You are here

Mac OS X Front-end For Linux

Cutting Edge By Dave Shapton
Published May 2000

Mac OS X's Aqua: the new graphical desktop front‑end for Linux systems?Mac OS X's Aqua: the new graphical desktop front‑end for Linux systems?

Dave Shapton muses on the possibility of Apple's OS X becoming a desktop front‑end for Linux, and explores the consequences of a new wireless communications protocol for musicians.

What follows is pure idle speculation — but the fact that it is highly unlikely doesn't stop the possibility being tantalising.

You've probably heard that the new Mac Operating System is called OS X. Whether you're supposed to pronounce it OS Eks or OS Ten I'm not sure, but what I do know is that it is a descendant of Next Step. Next Step was at the core of the Next Cube, a dramatically different computing platform conceived by Steve Jobs back in 1987. Despite its radical appearance and user interface, Next Step is basically Unix. This is where it gets interesting. Since OS X is based on Next Step, which is a Unix derivative, OS X must be part of the extended Unix family as well.

So what? Well, there's another member of the Unix menagerie that I haven't mentioned yet: Linux. Can you see where this is leading? Linux is (sort of) Unix. And so is OS X. Now, it just happens that what Linux lacks is a good implementation for desktop machines (as opposed to servers and the like) — and the most striking thing about Apple's OS X is its user interface, which Apple call Aqua.

Linux is given away free, but to avoid prohibitive download costs and to provide some semblance of organisation, companies such as SuSe (pronounced zoo‑zuh) and Red Hat (since you ask, it rhymes with dead cat) bundle it up with a manual, some support and a couple of thousand Linux applications, and sell it for between £30 and £80. It's cheap, then.

Now, if I were Apple (which I am clearly not) I'd be looking at Linux as a 'Trojan Horse' to get Aqua on to a potentially huge number of desktops. As Microsoft have so clearly demonstrated, if you 'own the desktop', then you effectively own the hearts and minds of the users. Linux is a powerful operating system. It's what computer programmers call a 'proper OS'. To non‑technical users this has traditionally meant "difficult to use and wilfully obscure", but if someone were to drape a sexy user interface over it — especially one with the Apple logo on it — I know I'd at least give it a try.

This is obviously a vast over‑simplification, because there are plenty of things about Aqua that are specific to the Apple platform. It wouldn't be a trivial feat of engineering, either, largely because of what has always held alternative operating systems back from mass‑market adoption — and what has always been Apple's strength because of its 'closed' environment — providing device drivers. But remember: this is just idle speculation..

Blue For You?

Ericsson have been prominent supporters of the Bluetooth wireless communications protocol, and have already launched a Bluetooth‑based cordless mobile‑phone headset.Ericsson have been prominent supporters of the Bluetooth wireless communications protocol, and have already launched a Bluetooth‑based cordless mobile‑phone headset.

Bluetooth is a new wireless technology which presumably gets its name from a Scandinavian warrior‑god, rather than a bizarre orthodontic condition. This would be appropriate because one of its strongest supporters is Ericsson, the Swedish mobile phone manufacturer.

On the face of it, there's no obvious reason why a mobile phone company would need a very low‑power, extremely short‑range radio communication system, when mobile phones are by definition wireless! But that's not what Bluetooth is about. It's actually designed to be a simple way to create an ad hoc wireless network, and if you've ever tried to connect a personal organiser (a PDA) to a PC you'll already know why this would be a good thing. The fact is that even with something as elegant as a Palm Pilot, with its dedicated docking station, you still have to make a physical connection to your PC or Mac to exchange information. And what if you're away from your own PC and want to download your email from someone else's computer? If they don't have a docking station, you're stuffed. Then there's the question of how you communicate with someone else's PDA. If it's the same type as yours, you've got a chance, but if it's not, then don't even think about it.

There is another brilliant application for Bluetooth that will make everyone want it: wireless earpieces. Have you ever tried using one of those 'personal hands‑free' kits that are usually bundled with new mobile phones? They have a single Walkman‑type earpiece with a cable that plugs into your phone. About 100cm down the cable (roughly where your chin is) is a blob of plastic with a hole in it that, in the field of mobile telecommunications, passe s for a microphone. What Bluetooth does is get rid of the cable. Instead, you have an earpiece combined with a microphone on a boom (like the kit that singer/dancers use on stage) that contains an A‑D and D‑A converter, and a Bluetooth transmitter/receiver. Your phone also has a Bluetooth transceiver and can talk to the earpiece, so your phone can stay in your pocket or handbag, where it can roast parts of your body that have less to do with your cognitive abilities. (By the way, Bluetooth transmitters radiate incredibly low amounts of energy compared to a normal mobile phone emission. Even though they use the same 2.45GHz frequency spectrum as microwave ovens, a standard Bluetooth transmission radiates less energy than a non‑transmitting device is allowed to leak under European legislation. They're even considered to be safe for use on aircraft. However, one side‑effect of using the microwave oven spectrum is that early tests have shown Bluetooth data‑rates to fall drastically if you heat up your take‑away in the same room!) So wireless headsets are one, very limited, use for Bluetooth, but I suspect it's the feature that is going to bring the first products to market.

