Obviously, I have not been blogging much lately. This is a good sign, inasmuch as the new gig is going well and keeping me ridiculously busy.

Another sign of the job going well and me being busy is that I have been spending a lot of time in aeroplanes lately. I have yet to beat my personal record of four flights in one day (LIN-CDG-CPH-CDG-LIN, since you ask) but I haven’t been idle either.

As you fly more, you start to notice and appreciate the little things: which airline has the better app, which airport is easiest to get around, and so on. European airlines (at least national carriers) still do decent on-board service too, which leads me to today’s observation.

I was on Lufthansa, and was served this beer:

First off, it’s the perfect beer to serve: it’s a Warsteiner, so a good German beer on a German airline. I also like the aluminium bottle celebrating Lufthansa specifically. An aluminium bottle is clever for an airline, since the material is much lighter than glass, and I believe may also help extend the shelf life of the beer, while the bottle still seems more "premium" than a can.

Lufthansa is one of my two main airlines; the other is British Airways¹, and the contrast in their beer policies is instructive. It’s certainly a missed opportunity when the cabin crew respond to a request for a beer with "Heineken or Tiger?". This hardly helps British beers’ reputation abroad…

It’s all the more puzzling since BA is quite happy to stock London Pride in their (excellent) lounges. Why not stock that on board, with one lager as an alternative for wrong-headed people who prefer that, instead of the indistinguishable choice between two non-British lagers?

I haven’t blogged in ages - which is a good thing, I hasten to add! It’s just that I have been drinking from the firehose at my new gig. It’s now more than a month since I started at Moogsoft, and I think I can begin to talk about what it all means.

I joined Moogsoft from BMC, but it’s important to note that I did not join BMC, I wound up there as part of the BladeLogic acquisition. BladeLogic was my first startup, and it was a huge amount of fun, a great learning experience, and probably my period of fastest professional development to date. Before BladeLogic I was at Mercury, but I quit to join BladeLogic, due in no small part to the acquisition by HP¹.

What is BladeLogic?

Both BladeLogic² Operations Manager (BLOM) and Incident.MOOG are innovative products in their place and time. BladeLogic, together with Opsware, redefined what server configuration management meant, and both companies went on to be acquired by larger "Big 4" IT vendors: Opsware by HP, and a year or so later, BladeLogic by BMC.

For a while both products thrived in their new environment, but in the last few years, both have been flagging. There are many reasons for this, from internal politics at both BMC and HP acting as distraction, to the rise of open-source configuration management tools such as Chef and Puppet. However, I wonder if those tools were simply the end of an era.

This is a known pattern: technologies reach their peak right before they get displaced by their successor technologies. The speed record for steam engines was set in 1938, but a diesel engine had already exceeded that speed in 1936, and by the 1950s [diesel locomotives were well on track to replace steam traction](

https://en.wikipedia.org/wiki/Diesel_locomotive#History)³.

This pattern even agrees with disruption theory: investment continues in the old technology, but it simply becomes overly complex and uneconomical compared to simpler, (initially) low-end competitors.

Pets vs Cattle

This disruption is exactly what I see happening now. The BladeLogic model of painstaking management of single servers is still relevant, but mainly for legacy or specialised systems. Much new-build development is already moving to disposable or even stateless VMs or containers, according to the classic "pets vs cattle" model.

In this brave new world, there is very little need to worry about the configuration of your "cattle" containers. Something like BladeLogic is arguably overkill, and users should instead focus on their provisioning process.

Of course it’s not quite as simple as that. Cloud zealots have been talking about replaceable short-lived cloud servers for a while, but it hasn’t really happened outside of some rather specific situations. The lifetime of VMs in the cloud has often ended up being much longer than this model would suggest, meaning that there is plenty of time for their configurations to drift and to require management in place. Part of the reason for this is that management processes and techniques that are still based on the paradigm of a persistent physical server. Much of this Weltanschauung has been adopted wholesale for virtual and cloud-based servers without much reconsideration.

There is also the topic of security and policy compliance to be considered. Given long system lifetimes, it is not sufficient to be able to validate that something was deployed correctly, as its configuration may have drifted away from that known-good state. The desired state may also change as vendors release updates and patches, or new security vulnerabilities are disclosed. In all of these cases, some mechanism is needed to check for differences between the current live configuration and the required configuration, and to bring the system into compliance with that desired state.

