Random Disconnected Diatribes of a p&p Documentation Engineer
So at last my new company computer arrived. And, like all competent technology users, I carefully perused the operating manual before starting it up the first time. You never know, it might say something important. And what I found has made me wonder if I actually dare use it at all.
For a start, the machine is supposed to be a Windows 8 equipped touch-screen laptop, but amongst the dire warnings on the “Safety Precautions” page is this:
Perhaps the latest update to Windows incorporates voice recognition or mind-reading capabilities instead. Though, I’m not sure the machine is actually a laptop at all. It also says:
Obviously that’s why they have to call them “Notebooks” now. It’s probably some EU directive designed to protect careless users from “hot leg syndrome”. In fact there’s even a section labeled “DC Fan Warning” on page 5 of the manual that says “the DC Fan is a moving part that may cause DANGER” (their capitals) and that you should “ensure you keep your body from the moving fan blades”. I reckon you’d need to have a body even smaller than a size zero fashion model, or at least have very thin fingers, to fit through the grille.
But, moving on, there’s also useful advice for that situation when you have a problem with your domestic heating system, and decide to email the supplier to ask for advice:
And should you decide, after all these warnings, to take a chance and use the computer there’s also some additional precautions you should take when selecting USB leads and other peripherals. For example, it seems that I should only use leads that are made here in the UK rather than any that might be classed as “foreign”:
Mind you, because the computer was (like every other item of technology I own) made in China, maybe it means I should only use Chinese leads and peripherals. Or does it mean that I CAN carefully insert foreign objects into it, and it’s only a problem when I carelessly or brutally “shove” them into the machine?
And then, last of all, it says that “incorrect installation of battery may cause explosion and damage”. Perhaps I should get our ‘Elf and Safety people to have a good look at the machine first before I risk my life (and sanity) by switching it on...
With our Big Data guide now finished (at least for the preview release) I managed to grab a few days off to clear my head of pigs, ants, beehives, and furry yellow elephants. And to try and figure out where we go next on the Azure guidance path. Mind you, much of my vacation time seems to have involved meeting several other types of wildlife.
It started with a trip to Monkey Forest. Compared to Monkey World in Dorset, which we visited some while back, it’s much less dramatic; but still worth a visit. It’s basically what it says on the tin - a forest with a hundred or so Barbary macaques roaming freely.
They’re docile creatures that seem quite happy to have humans wandering around, and - unlike the wild monkeys you see in places like Gibraltar that steal handbags and sunglasses - they are neither aggressive nor noisy (in fact, mostly they seem to be asleep). However, a few did manage to stay awake long enough to have their photos taken, including several very young ones that are as lively as you’d expect. And most did show some signs of activity at feeding time, as you can see in the photos here.
Another day out was to Ripley Castle, a place my wife had seen on TV and wanted to visit. As there’s a town called Ripley only a few miles away from our house I gladly agreed, thinking it would save me fuel and a long trip in the car. Except that the Ripley with the castle is in North Yorkshire instead. It’s a “planned village” that was built by the owner of the castle and is a heritage area, so it’s still very old-world and historical. And very pretty.
The castle itself is part occupied by the owners, but has large sections available for weddings and other functions, beautiful gardens, a deer park (though we never saw any deer), and a superb selection of rooms that are part of a guided tour. The lady guide was excellent and the historical tour was fascinating, including tales of visits by kings and by Oliver Cromwell and his army. And interesting details of the huge range of artifacts on show, including the fact that in bygone times shoes were not made as a left and right pair - you just bought two the same and wore them till they fitted your feet (or until your feet fitted them).
Our final trip of the week was to the delightfully named seaside town of Mablethorpe. We have a rule in our house that you can’t have a holiday unless you get to see the sea at some point. My wife had never been to Mablethorpe so it was a voyage of adventure for her. Though, purely by chance, we parked the car next to the Mablethorpe Seal Sanctuary and Wildlife Park. So it was immediately obvious where we’d start our adventure.
And I have to say I was surprised to find how extensive it turned out to be. The entry price was just a few pounds, but - as well as the seals we expected to see – there’s a tour through pre-history that shows how the area evolved over the last 300 million years, followed by a wide range of rescued birds and animals including owls, ostrich, various seabirds, lynx, meerkats, and more. And monkeys, just in case we hadn’t seen enough already.
