already jotted down the morse code for BKY) and started to use that to triangulate my position. As I am still flying under Visual Flight Rules (VFR) this was unnecessary but having written flight simulators and played with them for most of my computing life it was great to be using these instruments in anger.

Bury St. Edmunds turned up right on the nose and then we turned towards Felixstowe. The routing I had chosen took us directly over Wattisham military air base but of course, it was night time and they were closed for business. We didn’t need to speak to them as we were above their airspace but we did see a smattering of airfield lights – clearly the runway lights were all swiched off – they weren’t expecting anyone.

Felixstowe came up bang on the nose – which was gratifying and I remarked on how small it looked compared to the other towns in the area. Lots of flashing yellow lights told us that it was a busy port that was working at that time of night. I counted about four cargo ships in port, all being loaded/unloaded.

From Felixstowe we turned back towards Clacton and therefore home. Using the VOR we pointed ourselves directly at Clacton. The instrument in the cockpit that locks onto the VOR and helps with the navigation has a “To” and “From” flag to give you some idea of whether you are flying towards it or away from it and as you go over it the instrument can get quite confused. This is known as the “Cone of Confusion” and we had decided that as we entered the “Cone of Confusion” we would call up Southend Approach and let them know that we wanted “Rejoin” instructions.

It was very much a direct approach from there – no messing about in the circuit, just a straight in approach. If it hadn’t been for my instructor I wouldn’t have started looking for the runway lights as far out as I did, but I was quite surprised as to how far away you could see them if you were looking for them. I had actually misunderstood the orientation of the runway, so when I spotted it and we were already pretty much lined up with it, I was quite surprised.

It was strange to me that doing a straight in approach left me unsure as to what to do when. In a normal rectangular circuit I was geared up for doing things at particular parts of the circuit so this time I was caught off guard.

It was an uneventful landing – nose wheel juddering aside and then a backtrack to the club and some banter with the control tower.

… and that was that as they say. So … I am now 3h40m in … only 1h20m to go.

SOA is not the destination – it is the journey

SOA is not an end in and of itself, it is the way you move to a more agile environment.

I hear a lot of talk about Service Oriented Architecture (SOA) and it all seems to give me the impression that SOA is some form of goal that we should all aspire to! In many respects, like many other IT “revolutions”, it has taken on a life of its own and has become the battlecry of the faithful. I don’t see it that way at all … I don’t think anyone will ever get to a point where they can legitimately say that they have a Service Oriented Architecture – and even if they do, is that really where you want to be?

The whole question arises because it is driven by the IT side of the industry as a potential solution to the problem of constant change. That’s why I think that SOA is more about becoming agile. I think that those of us that are involved in business in any way will identify with the need to be able to change and morph the organisation rapidly and effectively to handle the marketplace as it currently stands. Actually we need to be able to adapt our business so that it reflects what we expect the marketplace to become while still maintaining support for the current state.

In that respect, SOA is not really a goal, Agility is the goal and SOA can enable that goal to be realised. However, it is necessary for SOA to really be about a Service Oriented Architecture. The reason I emphasise this is that I see that there are five components to an architecture and they are;

1. Business
2. Information
3. Service
4. Application
5. Infrastructure

For the Architecture to be truly an SOA, the whole stack needs to be Service Oriented and currently I don’t think that is the case.

I was having a conversation with some Japanese colleagues this week where we were discussing a data-centric approach to SOA and it made me think about the fact that we haven’t really begun to recognise the need to examine our data sources and build some form of Service Orientation into the way that we handle the data. When I was working with Ford Motor Company I was involved in a project to create an “Information Bus Service”. The thinking at that time was around creating a service that would allow anyone in the company to request data from any source in the company using this service. This would then bring about a rationalisation of the data sources or, at the very least, enable us to determine which data sources were the most popular for a particular data entity.

The real problem with this was that instead of setting the service up and allowing sources to wither and die through lack of use, the aim was to create a service that would pull back a huge aggregated record for an individual data entity. This seemed like a great idea at the time except that it is almost impossible to support CRUD (Create, Read, Update, Delete). If your record is an amalgam of multiple records from multiple databases, how do you Create, Update or Delete a record in the underlying databases? If all it does is allow you to read the data then it hardly can be classed as a service.

I believe we should have looked at providing access to data sources and then let natural selection take its course. That would then have enabled us to assess which databases were significant and which were not.

This is the “Journey” that I am referring to. Unfortunately there may be no end to the journey which is why SOA cannot be the destination. SOA needs to happen at all levels within the architecture of the business and that means that it could affect the way that a business is structured. The data sources will need to change to adapt to the new and generic business model right the way through to the infrastructure.

What’s next?

Work has, not surprisingly, got in the way of going and doing any more flying for the moment. Actually, it is a combination of work and some expensive work I am having to have done on my teeth (Root Canal). It really phased me to find that I needed some work on my teeth as I have always had no problems with my teeth.

However, take a look at the current page of my logbook if you are interested. I’ve posted it on docs.google.com where I am able to upload Open Office and MS-Office spreadsheets and text documents. I’ve posted the last page of the flying log together with the Aircraft flown and Aerodromes visited page. I was in New Zealand recently (welcome to all my NZ readers) where I was hoping to have a flight in a “Floatplane”. For the non-Kiwi’s among us, that is what we in good ol’ blighty call a “Seaplane”. Unfortunately the winds around Wellington were somewhat prohibitive as well as the personal schedule and so it never happened. I have, however, flown in the co-pilot seat of a seaplane in Seattle when I was over there visiting Microsoft in Redmond. It was an unusual sight coming in to land on Lake Union with little boats and other sundry jetsam and flotsam floating around on the lake in front of us as we descended. The striations on the picture seem to be something that the camera picked up from the propeller.

If you do take a look at my logbook you’ll see that I fly out of EGMC which is the code for London’s Southend Airport. Sounds grand but in fact it is a fairly small airport from which the Ford BAE146 flies to Cologne every day. They have a really good selection of jet aircraft sat around and seem to do quite a lot of maintenance work. We do have a Vulcan Bomber right next to the club I fly from.

I’ve bought the latest version of the Aeronautical Charts for Southern England and Wales (Edition 33). It means I can fly again as I am not allowed to fly without the latest charts. So, looking at the chart I was checking out what I need to do next in the night flying arena. Talking to my instructor we have discussed the idea of flying from Southend up to Bury St. Edmunds which means flying over Wattisham military airfield, we’ll probably hanger right there and head back over Wattisham to Felixstowe. From there we can follow the coast back to Southend although there a couple of danger zones in the area which we’ll need to steer clear of. It’s a distance of about 100nm. Problem is, of course, we have moved across to British Summer Time (BST) which means that dusk arrives later than it was doing. Interestingly enough, landing fees go up after 10:00pm at Southend so I’m going to have to get a move on if my final five circuits aren’t going to cost an arm and a leg.