However, this is now. As Docker and other container-centric infrastructure technologies become more prevalent, and as business functions continue to migrate from legacy to new-build applications, I would expect that that paradigm will evolve to replaceable plug&play infrastructure components, and do so everywhere, not just at the "unicorn" companies.

What does it all mean?

Lots of smart people are working hard to enable infrastructure to be managed as code. One of the characteristics of code is that you don’t change it in production, you develop it offline, then release it and don’t change it until you overwrite with a new version. The big variables that I think will affect the speed of the transition to this new model are firstly, the rate of replacement of legacy applications, and secondly, the evolution of IT management processes and culture to take advantage of new tools.

BladeLogic itself has the opportunity to evolve to have a role in the new model, of course. Regardless, BladeLogic was a huge part of my career development - and just huge fun, if I’m honest - so I will be watching development of the IT infrastructure management market intently, but no longer from the front lines.

I had kind of lost track of Duff McKagan after Guns ’n Roses imploded. Axl and Slash stayed more or less in the public eye, but Duff, Izzy and the others dropped off the radar.

Now here he turns up doing Sweet Child of Mine, accompanied by Krist Novoselic of Nirvana - on the accordion, of all things, and playing both the melody and the vocal part. Check it out:

But what I really wanted to point out was that Duff is wearing a tie, of all the unexpected things for a member of Guns ’n Roses to wear. This just goes to prove my ongoing contention that a tie is the most alternative thing you can wear, now that showing up for business meetings in Chucks barely raises an eyebrow.

Since I’m in the US this week, I thought I’d go check out an Apple Watch.

You can’t just rock up at the Apple Store though - you need to make an appointment beforehand through Apple’s concierge service. I duly did this, and was greeted at the door by a blueshirt - by name!

The device itself was pretty cool - I particularly like the Watch with the Milanese Loop band - but the try-on experience was underwhelming. The watch you get to strap on is running a canned demo video, which is not interactive in any way. To test the applications, you have to use the Watches that are in the display units with the iPad controllers.

The problem with this experience is that it doesn’t let you try the features I was most interested in: Glances, and the Taptic Engine. I was really curious about these two, and in fact I still am, since I didn’t get to check out either one.

Both of these features relate to notifications. The Taptic Engine is what generates the "tapping" sensation on your wrist that tells you that you have a notification in the first place. It can also be used for other things, such as giving you silent navigation directions. Meanwhile, the idea of Glances is that when you receive that notification on the Watch, you can rotate your wrist towards you, and the Watch wakes up and displays the notification on the screen. At this point you can either rotate your wrist away again, in which case the notification is dismissed, or start interacting with it in whatever way is appropriate.

Based on everything I’ve read and heard, I expect the main value of the Watch to be in its ability to help wearers process notifications. If you can take a quick look at your wrist to see whether a notification requires your immediate attention or not, that is much less disruptive (and rude!) than getting your phone out, unlocking it, and so on. However, I can’t tell how good the Watch actually is at doing that on the basis of a canned demo reel or interacting with stored notifications!

I have also seen some comments that the watch is sometimes slow to display the actual, y'know, watch face. The screen is off when you're not looking at it, but it's supposed to turn on when you rotate your wrist to look at the watch. This was something else that the try-on experience doesn't allow you to test.

I really hope Edition buyers get a more complete experience. Maybe I should have tried for that instead? I was after all conspicuously the only person in the Apple Store wearing a blazer…

If nothing else, one good thing that has come of the Watch is that Apple has redesigned its UK power adapter.

Detail of new-style UK Apple USB plug (included with Watch, or available separately). pic.twitter.com/8PlA9zdz8E

— Matt Gemmell (@mattgemmell) May 6, 2015

I really hope that when the Watch launches in continental Europe, that plug also gets redesigned. The US Apple power adapter, with its folding prongs, is definitely the best to carry around. The UK & European¹ versions always seem like afterthoughts in comparison, sticking out and catching on things.

(╯°□°)╯︵ ┻━┻

After 7 years, it's time to move on. Today is my last day at BMC, then I get a whole one day off - and even that only because it's a public holiday - before I start my next gig.

Today

I hand in my badge and gun - er, I mean, MacBook - and on Monday morning, bright and early, I will be in San Francisco to start my new job at [Moogsoft](

http://www.moogsoft.com)

. I could not be more excited!