The real strength of Bluetooth is that it is conceived as a genuine wireless network. Several devices can take part in a Bluetooth conversation at once, and the protocols within Bluetooth will take care of all the gristly stuff that you normally have to deal with when you set up a network of any sort. This is going to be extremely useful when you consider that mobile phones are going to merge with PDAs, which will also become personal MP3 players, Internet radios, and probably digital cameras as well. There isn't enough space here to go into all the possibilities, but just think about this for a minute: you'll be able to exchange your MP3 collection with someone by just standing near them! In fact you could set your PDA to search out other MP3 files and add to your collection while walking down the high street (although you'll have to walk pretty slowly because Bluetooth's data‑rate is only 1Mbit/second — so a four‑minute track will take about 30 seconds to download!).

A Studio With Teeth?

Mac OS X Front-end For Linux

Bluetooth won't revolutionise recording studios. Its data rate is high enough to support multiple MIDI channels, and I certainly like the idea of a rackmount synth module recognising my master keyboard simply because it is in the same room, but I wouldn't fancy using it at a wedding gig when the cook puts 200 sausage rolls in an industrial microwave oven. However, Bluetooth uncovers, and to some extent solves, some issues that could have a big influence on the way we work with music technology.

Let's assume that wired and wireless technology has reached the stage where we can confidently expect any two devices to be able to send data to each other. This would mean that we can ignore issues around connectors, low‑level protocols, and so on. In effect, every device will effectively have a bi‑directional socket (or radio connection) that might as well be labelled 'data to and from other devices'. Simple as that. The trouble is that it's really very complicated getting things this simple.

The way we've done it up to now is to have dedicated sockets and connections for specific functions: audio in and out, for example. To avoid the possibility of plugging an audio cable into a MIDI socket, not only are they distinctively labelled, but they are completely different shapes! So what happens when you have only one connection that does everything, or indeed no physical connection at all? What happens is that you have to build into your data‑transfer protocols a way in which pieces of equipment can 'discover' what they are connected to.

Bluetooth will never progress beyond a wireless way to connect mobile phones to earpieces unless it has some sort of discovery protocol. As it happens there are several: Jini (pronounced 'Genie'), Universal Plug and Play, and Salutation are just some of them. The idea behind all of them is to set up communication between dissimilar devices to allow them to understand how each device could benefit from interaction with another device. A simple example: let's say you've downloaded some email or a digital photograph onto your mobile phone and want to print it. As you walk into your office, into range of a Bluetooth mini‑network, your phone 'discovers' that a printer is available. The discovery protocol puts the appropriate drivers into place and starts printing on your command.

It's when we put this into a recording‑studio context that this starts to get really interesting. If your sampler could 'discover' your digital mixing desk, then by discussing available digital inputs and outputs, together with effects and other facilites, both mixing desk and sampler could make optimum use of each other's processing power.

In future columns I will explore this subject in a lot more detail. We have the raw power to make an all‑digital studio, and what we need to do now is make it work properly. It seems to me that as a new technology emerges, things get more complicated for a while (as they have done over the last 10 years, in the case of music recording).

We're now in a position to simplify things. Firewire and mLAN (Yamaha's new music and audio transport based on Firewire) give us the speed, but we need to invent our own music‑technology‑specific discovery protocols. How about a Mix Description Language — a way of describing every aspect of a mix so that it can be run on any suitable mixing system?


For years, most mobile computers have been fitted with a wireless communication system that uses an even higher frequency spectrum than Bluetooth: infra‑red. But merely putting two devices together won't get them to work with each other. First you have to ensure that they can recognise each other as capable of infra‑red communication. Then they have to start a dialogue. To do this they have to speak the same language (or 'protocol'). Of course, they have to both be a type of device that can actually have a meaningful conversation together (what, for example, would a bicycle say to a washing machine?). I was one of the last people to buy an Apple Newton, before it was tragically deleted from Apple's product range — just as it was evolving into a Really Useful Thing — and it actually did useful stuff via infra‑red. But I could only make this work with other Newtons, and I have to say that finding someone else with a Newton was the hardest part of establishing this communications link. I also tried handshaking with a Sony Viao Notebook computer, with the result that both devices acknowledged each other, but could not agree on a way to exchange actual data, producing messages like 'data‑link error 2f345e'.

This is all very frustrating because infra‑red communication is very cheap to implement, and can support very high data‑rates — easily good enough for real‑time digital audio. OK, you need line‑of‑sight contact with both devices, but that's a small price to pay for not having to crawl under your desk every time you want to read your email on your PDA.