Monthly archives of “March 2007

A bright idea?

UPDATE: See this more recent post for information and photos of how to get a 2-pin bulb to fit in a BC3 fitting.

This may well be the example which involves the most different ‘architecture of control’ issues so far – by a long way. It is a complex case with a number of aspects, intentions and effects to consider. My mind isn’t made up on the rights and wrongs of this: it’s certainly an architecture of control, it’s certainly devious and it’s certainly a case of introducing a razor-blade model (product lock-in) into a field where there was previously none; it will also end up costing many consumers more money, yet it’s founded in an attempt to ‘encourage’/force more environmentally friendly behaviour.

A couple of weeks ago, George Preston let me know about Eaton MEM BC3 light bulbs and fittings. These are compact fluorescent lamps (CFLs or ‘energy-saving’ bulbs) which have their own kind of three-pronged bayonet connector (left), as opposed to the standard two-pronged bayonet (right):

BC3 lamp, photo by George Preston
BC3 lamp, photo by George Preston
BC3 fitting - image from MEMLITE brochure
BC3 CFL and standard bayonet CFL compared, and a BC3 fitting. Upper two photos by George Preston; lower photo from BC3 brochure [PDF].

Notice those three prongs are irregularly spaced. A normal bayonet bulb won’t fit in a BC3 fitting, and a BC3 bulb won’t fit in a normal bayonet fitting.

What’s the rationale behind this?

From Approved Document L1 [PDF], an amendment to the UK Building Regulations, which came into force in April 2002 (applying to new-build houses):

1.54 Reasonable provision should be made for dwelling occupiers to obtain the benefits of efficient lighting. A way of showing compliance with the requirement would be to provide at a reasonable number of locations, where lighting can be expected to have most use, fixed lighting (comprising either basic lighting outlets or complete luminaires) that only take lamps having a luminous efficacy greater than 40 lumens per circuit-watt. Circuit-watts means the power consumed in lighting circuits by lamps and their associated control gear and power factor correction equipment. Examples of lamps that achieve this efficacy include fluorescent tubes and compact fluorescent lamps (not GLS tungsten lamps with bayonet cap or Edison screw bases).

The idea is, then, that since ‘normal’ bayonet fittings can take normal tungsten incandescent filament bulbs as well as normal CFLs – something which has of course driven the more widespread adoption of CFLs – there is the likelihood/possibility that householders might replace any pre-installed CFLs with filament bulbs, for whatever reason (the usual reasons are the colour of the light, the aesthetic appearance of the bulbs, and the warm-up time). To prevent this possibility, a new type of light fitting and associated CFL cap design were required which were uniquely compatible, so that anyone with this kind of fitting would have to fit bulbs with the new cap design, which would only be available on CFLs.

(Note that the same objective could have been achieved by fitting these rooms solely with fittings for commonly available standard linear fluorescent tubes, i.e. strip lights.)

So, Eaton’s MEM 250 division created the BC3 (bayonet-cap-3?) range, being nominated for an Electrical Product Award for Contribution Towards Energy Saving in the process.

What’s interesting is that as well as complete BC3 CFLs and BC3 fittings, the BC3 range includes BC3 base units (with the ballast and control electronics in them) into which a four-pin CFL tube can be plugged:

BC3 lamp unit, from BC3 base unit, from
Left: A tube unit with four pin connector; Right: A BC3 base unit (including ballast) to allow the tube to be attached. Images from Ethical Products Direct.

This allows the tube to be replaced independently of the electronics – thus saving resources – but does not appear to be the focus of the BC3 system. (Just a thought: if more new houses were pre-fitted with these base units, or simply standard 2-pin bayonet base units, within the light fittings, so that a householder would simply go out and replace the tube rather than the whole lot, similarly to the linear fluorescent tube suggestion above, would it not have made for a more environmentally friendly solution?)