It's definitely a wrench to leave after so many years, and so many different roles. In particular, I have to admit that I will miss the view from BMC's Milan offices, with the Alps in the background:

Onwards to new adventures!

This is how things are now.

The other day I was shopping for some end-of-season discounts on ski gear. I'm okay for snowboard stuff, but I could use some new ski gloves, and you can never have too many socks and base layers.

I found a few things, put them in my shopping cart, but then I got distracted and didn't complete the transaction.

I came back the next day, only to find that the shopping cart had timed out and all my selected items had flown back to their various departments. I didn't really need anything, so I rage-quit the browser tab and thought no more of the issue.

…except that for the last week the retailer has been stalking me around the web, mainly but not only on Facebook, with ads for exactly the items I had been looking at.

So let me get this straight: you know who I am, you know what I was looking at, and you're willing to pay Mark Zuckerberg to show me ads for those products - but you can't be bothered simply to leave the items in my shopping cart so I could, I don't know, check out and pay you for them?

An ounce worth of common sense beats ten tonnes of "social intelligence", "sentiment analysis", and stalker-ish Facebook advertising - trust me on this.

Image by Elisabetta Foco via Unsplash

I travel a fair amount, mostly for work but also for pleasure. Despite travelling so much, I still enjoy it a lot. I see colleagues and friends who travel much less than me get all stressed out, and I think the difference is in preparation.

Part of that preparation is in taking the right luggage for the job - and yes, that does mean having multiple items of luggage that are around the same size!

Rule 1: Never Check Luggage

Let me just repeat that one because it's so important: never check luggage.

Sure, it can be annoying to schlep luggage around the airport - but airports are designed to make that easy. Airports have smooth floors, wide spaces, and lifts and escalators for moving between levels. The tough part is manoeuvering luggage around on the street and on public transport, neither of which environments is designed around luggage. The thing is, though, you will have to do that part anyway once you have (with luck) retrieved your checked luggage.

Why take hand luggage? Simple: so that it doesn't get lost along the way. Nothing stresses me out more than the idea that I might arrive at my destination without clean clothes or toiletries, so I'd rather just have them on me. This goes double if you're travelling with kids: you need at least one change of clothes per child, nappies if the children are still using those, favourite toys and books, wipes, snacks, drinks, etc etc etc. Those items must be at hand at all times, during and after the journey, so you have to have them in hand luggage.

Bottom line, this means that you have your stuff with you even if you get delayed or rerouted suddenly, and there is no way that your luggage can end up in one place when you are in another. This already removes a major source of travel-related stress.

Rule 2: What Goes in the Luggage

The usual question I get at this point is "but how can you go away for a whole week with just that dinky suitcase?". Well, it takes planning and the right luggage, but it can be done. And no, I don't just show up wearing one of five identical black T-shirts or something like that! My typical work trip load-out looks like this:

• Suit (a minimum of one suit per two days' wear)

• Shoes for suit (you don't want to travel long-haul in suit shoes if you can avoid it)

• Shirts (a minimum of one shirt per day; a spare shirt is good if you have room)

• Ties, pocket squares, etc (even if you meet the same people wearing the same suit on subsequent days of a trip, if you change shirt, tie and pocket square, it comes off as a different outfit)

• Sweater (one in the case in addition to the one you travel in - even if you don't have different trousers to change into, losing the tie and opening your collar in the evening helps to relax)

• Toiletries (because they're going in hand luggage, these need to be sub-100ml - but these days, you can buy most things in travel sizes, and decant anything else)

I realise this is a bit male-centric, both in the talk of ties and the assumption that a single pair of shoes - well, two: one worn and one packed - can possibly be sufficient. This is my experience that I'm sharing here, though. I may try to get SWMBO to add her recommendations for women travellers at a later date.

On the topic of shoes, here's a pro tip: save space and protect your shoes by stuffing your underwear into the packed shoes. This will prevent the shoes from getting crushed and also free up the space that the underwear would have taken up inside the case. Just try to avoid mixing the "clean" shoe with the "dirty" one if you are on a multi-stop trip…

If I'm travelling for internal meetings, I probably won't bother with a suit & tie, but I'll have trousers and blazers instead, which take up at least as much room if not more.

It's all about the outfits, though: try to plan outfits that you can mix & match, so you wear a pair of jeans with your travel sneakers for informal activities, but dress up the same jeans with brogues, a shirt and a blazer for "business casual"¹ events.