The only real downside of the week was discovering that the unusual badger that comes to visit us at home had been killed (probably hit by a car). We’ve been watching him for a while on our garden and tree cameras. Mostly we see him at night illuminated by the infra-red lights, and so the pictures are monochrome rather than color. But it’s clear that he didn’t have the usual black stripes of other badgers; instead he seems almost albino. As you can see from the photo below, taken after his demise, he really was an unusual color.
We’d be interested to hear from anyone who has seen badgers like this, or can provide any more information. Perhaps, with luck, we’ll see some similar colored cubs visiting later this year...
Would you buy a vacuum cleaner just for your cat to play with? After my wife watched some videos on YouTube where kittens were chasing and sitting on top of one of those robotic UFO-shaped vacuum cleaners, I was informed in no uncertain terms that we should have one for our two cats to play with.
And by happy coincidence it was coming up to my wife’s birthday, so it seemed like an obvious gift choice. But here’s the problem – these days it seems like giving household necessities as gifts is a thing of the past. If I bought my wife a new ironing board as a Christmas present I’m sure I’d be in the dog-house.
Is this something that’s gradually changed over the years? I can remember, many years ago when I worked in a hardware store, that our Christmas window displays were full of useful gift ideas. Power tools, a lawnmower, or a toolbox complete with a comprehensive range of wood screws for him; and for her a washing machine, the latest design in enamel saucepans, and deluxe sets of clear cookware. And, of course, an ironing board.
But casting fears aside, and taking my chance in the domestic relationship stakes, I bought an iRobot cleaner and presented it to my good lady on the day. And, thankfully, she was delighted. However, the nub of this week’s post is that, having added yet another computer-controlled example of modern technology to our household, I can’t just relax. As a typical geek, I need to know how it works.
OK, so I know that it learns the shape of rooms by bumping into things. But the fact that it slows right down before it bumps into something indicates there must be some other sensors in use. And how does it find its way back to the charging station when it’s not in line of sight? From memory? It’s not as though it follows any pattern that you can figure out as it seemingly randomly navigates its way from room to room. And every now and then it flashes up a message that it’s “spot cleaning” some area. How does it recognise the bits that need extra cleaning?
So much for thinking it would save me time. Instead of hoovering myself, now I have to follow it around pondering on the program it’s running and trying to figure out how it does it. It even seems to have a sense if fun. After pushing the tray containing the cats’ food dishes into the middle of the hallway, much to the chagrin of the cat who was watching at the time, it managed to find one of the table tennis balls the cats occasionally deign to pat across the floor. For several minutes it carefully nudged the ball across the lounge carpet, into the hall, and finally lost it under a cupboard.
Maybe, after a few days, it won’t bother cleaning the floor and will just play ball with the cats instead. I could have bought them a clockwork mouse for a fraction of the price…
It seems like everything is "HD" these days. Your laptop has a "high density" plastic case, your car has a "heavy duty" battery, and your TV has a "high definition" picture. So I guess it's only to be expected that your data analysis tools will be "highly distributed". And at last we're "happily done" with our guide to HDInsight.
Yep, after several months fighting with Microsoft's Big Data solution we've shipped the first version of our guide to Windows Azure HDInsight, based on the current preview release. It's been one of the most troublesome guides in terms of figuring out the structure and the content boundaries; and interfacing with the exciting world of open source technologies (as I ruminated about just a few weeks ago). But we got it all together in the end.
Our guide contains the obligatory "What is Big Data" section, as well as describing how HDInsight integrates with the rest of the Microsoft data platform. Then there's a chapter about loading data, one about performing queries and transformations, and one about consuming the data. From there on in the guide discusses automating the whole data analysis process; plus the usual management and monitoring topics as well.
What's likely to be of most interest to developers, however, are the scenario chapters and associated code examples that show how you can use HDInsight in four distinct ways: as an experimental platform for investigating interesting data, as an extract/transform/load (ETL) mechanism for data validation and cleansing, as a data warehouse that you can turn on and off on demand, and as a data source for your existing enterprise business intelligence (BI) systems.
Here's our tube map for the guide:
Unlike most other books and guides, we've concentrated on integration of HDInsight with your existing business processes, and combining it with data analysis and visualization tools such as Power View and GeoFlow, as part of an end-to-end solution. Yes, it's Microsoft-centric - but, hey, that's who pays my wages...