Some interesting claims are being made for the BC3 system. Somehow the idea of forcing the householder to buy one particular brand of CFL has been transmuted into a misguided suggestion that the BC3 system actually makes the houses more energy efficient – e.g., from a housing association magazine [PDF] in Wiltshire:

Residents in some of Westlea’s newer homes will know that we now fit special three-way bayonet lamp fittings as one way to make the property more energy efficient. Although the ‘BC3 eco bulbs’ needed for these lamp fittings are more expensive than ordinary lightbulbs, using them in a ‘standard’ house could save the resident around £100 each year because they use less electricity than ordinary lightbulbs. Some residents have told us they have had difficulty buying the three-pin eco bulbs locally, but we’re pleased to report that the following outlets are able to supply them from £6.35 upwards…

From £6.35 each is a lot of money. Standard ‘Tesco Value’ 2-pin bayonet CFLs started at 88p each (Tesco, Egham, Surrey) the last time I looked – that’s especially cheap, and they were only 11W, but 15W units are commonly available from about £2 – £3. Searching Froogle shows that BC3 bulbs start from around £10. Even Ethical Products Direct, to whom Eaton MEM’s own website directs visitors wanting to buy BC3 bulbs, charges £9.36 for the cheapest complete BC3 unit.

This is a lot of money for something which provides the householder with exactly the same function as a standard CFL a quarter the price. (It’s not as if the BC3 bulbs last much longer, for example, or are more efficient. They just have a non-standard fitting and are only supplied by one manufacturer.) In fact, one might suggest that standard CFLs offer the householder more benefit, since they can be swapped around, fitted all over the place, even fitted to replace incandescent filament bulbs in standard fittings, should someone – shock – actually want to choose a CFL without being forced into doing so.

The housing association quote above demonstrates an important point about the use of BC3s. Many householders’ first encounter with them will be when they notice a CFL going dim or actually failing, or want to increase the light levels in a room, and find that they have to spend much more than they were expecting to spend on a CFL anyway. George’s story demonstrates this well:

We have recently moved into a new flat which is part of a modern development in London. A few lightbulbs needed replacing when we moved in, so I went out and bought some (they’re all energy-efficient ones so I bought the same to replace them with). But oddly, none of them would fit in the fittings. I was under the impression that there were just Bayonet and Screw Cap fittings? These fittings were bayonet, but needed three, irregularly-spaced pins instead of the standard two.

I’m no stranger to energy efficiency, and it wouldn’t be so annoying were it not for the fact that the bulb I had bought as a
replacement was an energy-efficient type anyway, but it seems illogical and a shame that properietary fitting sizes have been introduced into something that has always been so simple – choosing a lightbulb.

(Equally, there is the problem of actually getting hold of BC3 bulbs. I went to the enormous B & Q in Slough on Sunday and couldn’t see any on the shelves. While the 8,000 hour lifetime may mean that there’s not a massive demand for them yet from the public, ordering online and waiting for delivery is not really a great option when a light bulb fails. It often causes inconvenience, and can be dangerous – until Incluminate‘s a production reality (!), the best option is to keep spare bulbs in the cupboard. But if you don’t realise that you need to keep special BC3 bulbs, and that these aren’t available from every corner shop or even every massive DIY store, this is going to be extremely inconvenient. The BC3 brochure does mention a “householder card… which can be left with the homeowner highlighting the ‘energy saving’ aspects of their new home” but how many people will remember to stock up on BC3 bulbs as a result?)

Anyway, I think the main issues are:

  • Razor-blade model: monopoly on fitting type means higher prices can be charged for same function, consumers locked in
  • Non-standard fitting likely to cause significant inconvenience to householders
  • But:

  • System does force householders to use ‘energy saving’ bulbs*
  • The BC3 range is also made in the UK, which aside from actually supporting local jobs, means that the units are not transported from China as, say, Tesco Value CFLs are. That saves on transportation energy, at least, and while – looking briefly – I couldn’t find a patent for the BC3 system, I presume Eaton have it protected somehow, otherwise there would surely be cheaper BC3-compatible bulbs available.

    (Another thought is what other proprietary systems – if any – have manufacturers evolved to meet the regulations in part L1? Are there lower-profile rival systems with their own fitting and cap designs? What would the implications be if a particular type were no longer available a few years down the line?)