Bag of Holding

So what sort of Bag of Holding is all this going in? Just a standard-sized item of cabin luggage - you do NOT want to be That Guy (or Gal) arguing with gate staff about the size of your roll-aboard. Rigid sides are better for this purpose because they are less forgiving, so you are not tempted to stuff just one more sweater in there.

Make sure your case has TSA-compliant locks, just in case you do have to check it for whatever reason. Also, have a plan for your liquids and any other fragile or valuable items that are in your hand luggage and that you don't want to entrust to the tender mercies of the baggage handlers. A drawstring bag is good for this, as it takes up next to no room when empty, and can also be used for other purposes if necessary: keeping a wet swim suit apart from dry clothes, impromptu shopping trip, or whatever.

Remember also that not only do different airlines have different luggage restrictions, but there are also regional variations. The default IATA size for carry-on baggage is as follows:

Cabin baggage should have a maximum length of 56 cm (22 inches), width of 45 cm (18 inches) and depth of 25 cm (10 inches) including all handles, side pockets, wheels etc.

However, many airlines have smaller size restrictions, especially low-cost and regional operators, and some will also enforce weight limits aggressively. Regional carriers operating smaller planes under the aegis of a larger carrier will generally offer "valet" services for hand luggage, meaning that you leave it at the air stairs or on the jetway as you board the aircraft, and it is returned to you as you deplane. However, this still means that your luggage is going in the hold, so you should lock it and remove valuable or breakable items as if it were being checked in. Also, sometimes valet-checked luggage is returned at the carousel, not at the exit of the plane - check this ahead of time, as it may affect your planning if you have a connecting flight.

Rule 3: Get the Right Luggage

There are a few different categories of luggage worth owning if you travel at all frequently. Your mileage may vary, but these cover most bases.

The Overnight Bag

If you have to spend just one night on the road, it's not worth hauling an entire suitcase just for a change of shirt & underwear. My personal choice is a rolling laptop bag with a separate pocket, just big enough for that change of clothes. You can wear the same suit both days, and switch it up by changing ties and other accessories.

I own a previous generation of this bag from Tumi, and I love it. Others do the same sort of thing with a backpack, but then you need somewhere to put your clean clothes so they don't end up in a ball at the bottom of the backpack with a packet of last year's mints and a cable for a device you no longer own. However, I'm aware that I am prejudiced against the suit+backpack look, so if that combo works for you, more power to you.

The Basic Roll-Aboard

This is your generalist piece of luggage, that can serve for trips from two nights up to a week or so. Rigid versus soft sides is a matter of personal choice, but four-wheel "spinners" are a must. This way you can squeeze the case sideways through much tighter spaces, such as for instance the aisle of an aeroplane, instead of picking it up and banging all the other passengers on the elbows, knees, or even head.

My personal choice is also from Tumi, but most manufacturers have something like this. Just one thing to point out is that there are two sizes of carry-on, and it pays to get the smaller one, sometimes labelled as the "international" version. There's no point in being able to fit more in your bag if you end up having to check it most of the time anyway.

The Duffel Bag (Optional)

A duffel bag is a very fashionable choice right now, but it's not super practical. It may look good in the departure lounge, but you may regret it if you have to hump the thing the length of FRA for your connection, or on that dreaded last day of the trip when you've already checked out of your hotel but your flight isn't for hours yet.

Why even list it then?

Because there are cases where a duffel bag is actually a sensible choice, and that is when you have chosen the correct one. I favour this one by North Face:

The point of this bag is that you can use it either as a rucksack or as a duffel bag. When it's a duffel it's rugged and waterproof - plonk it down on the snow if you need to, and your clothes will stay dry. Those loops on the outside are also super useful for hanging stuff onto with carabiners. The point of the rucksack straps, though, is that you can put it on your back and leave your arms free, which is especially great if you're travelling with children.

Duffel bags are subject to the same limitation that I was complaining about above with regard to backpacks: the contents tend to wind up all tangled together in a crumpled ball. To prevent that, I have a combination of packing cubes from Muji and Eagle Creek, but still, you're not going to pack a suit in one of these; it's strictly for informal trips or as your carry-on for family vacations. That's where this thing shines, in fact. You're almost certainly not going to get away with doing a family vacation with just carry-ons until your kids are quite a bit older than mine are now, so you're going to have to check some luggage - but pack one or two days' worth of essentials in one of these so that you are never separated from food, favourite toys, nappies, change of clothing, spare batteries, drawing materials, and all the other hundred and and one essentials of enjoying a family vacation and not getting lynched by the other passengers.

