Coming down, fast and floppy, from the exams. The week since the last (an account of which I still owe you) I’ve spent oscillating between nervous energy and nervous prostration.
There is so much I want to get done over this summer, but for now I’m mostly doing calm, therapeutic tasks. As I put it in a mail to a friend earlier, I’m reassembling myself. Plunging sinks. Designing CD covers. Configuring Linux. Oddly varied, when I think about it. Began summer work on my mother’s house. Planted some flowers – blue pansies. Cut the grass for the first time this year. Felt more like shearing a green sheep. And finally started to varnish the window frames, a job that’s been looking accusingly at me almost since last summer. It’s interesting, I’ve never varnished something before. Except the truth of course.
The Varnished Truth wouldn’t be a bad tag line for this blog, come to think of it.
Another exam this morning. Christ what a paper. Answer three questions out of four; was going to be out of five but they had to cancel a lecture or two so they curtailed our choice to compensate… Which meant that being weak in even one area was a big risk.
And I was weak in one. This paper was Information System Innovation, a strange mix of investment decision-making, Intellectual Property law, and Open Source idealism. At all costs I wanted to avoid a question on business metrics, the tedium of which makes my brain cry.
I got lucky. My favourite area – Open Source Software – came up in two questions. If anyone on the course had been trying to avoid Open Source on the other hand, they were pretty much stuffed and mounted. And this after they told us explicitly that there would be no overlaps.
My only real problem with the paper was that there wasn’t time to say all I wanted to say. So strange to be answering questions about the likes of Richard Stallman and Linus Torvalds, people who before this seemed more like figures out of folklore. Weirder still to think that when I graduated with my primary degree, none of the stuff on this course had happened yet.
So despite the stress I actually enjoyed the exam. This may not be a good sign, as it means I managed to go on at some length about things I have opinions on. Apple versus Samsung, Menlo Park versus Xerox PARC, IP in an age of 3D printing. Did they even want opinion? Did I show I was fully engaged with the material, or rave about stuff that was only tangentially related? Essentially, I can only have done either a brilliant or a disastrous paper.
Where did Linux come from? Strange as it may seem, its roots extend back to ’60s Counterculture. Not a lot of computers in the Haight-Ashbury of course. In those times the only place a young person was likely to access hardware was at university, and it was on campuses that experimentation with drugs and social non-conformity met the sort of person who doesn’t have a lot of friends but is really good at mathematics. Strange things developed out of this cross-fertilisation. Like, to a large extent, the Internet. It was the beginning of Hacker culture.
In this period one of the leading computer operating systems, on campuses and increasingly in industry, was Unix. It was interestingly designed and well-suited to the networked style of computing that was beginning to emerge, seeming almost a little anarchistic in itself. But it was still an expensive, licensed corporate product. A few brave (and possibly slightly high) young souls decided that, hey, they were programmers. They could make something just as good themselves.
The name of this project was GNU. (Standing for GNU is Not Unix. What else?) Like Unix, GNU was designed not as a single giant program but as a whole bunch of little ones, each with its task to perform. A lot of progress was made, but the project long lacked its most vital component: The one central program that organises all the others, known as the kernel. Until, that is, 21-year-old Finnish student Linus Torvalds created one for his own amusement. When the two projects were put together, a complete operating system was born. Purists to this day refer to it as “GNU/Linux”, but plain Linux does for the rest of us.
Don’t make the mistake though of thinking of Linux as an inferior imitation of Unix made by hippies. It’s true it was modelled closely on Unix. In one sense it is Unix; its commands and structures are much the same and a person who knows one can use the other. The difference is that Linux is devoid of any patented or proprietary technology, and so can be copied, changed and distributed freely. This openness has allowed countless people to improve the code – everyone from oddball geniuses just showing off to giant corporations motivated by the bottom line. The only rule is that if you make modifications you must give your work back to the community. This Open Source philosophy has allowed Linux to mushroom in capabilities and refinement, leaving the Unix it once emulated far behind.
This freedom has also led to the huge number of “distros”, as they are called. Linux comes in several major versions, and almost countless minor. Though let’s be careful to be clear about this – they are not different operating systems in the sense that Windows and Mac OS are different. They’re all broadly compatible, the differences reflecting variations more in philosophy than technology.
Nonetheless the sheer breadth of choice may be off-putting at first. Don’t let it worry you, the day is not long off when you’ll believe passionately that one of them is far better than all the others. But that doesn’t matter now (and to be honest, it won’t matter a hell of a lot then either) – what we care about is where to start.