    Overall, this is a clever commercial attempt to respond to a governmental decision made with environmental protection in mind, and as such probably ought to be filed along with optimum lifetime products as something where the intention is to benefit society as well as benefit the manufacturer, at the expense of additionally inconveniencing the user. I feel focusing on a system of built-in base units, with readily available standard replacement tubes (either CFLs or linear fluorescent format) would have been more user-friendly as well as reducing the amount of electronics needlessly thrown away, but it would not have permitted a razor-blade model to the same extent.

    It will be interesting to see how the BC3 story develops in the years ahead: will they become commonly available, and how high will public awareness be? There will probably be many more similar products and systems in the next few years using technology to enforce government policy, particularly in an environmental context, and the Eaton MEM BC3 will be an important case study.

    *Of course, there’s a lot that ought to be said about the real merits of a large-scale shift to ‘energy saving’ bulbs, particularly in relation to Australia’s decision to phase out incandescent filament bulbs entirely, the European Lamp Companies’ Federation’s focus on the same, Gordon Brown’s announcement on this, and campaigns such as Ban The Bulb.

    As a designer and engineer, I would suggest that in cold climates, 100W from an incandescent filament bulb means simply that 100 joules per second of heat is going into my room (probably wasting another 200 joules per second at the power station, but that’s another matter). Light bulbs do heat our homes. If we lose 80W from the light bulb, the heating will probably get turned up by 80W instead. Better insulation, so that that heat isn’t lost, may well turn out to be just as good, or better, than mass-replacement of thousands of millions of light bulbs with CFLs requiring significantly more resources to manufacture (and dispose of). Those electronics in the base don’t come from nowhere, and are likely to outlast the fluorescent tube: hence why the idea of replaceable tubes is much more sensible than throwing away and replacing the base unit each time as well. But the bandwagon’s set off and with heavyweight government and heavyweight manufacturers on board, it’s got a lot of momentum…

    Anti-user seating in Oxford

    Anti-user seating in Oxford
    Anti-user seating in Oxford
    Anti-user seating in Oxford
    Anti-user seating in Oxford
    Top two photos: A bench on Cornmarket Street, Oxford; Lower two photos: A bus stop seat perch on Castle Street.

    While from a very narrow specification point-of-view ‘they do their job’, what utter contempt for users these two seating examples demonstrate! The benches on Cornmarket Street are clearly intended to prevent anyone lying down on them (armrests, small radius of curvature) or indeed sitting for very long at all in comfort (height off the ground, vertical backrest, small radius of curvature). Why? Why despise the public so much?

    The designer must have been given a specification requiring all the above features: I can’t believe they just arose out of aesthetic or manufacturing considerations. That bench has been engineered to restrict, control and discipline users. Was it really necessary? Does forcing the homeless to lie on the ground instead, or preventing people sitting comfortably and watching the world go by really ‘solve’ any problems?

    The bus stop perch – in this particular location intended at least partially for Park & Ride users – is perhaps even worse. It’s angled such that a young child couldn’t easily sit on it without sliding off. An adult has to stretch out his or her legs just to perch. A parent couldn’t sit next to a young child. A shopper would have to put down his or her bags on the ground, since they’d slide off the perch. My girlfriend and I couldn’t rest our drinks on the bench next to us; we had to put them on the ground. OK, that’s not much of a hardship, but it’s just frustrating design, intended to serve objectives other than the users’ benefit or convenience.

    You wouldn’t want to wait any longer than necessary at that bus stop. If you were making the decision about whether to drive into Oxford or take the bus to go shopping (assuming cycling not to be an option for this) the unattractiveness of perching at an angle for 15 minutes on that mean strip of perforated sheet would begin to weigh heavily against the public transport option. Sure, you might end up sitting in your car in heavy traffic for 15 minutes, but it’s your car. The seats are comfortable, it’s warm, and you can shape and adjust the environment to suit you.

    I don’t want to go off on one here about solving (or easing) Britain’s transport problems, but I do feel that this kind of situation embodies some of the very important issues. By making bus users feel unwanted – despised even – you don’t enhance the image or desirability of the mode of travel. Little details such as this can make a huge difference to perceptions. The buses themselves are great, but if the experience of using the service seems to demonstrate contempt for the user, the user may develop contempt for the service.