So I considered naming this week's post "Climb Up On My Roof Sunny Boy", but I'd probably get sued by Al Jolson's estate. Though, as we're now officially inverted here at chez Derbyshire, it did seem appropriate. What I wasn't expecting was the domestic strife that is a direct result. They don't tell you about that in the glossy brochures...
The inversion we've encountered is, of course, a direct result of the team of local guys who came and nailed solar panels all over our roof and installed a box the size of my huge Dell workstation in the garage to invert the volts from the panels into real electricity. I did ask the guy if it meant they come out upside down, but he didn't laugh. I suppose he's heard that too many times before. But why don't they just call it a "converter"?
At first it seemed like a great idea. The little red light on the generation meter was flashing like mad and I could envision the pound notes rolling in. Until the RCD earth leakage thingy tripped out. Regularly. Turns out that, along with the seven 1000W UPSs and who knows what else in the way of technological magic there is plugged into the house circuit, the additional leakage from the inverter was too much for the mandatory RCD.
However, after they came back and moved the inverter and my server cabinet to separate consumer units (it really should have been done like that from the start), we're up and running again. It's just a shame that the sun hasn't come out since. If we have a miserable summer this year, you can blame me.
And then I discovered the next problem with going photo-voltaically green. I suggested to my wife that she should use the tumble drier only when it was really hot and sunny so it used the free electricity, but on days like that she wants to put the wet laundry outside on the washing line. When it's raining, and there's no free electricity, she uses the tumble drier. And you can imagine the response when I suggested she stay in and do the ironing on nice warm days, then go shopping when it's snowing.
Still, it's nice to know that for part of the day all my computers are costing me nothing to run; and it even diverts any spare volts we generate to the immersion heater to reduce the amount of gas we use to provide hot water. With the current rate of climate change supposedly due to global warming, I should easily recoup my investment by 2050.
But having a new technology device available to play with is, of course, too much for a computer geek to ignore. Having discovered that the inverter exposes its generation log data over Bluetooth, it was inevitable that I'd need to find a way to add the data to my weather website. And I can justify all this effort by the fact that I don't need to spend a couple of hundred pounds buying a rather boring-looking remote monitor device to see what's going on.
So out came Visual Studio and, after a couple of days, I have an automated system for generating data and charts; and I can expose the data over my internal network and publish the charts on the web. If you're interested, you can see them here. And if, by some remote chance you have an SMA Sunny Boy inverter (or you're prepared to modify the code to suit your inverter) you can download the utility and the .NET source code here.
What I need to discover now is if, on a typical English summer day, we're actually generating enough electricity to run the inverter and the computer that monitors it...
At a meeting of our remote workers group the other day I noticed how competitive people are about how far they are away from their Microsoft office. It's almost like you get a prize for being the remotest (hopefully measured by location and not character trait). What set me thinking was how the different people measured their remote distance.
For example, early contributors to the discussion expressed the distance in miles. Starting at about fifty the figure steadily increased until one person decided that their remote distance was "three hours". Though they did then qualify it with "180 miles" and mention that this was driving time in a car. I suppose if you don't own a car, three hours could be 40 miles on a bus (based on the journey times and regular stops in my neck of the woods), or 12 miles walking. So hours doesn't seem to be much of a useful distance measurement scale.
In fact it reminded me of a conversation I had some while ago (in my pre-Softie days) with a nice lady from Digital River in Minneapolis. At the time they were selling my software, and the nice lady had just been appointed as my marketing representative. She suggested I should call and see them when I was nearby and I happily agreed, even though I had no idea at the time where in the US the state of Minnesota is. But I often passed through Chicago airport in those days, and when I mentioned this she helpfully suggested that it was only a couple of hours from their office. "Great," I said, "Next time I'm over I'll hire a car and call in." To which she replied "Err, no, that's two hours by 'plane..."
So now I have another measure of distance - how far away somewhere is when flying there. Here in Ye Olde England a distance of two hours by air covers something like nine countries. Being a remote worker where there are a couple of whole countries between you and the office surely wins some kind of prize.
But getting back to the discussion at our meeting, I decided not to upset the rest of the group by mentioning that I'm 4,791 miles away from my office as the crow flies. Or, by plane via Amsterdam (my usual route) 5,935 miles away. And the trip, end to end, takes about 23 hours. But if I said I was 23 hours away, somebody would just suggest that I went the other way round the world because then it would take only an hour.