That’s all very well, but why would you – an ordinary person with no particular ideological bent or business need – want to use Linux? Obviously if you’re reading this you’ve already got a perfectly good computing device of some sort. It will have an operating system from Microsoft or Apple or Google that you’ve spent time – perhaps years – getting to know. You may have spent a lot of money on software that won’t work with anything else. Why would you even dream of starting over with a whole new system?
I admit, this applied to me too. Now and again I would install Linux, marvel for a while at how all that great stuff was available for free, and immediately go back to the system I paid money for. Not because it was better, but because I knew it better. This catch-22 of not using Linux because you don’t know it and not knowing it because you don’t use it could go on indefinitely, always keeping you from taking that last step over the threshold. Unless and until a situation arises where nothing but Linux will do. And this is what happened to me recently.
Twice in fact.
The first case was a family member who’d acquired a PC with no working hard drive. He could’ve bought a copy of Windows for about €100. But why? He didn’t need Windows in particular, hadn’t spent years learning its little ways. If he was going to get to know one system, it might as well be the one that wouldn’t keep asking for money. On top of this his main reason for getting the computer was to go online, and for that Linux could not come more highly recommended. Viruses that attack it are too rare to seriously worry about, and it is designed in such a way that if one did get on it could do little harm. So we resolved to set him up with Linux.
And there was my own case. As I was telling you earlier, I recently built a system with more memory in it than you could conceivably shake a stick at – 16GB. However, the ordinary 32-bit version of Windows can’t make use of anything like that much. Just as bigger cities need longer phone numbers, you need a modern 64-bit operating system if you want to call up a serious amount of memory.
And here’s an annoying thing, there is no Windows upgrade path to the 64-bit version. So adding RAM can mean you have to buy a whole new license. For about €100.
Or you give Linux a go, and never pay for software again.
Hmm.
So there are people in some quite ordinary situations who could save considerable money by using Linux. And needless to say, it has other advantages apart from low cost and security. It’s also the most customisable, flexible system. There’s so much sheer choice in fact that it can seem a little intimidating at first, so next time out I’ll talk about where to begin.
Linux – it’s like a shark or an iceberg. Most of it’s below the surface but it’s moving fast and you really ought to know about it.
The majority of people, if they’re aware of it at all, probably think of Linux as the non-commercial alternative to Windows or Mac favoured by people who use computers less to get things done than because they actually enjoy it. Which is a shame in a way, as it gives a wholly wrong impression of its significance. Linux is so much more than an operating system for nerds. Indeed you probably use it yourself, every day. Each time you visit a website the chances are good that you’re talking to a computer running Linux. Smart devices in your home like satellite boxes and DVRs, even TVs now, use Linux. If you have an Android phone, that’s based on Linux. Governments are adopting it, and it is far and away the favourite operating system for the world’s most powerful supercomputers.
So it should be surprising that on the workplace desktop – still the biggest, most visible, and most lucrative computer market sector – it runs a very distant third. How come?
Counterintuitively, because it’s free. After all the PC is not ruled by Windows and Mac OS because they’re cheap. Rather it’s because people can make money out of them. Huge ecosystems of supporting industries have grown around these computing platforms – software, hardware, services, training, maintenance, publishing – in no small part because there was a key partner there offering support and leadership.
How do you make money out of a system no one owns? With whom do you form a partnership? How can you be sure, when no one’s in charge, that it’s going to develop in a direction that will suit your business? It’s tricky.
Unless of course you take a leadership role yourself. It seems the best way to make money from Open Source Software like Linux is often to step up and be that key player. This is what Google did with Android, making it the world’s most popular phone OS. Or Red Hat, whose version of Linux is one of the most successful systems in the server sector. Each brought their own business model; Google of course is ultimately selling advertising, Red Hat its expertise and support.
If anyone is going to turn Linux into a household – and an office – name it is surely Canonical. This is the power behind Ubuntu, the most popular and user-friendly desktop version of Linux so far. And their vision doesn’t stop at the desktop – nothing worthy of the name could these days. In recent weeks they’ve launched Ubuntu editions for tablets, phones and TVs. It seems they plan to have devices running their software in every major market sector.
Ubuntu TV – It’s a lot like Windows Media Center except for the giving money to Microsoft part.
It’s an extraordinary ambition, and if they can pull it off then Canonical/Ubuntu will be up there with the big girls, sitting proudly alongside Google, Apple, and Microsoft. But can such a fabulous commercial edifice really be built on open foundations? So much of Linux is being developed by people who work for competing organisations – or who aren’t being paid to do it by anybody.