The Suit Bag (Optional)

This one definitely depends on your lifestyle and wardrobe - but if you need to wear a suit or similar formalwear at your destination more than rarely, you need one of these. Basically, the idea is that instead of folding your clothes, you just place them in the case, still on their hangers, secure them in place with a strap or two, zip up the case and off you go. At your destination, you can either hang the entire bag if you have a suitably sturdy wardrobe, or simply remove your clothes, still on their hangers, and hang them right up. While they won't be 100% fresh, they will be far less rumpled than if you'd folded them.

Once again, my own choice is from Tumi - the previous generation of this bag:

The thing is genius. If it does have one flaw, it's that it is pretty wide, so it can be unwieldy in airports where they have those width restrictions to stop people going through with their trolleys. It is still within IATA dimensions, although I've had to argue the point with gate agents a couple of times. Overall, the results are worth it though.

Other Notes

Hot / Cold

The hardest trips to plan for are the ones that involve multiple climates. The default is to take your outer coat on the plane, but that gets old fast, especially if you have connections & layovers. Instead, see if you can get away with a lighter-weight outer coat that you can pack, and layer it. One combo that works well is a sweater or a vest worn under a blazer, and a raincoat or similar over the top.

Cotton-cashmere blends are great for this purpose, and are also super-comfortable for travelling in - although I generally prefer to use older sweaters for this role, as sleeping in them is tough on their appearance.

You also need to plan your footwear, both socks and shoes. This may be specific to me, but I can deal with temperature extremes in other parts of my body much better than I can when they affect my feet. This is why I don't travel in emergency exit rows, despite the extra legroom - that, and not being able to keep my laptop bag handy.

Tetris

The enemy of efficient packing is space. No, this isn't some Zen paradox! Just don't leave empty spaces around your items, but instead figure out how to slot everything together like a jigsaw. This will both let you fit more stuff into the suitcase, and at the same time make sure that your stuff doesn't shift around in transit.

Some experience with Tetris is recommended.

Recap

Planning is everything. Get the right luggage - or an assortment of luggage to suit the types of trips you take. Sort out all your toiletries, medications, and so on. Ideally have travel-size duplicates that can just stay in their transparent baggie, ready to be dropped in your bag.

Same thing with clothes: plan an assortment of clothes that can be combined in different ways and that will cover the entire spectrum of expected climates on your trip. Yes, this can go wrong - ask me sometime about landing in New York in February, straight from the Caribbean, and being informed that my luggage had not made it.… This is why you also need a plan B, C, and several other letters (add more letters when travelling with kids).

With preparation and practice, you can pack in a few minutes, enjoy a stress-free trip, and arrive at your destination far less frazzled than someone who threw everything and the kitchen sink into a piece of luggage the size of the monolith from "2001".

Happy travels!

Update

I had no sooner published this post, than I found this list of "bad travel gear" from The Economist in my RSS queue.

The very first item in the list is:

Luggage with spinning wheels, which don't work well on rough surfaces and cost you precious overhead bin space.

Obviously I disagree with this one! Spinners have the advantage that they can roll in any direction - and if yours doesn't work on rough surfaces, I suggest you bought the wrong one. Mine works just fine on rough surfaces.

Much of the rest is common sense - items like money belts and travel wallets just mark you out as an inexperienced traveller. I do have an inflatable travel pillow because it takes up next to no room when deflated, but I rarely use it and should probably ditch it.

However, this item brought me up short:

Your tablet (unless it is your primary computer). Your phone and your laptop can do anything it can do. You don't need it. Leave it. If you want to read a magazine, buy The Economist in the airport.

Sorry, but this recommendation is even more bone-headed than the "no spinners" one above. Maybe this is true if you're in business class, with plenty of room and in-seat power, but at the back of the plane where we plebes travel, there is rarely enough room to get out even a 13" laptop. A phone on the other hand has a much smaller screen, so while it's probably fine for a short-haul trip, your eyes would get very tired of squinting at the screen for a long trip. Also, my iPad has a case that lets it stand upright so I can watch a film or just rest my arm while reading; my iPhone doesn't, because most of the time that stand would just be useless extra bulk in my pocket. The iPad's case also means that the device can be stood on its edge on the tray together with my dinner, while I'd have to close the laptop to eat - and then I couldn't read or watch my film. If I had bought that paper copy of The Economist - because I hate trees or something - I'd be trying to read it one-handed while eating, a sure-fire recipe for ending up wearing your dinner.