Or, if I was walking, it would only be four miles...
It's easy to imagine that the computer is a recent invention. A search of the web reveals a host of machines claiming to be the first electronic computer, and all are mid-20th century. However, what's harder to determine is the first appearance of a truly programmable computer. After watching a fascinating TV documentary this week, it seems that amongst the first was a model of a small boy writing on paper with a quill pen. And it was built more than two hundred years ago.
The automaton named The Writer was built by a family watchmaker business in France in 1774, and is just one of a series of clockwork-powered machines created around the world during that period. The Writer sits at a desk, dips a quill pen into a dish of ink, and then slowly and delicately writes beautiful cursive text onto a pad of paper on the desk. You can see a full description and pictures on the IW Magazine website.
OK, so clockwork automatons and toys had already been around for ages at that time. What struck me about The Writer is that the mechanism uses a large wheel containing details of the letters that will be written. But the segments of the wheel are removable and replaceable, so they can be changed at any time to write completely different text. It's effectively a stored program computer that converts a set of instructions into some recognizable output.
Yes, you can argue that it's a fairly simple transformation from program to output; there is no intermediate processing as such. Each interchangeable segment of the control wheel simply defines the set of movements of the automaton's hand. It's not a general purpose computer either - you can't tell it to play chess, or calculate the trajectory of a cannonball. But it's a fascinating stage in the development of adaptable stored program machines.
Mind you, the documentary also showed an automaton called The Turk, built in 1770, that supposedly could play chess - and could even beat the most skillful players of the time. This completely astounded those who saw it in action, and certainly would have been an incredible achievement if it hadn't turned out to be powered by a real human chess player hidden inside.
The documentary was produced by the BBC here in the UK, written and presented by the enigmatic Simon Schaffer, and you can even see a clip of The Writer on the program's website page. Yet, in a remarkable contrast to the capabilities of our ancestors, news this week revealed that the BBC has removed the clock showing the current time from all of its website pages.
Why? Because somebody complained that all it did was show the time on the origin server that generated the page, which might not be accurate for the location of the viewer (and presumably, if they left the page open for a few minutes, would be the wrong time anyway). The comment from the official BBC spokesperson was that it would take 100 developer days to change it so that the clock showed the correct time, and this could not be justified.
It comes as a bit of a shock when you phone your bank with a routine enquiry, only to be told that they're no longer a bank. When I mentioned it to a colleague she asked what they're turned into. Perhaps now they are a greengrocer? Or maybe a pet grooming salon? Have they spent my meagre savings on hair dryers for dogs? Or cabbages and carrots?
Fortunately, the change is only due to the consolidation and reorganization of the several banks that are now owned by the government (or, to be more accurate, by us taxpayers) and my bank is continuing to operate but not accepting any new business. However, it also is not accepting any changes to existing accounts either, and their advice was (amazingly) "I suggest you move to another bank!" So I have done.
Of course, the problem is that our lives are now so complicated by direct payments, standing orders, bank credits, and other bank-related puffery that you'd assume switching to another bank would be a complete nightmare. Even though they say they have all these clever systems to move everything over automatically. But, hey ho, what else can I do?
So I chose a bank that's UK-based, part of a major banking group, and has top ratings for customer service. And that doesn't charge an arm and a leg every month for "account maintenance," or for "valuable additional services" that I don't need. And when I phoned their customer service department to ask a few questions it was answered immediately by a real person instead of the expected "press 1 to be annoyed and listen to inane music" message.
Even setting up an account was easy and quick online, and for a while it looked like a wonderful financial honeymoon was on the cards! Until it came to setting up some additional services, where the system seems incapable of selecting the appropriate one of the two joint account holders. So I log off and apply for the additional service as a new customer, but now it says I can't do that either because I'm already a customer.
No problem, just call the nice man on the free-phone number and ask him what to do. That's when you discover that (a) it's not a free-phone number once you become a customer, and (b) on the number you have to use now they do have a "press 1 to be annoyed and listen to inane music" message. So I'll submit my question over their secure messaging system instead. Except that it won't let me tell them which account I want to talk about because "the application is pending".
Maybe I just expect too much from technology...