Indeed community disenchantment may already be starting to show. Ubuntu is no longer flavour of the month. Once hugely popular with the sort of Linux user who doesn’t actually want to reinvent the wheel but just needs something that can be installed and maintained with the minimum of fuss, Ubuntu – and its slightly geekier sisters Kubuntu and Xubuntu – drove all before it. No longer; the flavour now is Mint.
This is a different Linux variant (or ‘distro’, to use the jargon). Indeed it’s very much a variant of Ubuntu, just with Canonical’s more commercial ideas stripped out. (This ‘forking’ is perfectly legal in the OSS world – in fact it’s the whole idea. Ubuntu itself is based on the well-respected Debian distro.) Mint’s popularity though was given a huge boost when Canonical introduced their ironically-named Unity interface.
Some of the resentment of this was silly. Unlike Windows or Mac where the graphical user interface is part and parcel of the system, much of the beauty of Linux is that you can choose – even create – your own. For some however it’s a cause. There have long been two main Linux desktop camps: Gnome and KDE. Ubuntu had been on the Gnome side, so its defection – to a third camp of its own invention yet – was seen by many as betrayal.
More seriously though, Unity is very clearly a touch-orientated interface. As the name suggests, it’s meant to be similar on all types of device. Rather like Windows 8 this makes it less efficient – or put it another way, more annoying – for users stuck with an old-fashioned mouse. And as Linux tablets barely even exist yet, that means pretty much all of them. For the first time, Canonical were allowing their commercial vision to degrade the user experience.
But that was as nothing compared to the next change. A feature of Unity is that you can find a file or application by typing its name in a search box. If what you type isn’t on the computer, the newest version of Ubuntu continues the search on the Web – specifically, to Amazon.com.
“Crochet Patterns not found. Do you want to purchase Crotchless Pants?”
This unasked-for advertising feels a bit like an invasion of privacy. The building of it right into the operating system feels a lot like a kick in the teeth to the non-commercial ethos that engendered Open Source.
Making money in itself is not the problem. Google, Red Hat, even that old devil IBM make a lot of money out of Linux. Where I think Canonical sail close to the wind is in identifying themselves with Linux more closely than any company before. Independent computing creatives will resent it deeply if they come to be perceived as dupes – or worse, minions – of a commercial giant.
There are two questions here really. The first it whether Canonical/Ubuntu can maintain the goodwill of the wider Open Source Software community. The second is whether they can realise their vision without it. Perhaps they can, but I think it would be a minor tragedy if they did.
One thing over which there’s no question though: Open Source Software can continue without Ubuntu.
The window of McCambridge’s is one of the great places in Galway to have coffee. Looking onto the main shopping thoroughfare, it combines all that’s best about walking around town with all that’s best about sitting inside not doing that.
With our weather – and the last time I was here I watched a wooden forklift pallet being blown down the road – it’s a priceless resource.
The name of that thoroughfare by the way is Shop Street. I’ve always liked the excessive literalness of that. The adjoining High Street meanwhile is full of pubs. All we really need is for the banks to be down Arsehole Avenue.
But I must stop avoiding the issue, I’m here to apologise. This has been one of the longest breaks I’ve ever taken from writing here. What siren has lured me away with her haunted song? I’ll tell you honestly. Flagrant geekery. Part of the time it’s been Java. Not the coffee, the programming language. Part of the time it’s been Linux. All of it, in short, stuff that most people neither understand nor – and here’s the really tricky part – particularly want to understand.
So writing about them in an entertaining way may be a little tricky. But I will give it a go.
You might expect someone who thinks so much and so deeply about mathematics and science to be socially uncomfortable with the normal humans, but Randall Munroe is an engaging and amusing speaker – both in front of a lecture theatre audience and over dinner later.
OK, how many people have I lost already? I need to remember that not everyone is immersed in Geek culture, even now. Randall Munroe is hugely famous; possibly the second most famous person I’ve met. After Colonel Sanders.
But he’s not… TV famous, for want of a better term. Not public property.
I spend much of my social life in a world where everyone but everyone knows who Randall Munroe is. That world is congruent with the other one – you’ll find XKCD fans in every country of every continent – but it is still a shadow world. You’d be forgiven for thinking that Geek culture is mainstream now, what with everyone being on Facebook and all, but really most people just hang around the gateway to the other realm.
And so he can have literally millions of fans – of which I am certainly one – yet still speak at an obscure conference in a small city, and have a quiet dinner after with friendly and only slightly overawed strangers. If you’re going to be famous, I think that may be the way to do it.
Two years ago Nokia’s future was going to be Maemo, a cutting-edge operating system based on Linux with a sleek interface to replace the veteran Symbian OS. The phone that would bring it was the N900. This had the same processor as the rival iPhone 3GS, plus a much better screen and camera, neat slide-out keyboard, a kickstand and great stereo speakers for watching video, and even features the iPhone lacks to this day like real multitasking and a memory card slot.