Let's also talk about battery life: I have never completely zero'd the battery on my iPad on even the longest trips - and if it comes to that, I have external battery packs in my bag (note plural). If you watch movies on your laptop, just one will take a huge bite out of your power reserve. If you try to actually use a laptop, things can also go south surprisingly fast. For instance I have one corporate Word template which just makes Word craaazy for some unfathomable reason, and it starts thrashing for the whole time any doc using that template is open.

For once, I can't agree with Gulliver and The Economist. Unless it's a short-haul flight - meaning, an hour's duration or so - you almost certainly want your tablet with you on the trip.

Image by Matthew Wiebe via Unsplash

In case you are new to this Internet thing, there has been a long-running debate about advertising on the web. Basically, what it boils down to is that nobody has really found a way to pay for doing things on the internet at scale that does not involve displaying advertising to users.

While it is true that display advertising involves less friction than a paywall, there are all sorts of problems with the model. One is simply accounting. Most adverts are paid for based on the number of times they are displayed to users. However, on the web, that just means that the URL was invoked - so it is possible to sabotage a rival's advertising campaign by retrieving the URLs for their ads automatically and very rapidly, until the budget for their ad campaign is exceeded. This prevents humans from seeing the ads, while costing the advertiser a chunk of cash.

The other problem is a variation: just because the ad has been retrieved by a web browser that has an actual human customer - or at least a prospective customer - sitting in front of it, does not mean that the ad gets displayed to that customer.

These days, most advertising is fairly benign, but in the early days it was a bit of a Wild West out there, with pop-overs, pop-unders, interstitials, and so on. In those dark times normal ad blockers were insufficient, so I used a Firefox add-on called Greasemonkey to actually rewrite web pages on the fly and remove the worst excesses. Nowadays, apart from a few sites (ahem Forbes ahem) that stick by the interstitial, most of those early techniques have mercifully died out.

However, old habits die hard, plus some people seem to be actively offended by ads - in the content they are getting for free, let's not forget - and so very many people routinely use ad blockers in their web browsers.

There has been some debate about whether this actually constitutes theft, although a German court has recently ruled that it does not meet the legal definition of theft.

Regardless, I feel that it meets my moral definition of theft. Companies put out content with the expectation of being paid for it, so it seems churlish at best for me to enjoy the content but refuse them the chance to make a fraction of a penny from their advertisers off my enjoyment. There is a line, but nowadays I am more likely simply not to visit offending sites than to try to bypass the ads.

Problems do remain with this model, not least becase of the sketchy things that it forces people to do:

The second gross thing is that we've given just one more piece of information about our customers to Facebook, but not in a way that is directly useful to us. Even though we're doing the leg work to build up this dossier on our customers, we don't actually get to look into the file. Only Facebook does. When Facebook eventually goes away, the information is gone. When Facebook becomes more extortionate, the information is gone.

The ad blockers are no better on the moral plane:

In a particularly evil move, some ad blockers have started charging ad networks ransom. Ad blockers work by whitelisting some sites and blacklisting others. One of the most high-profile examples of pay-for-whitelisting is a deal struck back in February between Adblock Plus and Google, Microsoft, Amazon, and Taboola. The price: "Thirty percent of the additional revenues it would make from being unblocked."

In this situation, the best users can do is to try to tread the middle ground, watching their cookies and occasionally clearing them out, without going so far as to block ads programmatically.

The good news is that I am beginning to see new services that appear to be making subscription models work. I don't know whether this is a function of the particular niches they are in and that catch my interest, my age (I'm a leading-edge "digital native", surfing just ahead of obsolescence), or my disposable income and consequent willingness to pay for good content - but the bottom line is that I don't even see banner ads any more, so it's an easy moral decision to make.