    Japan may have some of the most explicitly user-unfriendly public benches we’ve come across so far, but there’s also something rather disturbing about the sheer blandness of the bench implementations shown above. Their starkness embodies the thinking behind the design: all possible interaction methods to be reduced down to one sole, pre-defined utility function, with the user not permitted to do anything outside that intentionally myopic definition.

    (Incidentally, to be fair, there were some lower seats with horizontal platforms on the other side of the bench in Cornmarket Street. They still had armrests to prevent lying down (or even sitting close to someone), but were not as awful as the curved ones.)

    Plug: Wilson Brothers’ blog

    Nike bike by the Wilson Brothers

    Bit of a design-related plug: London designer/maker Ben Wilson (with whom I’m currently working on a project for Sir Clive Sinclair) and his brothers, Oscar and Luke, have just launched their own collaborative photo blog, which I helped set up using, a mildly modified Sandbox theme and automatic email-to-blog (via Flickr) to allow the simplest method of photoblogging I could think of.

    Between them the Wilson brothers take a lot of great photos of interesting and inspirational design, places, vehicles and people, as well as chronicling their own projects, and I think the blog’s going to get quite a bit of attention. The blog’s starting with a look at the building of a one-off bike commissioned by Nike (shown above), with some extraordinary detailing (cut leather decals and intricate stainless steel lugs).

    West Coast code meets Far East code

    Thanks to Mr Person at Text Savvy, I’ve just learned that this blog is blocked in China:

    Images from the Great Firewall of China test.

    I don’t know if that’s good or bad. From a censorship point of view, it’s bad, but it’s certainly interesting to be able to say that the blog’s blocked in China, even if it’s just for a rather prosaic reason (using WordPress?) as Mr Person suggests, and not the incendiary demagoguery contained within these posts and comments.

    (Additionally interesting is that as the whole of seems to be blocked, I might not have any more of my portfolio items appearing on Chinese design sites. One site even had me listed alongside Karim Rashid for a while, which was odd and flattering, perhaps, though I don’t think he’ll be losing sleep over it!)

    What I’ve learned so far as a freelance designer/engineer/maker: Part 1

    The sign on the door

    This is the first in a series of essays where I’ll try to look at some of the realities of working freelance in this field; I hope these will be interesting and possibly useful to others contemplating this kind of work. Please note, these are only my own musings and ramblings, written mostly on train journeys across North London, and I might look back on them with embarrassment and disagreement.

    At the moment, I’m a freelance designer/engineer/maker. What that means is hard to define. There are no obvious boundaries: I’ve said ‘Yes’ to almost every project, mostly out of necessity but partly out of trying to determine what I’m any good at. In practice that means that in the last year-and-a-bit I’ve worked on some diverse stuff, from developing ultra-lightweight bikes to designing novelty packaging, from researching multinationals’ brand architectures to doing toothed belt calculations for gearboxes. I’ve tested radio-controlled things in the Thames looking across at Windsor Castle, and grappled with CSS while sitting in an abandoned factory in Dalston. I’ve hand-lettered sandwich shop menu blackboards and sprayed T-shirts with the logo of a new telemetry spin-out company. There’s mechanical engineering in there, some graphics, some electronics, prototype building, even copywriting.

    What it’s shown me is that a jack-of-all-trades is not necessarily master of none, but unlikely to be any more than master of some, few in fact. And the main reasons for that — so far as I can tell — are time and money.


    If every project is different, you pretty much have to start by spending time simply finding out what you’re doing, what the precedents are in that field, what important things you need to know, even what equipment you’ll need to do the job properly. Some clients tend to assume that anyone ‘technical’ can fix (or indeed design) absolutely anything involving engineering materials, electronics, computers, etc, and while to some extent I don’t think that’s untrue, given experience, it’s probably not the best policy always to say ‘I’ll give it a go’. But you do need to test your limits before you can know them.

    Back to the point: if you have to spend a significant amount of time on each project learning about the field, each project is going to take you longer than it would for someone who already knows what’s what. And you will make mistakes, of course.