As Nokia had been and was still the leading maker of smartphones, you might imagine a pent-up market of loyal customers gagging for a device that would show that upstart Apple. Yet when it arrived they stayed away in droves. The N900, very arguably the most advanced and capable phone yet made, was a magnificent, absolute, utter market failure.
What the hell went wrong? More than one thing, clearly. A phone with that much going for it could surely have survived one weakness, maybe even two or three, and still gone on to be a big seller.
Four though is pushing it.
The other week I showed it to an iPhone-using friend. She was interested, but asked – quite innocently – “Why is it so thick?” It’s true, it’s as thick as a brick. Downright chubby. Though its other dimensions are almost identical to the iPhone 3GS, it’s half again as deep. Is that such a bad thing? Thickness is actually a practical advantage in a phone, there’s little chance of cracking this one in your pocket. But such arguments weigh little in the scales of fashion. People have to be slim, therefore our phones have to be slim too – that seems to be about all the logic there is to it.
An even more damaging shortfall though was the strange absence of multitouch. Perhaps Nokia had too little experience with the necessary screen technology, but launching the N900 without a modern interface was like naming a ship “Abandon”. A resistive screen with a stylus for tapping little icons was an interface from obsolescent Windows Mobile and historic Palm devices. Nobody wanted one now.
Nobody, except a few freaks such as myself. Resistive is far better suited to drawing than the capacitive type of screen used for multitouch. Capacitive requires a large contact area with the surface, making precise detail impossible. Plus it lacks any dimension of pressure sensitivity, while resistive screens can be highly responsive to changes in the pen press.
Combined with pressure-aware Linux drawing applications like MyPaint, this ‘outdated’ resistive interface allowed realistic pen-like or brush-like drawing strokes. This made the N900 the best phone ever created for art,¹ a powerful but sensitive digital sketchbook you could carry in a pocket. Many of the cartoons appearing on this blog were done with it – pencilled, inked, coloured, lettered and uploaded without ever seeing paper or PC. You can even edit images with GIMP, a program with capabilities comparable to the full desktop version of Adobe Photoshop.
But the very thing that most endeared it to me was a huge turn-off to the wider public.
Then there was the lack of apps. As Apple were first to realise, shopping is part of the experience now. A phone is nothing without stuff you can buy for it. There are some very good apps available for Maemo – but almost none to buy. Its Open Source Software roots meant that people were keen to contribute useful stuff. With a little tweaking it could even run apps built for desktop Linux. But that actually worked against a market for the Maemo platform. Professional app developers were discouraged by having to compete with free.
And this cultural clash, Open Source on one hand and commerce on the other, created other unforeseen problems. If you’ve got an issue with a community-developed program, to whom do you complain?
You don’t. In the cooperative world of OSS you file a bug report, documenting the issue and the circumstances that produce it. Which is lovely, but customers who’ve paid money for a fancy phone hardly expect to have to help out as well.
Nor do they expect tech support that tells them to open a terminal window and enter Linux commands. That isn’t actually as intimidating as it might sound, but “Buy this and soon you’ll be learning Linux” is not the sort of slogan that say Apple would use. Or indeed anyone who wanted to sell anything.
And yet… It was so damn promising. If they had moved quicker to smooth off the edges of Maemo, if something like the N9 had arrived a year earlier – while people were still actually waiting for it – it might have been a hit instead of a peculiar footnote². Instead, Nokia paid brutally for not getting their collective arse in gear.
But, it was a remarkable achievement and a fascinating experiment. Even when it’s no longer my primary phone I’ll keep the N900 around, especially for travel, as an incredibly miniaturised PC. They can be picked up new on Ebay for under $200 now, I recommend them highly.
Though Samsung were later to make the impossible dream come true by putting an actual Wacom digitizer on a phone.
A footnote to the footnote: There are rumours that Nokia have quietly continued their Linux-based development – just not for smartphones. Dubbed Meltemi, a descendent of Maemo is rumoured to be the future replacement for the S40 “dumbphone” system that has done so well for them for so long, and could be used to bring smartphone-like features to low-cost devices. That might prove competitive against the rising tide of (frequently awful) cut-price Androids. A sad end for the noble Maemo maybe, but it could save Nokia’s bacon – and of course make them less dependent on Microsoft. Who’s to say that a Linux smartphone will not rise again?
All the time I’ve been writing articles under the heading “What Phone Is Right For You?“, I have never confessed to what I actually use myself. Do I favour the Apple or the Blackberry, Windows or Androids?