Image by [Joshua Earle](http://www.joshuaearlephotography.com/ "

JOSHUA EARLE PHOTOGRAPHY") via Unsplash

It seems that iOS 8.3 changed something in the way multiple keyboards are handled. If you don't know, you can add keyboards to iOS from Settings > General > Keyboard. This is worth doing even if you only type in one language, because it's how you get access to the Emoji keyboard. Enabling multiple keyboards adds a little "globe" key between the numlock and dictation keys:

Simply tap that "globe" key once to switch to the next keyboard in the list, or hold it to see a menu and select the keyboard you want.

The advantage of having multiple keyboards is that it enables predictive text to work in other languages. It also allows you to choose alternative layouts, e.g. AZERTY for French, QWERTZ for German, or QZERTY for Italian - but I find that confuses me more. Luckily, iOS lets you set all keyboards to use QWERTY.

Now, here's the problem. Before 8.3, if you had a primary keyboard (generally corresponding to your locale), you could switch to another language to type some text. The next time you hit the key, as long as it was within a reasonably short period of time, it would switch you back to your default keyboard. This is great for me, as I type mainly in English, but switch to other languages several times a day.

With 8.3 this behaviour has gone, and the "globe" key always switches to the next keyboard in the list.

This change is probably invisible to almost everyone, and only a minor irritant for those few of us who use multiple input languages frequently, but it is surprisingly annoying when you are used to the old way of things.

I can even understand the rationale, as I have seen people get confused by why the switcher would sometimes go to the next keyboard but at other times revert to the default - but the solution there is to give us preference settings to disable the behaviour entirely or change its timeout. I don't even mind if it's turned off by default, as long as I can turn it back on - but that's not the Apple way.

Sigh.

There are theories which seem just intrinsically right when you hear them. Clayton Christensen's famous "disruption theory" is one of these. I was recommended to read "The Innovator's Solution" by a friend of mine who had previously worked directly with Professor Christensen, and it definitely shaped my thinking about the technology business.

When the core business approaches maturity and investors demand new growth, executives develop seemingly sensible strategies to generate it. Although they invest aggressively, their plans fail to create the needed growth fast enough; investors hammer the stock; management is sacked; and Wall Street rewards the new executive team for simply restoring the status quo ante: a profitable but low-growth core business.

In sustaining circumstances—when the race entails making better products that can be sold for more money to attractive customers—we found that incumbents almost always prevail. In disruptive circumstances—when the challenge is to commercialise a simpler, more convenient product that sells for less money and appeals to a new or unattractive customer set—the entrants are likely to beat the incumbents.

Disruption theory explains a lot about many markets, although it is not without its critics. In particular, Jill Lepore caused a minor furore with a piece in the New Yorker entitled The Disruption Machine, in which she accused Professor Christensen of cherry-picking his evidence.

There is one famous exception and objection to disruption theory, and that is Apple. According to the orthodox version of the theory, Apple should have been disintermediated by now by smaller, more agile modularised competitors. Every year seems to bring its candidate as the disruptor of Apple: whether it's the Nexus, or Samsung, or Xiaomi, or whoever. Apple somehow survives them all, and not just survives, but goes from strength to strength.

Why is this?

Ben Thompson wrote about how classic disruption theory applies mainly to enterprise products, where the buyer is not the user, and so "feeds & speeds" that can be mapped on a checklist rule the purchasing process. The buyer is looking for a product that can satisfy some simplified criteria. Beyond that binary fit, the decision is primarily based on price.

Apple emphatically does not fit this model, with its focus on design and the user experience. The buyer is the user, and once their basic criteria are met, they can still be swayed by different user experience and personal preferences - of course, up to the budget they have available or are willing to assign to a piace of electronics. Therefore, Apple continues to be able to command much higher prices for devices that are - on paper - comparable to their modularised competitors. Users return again and again for newer versions of their device, every year or two, and Apple's profit margins are legendary.

Enterprise software had always seemed to be on a much more classic track to disruption, with procurement departments working from Request for Proposal (RfP) documents that generally allow yes/no or at most grading on a short scale, typically from one to four. Recently, though, the market has been changing, and incumbent vendors are being disrupted by offerings which bypass the traditional buyer in Procurement and appeal directly to the end user. One model is open-source, where sufficiently technical users have been able to download and use free software without support from central IT for at least the last fifteen years. The main roadblock to this avenue of disruption was in users' willingness to futz around with graphics drivers or whatever. More recently, a new avenue has opened up, namely software as a service. SaaS offerings require much less technical acumen, and much less effort even when the technical acumen is available. Even users who are able and willing to get their hands dirty only have a limited amount of time available to do so, and are quite happy to hand the responsibility off to someone else.