    What the above implies is that, as it’s going to take you longer, you’re going to have to work out how to charge. Should the client pay for your learning process? How fair is that?

    One point of view would say that no, you’ve created an (intangible) asset for yourself, and the client should only pay for your time once you know what you’re doing. The other point of view says that acquisition of knowledge is a prerequisite of being able to deliver what the client wants. Just as you charge for the acquisition of materials, so should you charge for the acquisition of knowledge. I think the answer probably lies somewhere in between, but it’s difficult for a freelance person — reliant on a sporadic income anyway — to ‘write off’ days as ‘knowledge acquisition’. If you have zero income (and maybe some expenditure) for those days, then you’re going to have to budget for that somehow, and that’s something that’s difficult to plan.

    A second major point regarding money is that, well, the client wants to spend as little as possible. Why has he/she/it employed you, a freelance individual with (probably) few facilities other than your brain and your hands, rather than a ‘proper’ design consultancy? Unless the client genuinely thinks you are wonderful, or are likely to come up with stunning insights or innovation which someone else wouldn’t, the reason is probably because you’re cheap, or the client thinks you’ll be cheap (‘Because you’re young, and have lower overheads, right?’).

    Wexelblat’s Scheduling Algorithm

    But — the client also wants you to be good. So you have to be good and cheap. And on a smaller budget, and with less expertise and experience to call on than an established consultancy. How are you going to do it?

    When I was working for a couple of weeks at a well-known design consultancy in London, two experienced freelance designers, David Baird and Simon May were also working on (more important aspects of) the same project. One morning, one of them (I can’t remember if it was David or Simon) drew out on his sketchpad, this diagram…

    Wexelblat's scheduling algorithm: fast, cheap, good: choose two

    …and said ‘You can have 2 out of 3. It’s either good and fast (and not cheap), good and cheap (and not fast) or fast and cheap (and not good). That’s what I try to tell clients.’

    This stuck with me at the back of my mind; I’ve since found out it’s (sometimes) attributed as Wexelblat’s Scheduling Algorithm (presumably after Richard Wexelblat?), though also apparently an ‘old designer’s adage’ (Jason Kottke) and an ‘old Hollywood maxim‘. The impossible triangle used to illustrate it here is cleverer than what I’ve drawn above, but the principle is the same. (As with so many principles and maxims popularised through software development, it also seems to apply very well to design and physical product development.)

    As we’ve seen, the client wants a project to be good and cheap. Hence, if Wexelblat is true, it’ll be slow, even if some of that slowness is accounted for by knowledge acquisition, and mistakes. But if you’re charging for that time, you’re incurring costs in the process, which tends to counter the ‘cheap’ aspect of the project. So, there’s an inherent difficulty with applying Wexelblat to jobs with a significant learning curve. If your costs are proportional to the time you spend, you can’t be cheap without also being fast, and bad (since you possibly don’t even know what you’re doing). For the inexperienced, cheap and fast and bad is possible, but good implies not fast and not so cheap unless — as we considered earlier — you’re willing/able to write off your learning time.


    If the above sounds negative, I don’t mean it to. It’s exciting working on new things and building up expertise, but when clients’ primary reason for choosing you in the first place may be cheapness, you’re going to have something of a difficult compromise and balancing act on your hands, just in terms of scheduling your work and budget, let alone the specific challenges of the project in question. It might mean that your definition of ‘1 day’s work’ slowly seeps into becoming ‘7.30 am to 2 am’ just in order to get everything done in the same number of days you promised, and for the same cost. That’s fun for a while, but gets pretty tiring for those around you even before you get fed up.

    An implication of all that is that to be competing on price alone can be a stressful game, especially when having to do so simply to get enough work means that you have a lot of learning to do for every project. It’s something of a positive feedback loop, a vicious circle. But, if you can build up enough experience in a particular field, and are able to use knowledge acquired (or problems solved) on a previous project, you have the start of something more edifying. You may still be able to compete on price, but you can now be cheap, faster and better, since you know what you’re doing. And, slowly, gradually, you might even be able to specialise in a certain field, no longer jack-of-all-trades, but actually mastering something.