None of the above. There was a reason I didn’t bring my own choice into it – the phone that’s right for me probably isn’t right for you. Or most anyone, it seems. Don’t get me wrong, it is an extraordinary piece of kit. But it sold in the hundreds of thousands at best, rather than the millions a hit mobile device needs.
Which is a shame, because it could have been the future.
That device is the Nokia N900 – less a phone really than a tiny PC with phone capabilities. Of course that can be said loosely of any smartphone, but it’s true of the N900 in spades. On this one you can have an actual desktop, and edit office-standard documents in powerful applications. Not bad for something you can carry around in your pocket.
Nokia achieved this with the help of the Open Source Software community – people who create free programs because they believe information technology should be free. Or just for the hell of it. OSS has brought us many great things, famously the Firefox browser of course, but also the free alternative to Windows and Mac OS called Linux.
The N900’s “Maemo” operating system is, like Ubuntu, based on the respected Debian¹ variant of Linux, so installing full-scale desktop applications is relatively straightforward. One great example: The GIMP², an Open Source alternative to Photoshop with all the image-editing power you could want. Thanks to this I was able to take, correct and upload the image above entirely on the phone.
And the usefulness only begins there. By hooking up a mouse and keyboard and using the TV-out as a monitor, I can have a desktop PC in any hotel room. Or indeed I can use the phone itself as a Bluetooth mouse and keyboard to control a PC from bed – which came in darn useful when I was ill – or control TVs and so on with its infra-red output. It’s versatility is matchless. Furthermore it excels as a media player, with powerful stereo speakers and a screen that wasn’t surpassed until the iPhone 4 came out.
Its greatest strength though has to be its browser. Nokia had been making specialised “Internet tablets”, portable Web browsing devices, as a semi-experimental sideline since 2005. The N900 just took that and added phone call abilities. Its browser therefore is proper Firefox, just with a finger-friendly interface, and brooks no ‘mobile view’ nonsense. It displays Web pages exactly as they’d appear on a desktop. With Flash, of course.
And though the phone aspect is a late addition, its integration is exemplary. Skype is just there – you can make and receive VoIP calls just as easily as ordinary ones. Similarly, IM services like Facebook and Google Chat integrate seamlessly with the phone’s text messaging. Contacts on different social networks can be merged into a single address book entry too. That’s just the way it should be.
So, what the hell went wrong? If this was such a paragon of smartphone virtue, why did no one buy it? Hopefully I’ll have the strength to face that sorry tale tomorrow.
Poignant little story – Debian was named for its creator, Ian, and his girlfriend Deb – before they broke up. He’s going to have a hard time forgetting her now.
The GIMP stands for The GNU3 Image Manipulation Package.
GNU stands for GNU Is Not UNIX4. Recursive acronyms are kind of an OSS in-joke.
Well we finally got the iBook done. Possibly due to its habit of regularly becoming misconnected, the AirPort (WiFi) card had well and truly failed. You could tell this because if you plugged it back in and switched on, the computer wouldn’t boot or even beep. Its fan would just start spinning at maximum speed, and the card would become too hot to touch.
Generally, not a good sign.
Buying a new internal card from Apple would have been expensive, time-consuming, or both, so we went looking for a USB WiFi adapter instead. I was surprised – the shops were full of very reasonably-priced stuff from Netgear, Belkin and Cisco, absolutely none of which seemed to be compatible with Macs. You’d think it would be worth the small cost of developing drivers. Sure, all Macs come with WiFi built in now, but so does virtually every PC laptop.
We found a nice one eventually though, from a maker called Edimax. It was a bit more expensive than the others, but it’s cute as a button. The same size as the smallest Bluetooth adapter, yet it seemed to have no problem receiving a signal throughout the house.
Any catch? Well yeah… The drivers come on a CD. But to save on packaging – laudable as that is – it’s one of those mini CDs, maybe half the diameter of the proper thing. Not a size you see much since the demise of the CD single, which was the same day as they came out. These work perfectly in most CD drives of course, but Macs have slot-loaders. So basically you slide that little disc on in there, and… you start figuring how the hell you’re going to get it out again.
Except possibly by sheer luck once in a while, a mini CD is not going to play in a slot-loading drive. But no matter, you can download the drivers instead. Just connect to the Wi… Oh right.
Where there’s a will; I happened to have a 3G modem with me, though I suppose we could’ve dug up an Ethernet cable too. After that it was relatively simple. Except that the download link required you to enter an email address despite clearly saying it was optional – a double irritation this time. Otherwise though, it seems a lovely little product. And not only does it come with drivers for most versions of OS X and Windows, it even has them for Linux. One to remember.