This is how you get the infamous "shadow IT" - enterprise IT's particular incarnation of disruption theory. However, I do wonder whether enterprise software might not also have exceptions to classic disruption theory. Buyer inertia may prevent modularisation, or at least complete modularisation, from taking hold, or delay it for a long time.

From integrated to modular to orchestrated

Apple cannot easily be replaced by modular competitors, even when those competitors offer lower prices and nominally higher performance, because the overall user experience delivered by those competing devices is inferior. There is an equivalent mechanism in enterprise software - although, unfortunately, it does not often manifest in attractive user interfaces and satisfying interactions. Rather, it is the experience of the buyer which is important.

Many mature, established companies have a "vendor rationalisation" initiative of some sort. Some may even go so far as to have an "Office of Vendor Management" or equivalent. One way of looking at this is the "one throat to choke" school of thought taken to its extreme, but there is something else going on here.

As software becomes more complex, and user requirements more varied, there are fewer and fewer one-stop software packages. Even within a single vendor's offerings, users will need to select multiple packages, many of which will have been developed by different teams or even by different companies acquired by the vendor over the years. Customers are looking for a trade-off between best-of-breed solutions from different vendors or open-source tools that require substantial work to integrate with each other, versus vertically integrated solutions from a single vendor that may not excel in any one area but can deliver on the whole task.

The variable that will drive choice in one direction or the other is the rate of change. If the integration between the best-of-breed packages remains valid, once developed, for a significant length of time, then the modularised, disrupting solutions - whether commercial on-premise, open-source, or SaaS - will win. If on the other hand integration is a constant effort that never fully stabilises, requiring never-ending development to chase a constantly moving target, then the benefits of the pre-integrated solution become more attractive in their turn.

Timing is everything

The twist is in the incentives. Developers of the modularised solutions are in a race with other modularised solutions, hired to do the same job, in Christensen's terminology. The way they keep ahead in the race is by evolving faster, adding more functionality sooner than their competitors. They have no direct incentive to stabilise their solutions. For the same reason, they have no particular incentive to stabilise the interfaces to their solutions, as this makes them more easily replaceable by their competitors (less sticky).

The upshot of all this is that a vertically-integrated company can stay ahead of the curve of disruption by innovating just enough to maintain stability for its users, while supporting a certain speed of evolution. This is the job that their customers hire them to do.

The commercial Unix platforms were displaced by Linux because both followed standards (GNU, Posix, and so on) that made them largely fungible from the point of view of their buyers, once Linux had developed beyond its beginnings. iPhones were not displaced by Android phones because they were not fungible to their buyers.

How not to be fungible

What are the characteristics of enterprise software that can make it non-fungible? Simply put, it comes pre-integrated, both with itself (or rather, between different components of itself) and with everything else. This is why content is so important. An API is not enough to avoid being disrupted; any open-source project worth its salt comes with an API - probably RESTful these days, but the principle is independent of technology. What prevents disruption, making the software "sticky", is content: pre-built integrations, workflows, best practices, and data transformations, that make the software work seamlessly for customers' needs.

Enterprise software needs enterprise-grade content that takes advantage of those integrations. Relying on technical capabilities alone leaves enormous vulnerability to motivated developers and agile start-ups. A would-be enterprise vendor must focus on what it can do to prevent disruption. Agility - chasing the bleeding edge - is not the job that buyers hire it to do.

However, there ain't no such thing as a free lunch: those integrations have to keep up with those other fast-moving targets. There is no "we support these two products, and we will add the third-placed platform with the next release of our software in a year's time"¹. The integrations themselves have to evolve, and do so in a way that is both backwards-compatible (you can't break everything your users have built every time you upgrade) and fast-moving (you have to keep up with where your users are going, whether to new platforms, or to new versions of existing platforms²).

The bottom line

All of this represents yet another level of abstraction. The competition moves to a different layer of the stack: the content and integrations. Enterprise vendors who refuse to follow along are simply ceding to their competition; users - and importantly, buyers - are there already.

Hardware got commoditised, then operating systems got commoditised, and now it's the next layer up. It's not what you have under the hood, it's what you do with it - and both people and enterprises will buy the tool that enables them to get the most done.