Containers being standard makes the world a better place
Or at least a cheaper place. Which may or may not be the same thing, but for me professionally it is basically true. A way of achieving the desired output that is cheaper than the last way is a good thing and allows the money saved to be spent elsewhere.[1] This should make the world a better place, albeit marginally.
So, containers. Specifically this post was inspired by shipping containers, but there many other things out there, like railway gauges, that have been beneficial by their being standardised.
Before shipping containers the speed with which you could load a ship was controlled by all sorts of factors: the size and type of ship; the sizes and types of the objects to be loaded onto the ships; the cranes available at the dockside or on the ship itself; numbers, strength and current injury status of dockhands and crew members; etc..
With containers someone loads the stuff into the container limited only by volume and mass, in much the same process, time and effort they would have needed in loading the vehicle delivering the goods to the dock. The container is then put on a train, an artic or a barge and moved to the dock where one type of crane lifts the container onto the ship. The containers are a standard size and shape and a known maximum weight. The ships need loading space that fits containers and can be scaled as necessary.
The end result is a method of shipping goods around the world in huge quantities at a small proportion of the costs before. Using containers is simple, safer and less time-consuming for everyone. Excellent. A simple and open standard saving us all money. Your container needs to meet a size and strength specification, have places for the crane to get hold suitable for the maximum gross weight of the container, and be marked appropriately as per the ISO standard.
Now in much of IT standards have been equally beneficial, and pretty much all the best ones have been equally open. The Internet only exists because the standards were and are open to all; the web as we know it today only exists because the specifications for HTML and HTTP, etc., were open from day one. Networking works because TCP/IP is a standard (the dominant protocol used to connect network devices to each other); because RJ45 is a standard (the network port and plug you probably have on your workplace wall); because Cat5 is a standard (a standard for network cables, most likely the cable connecting your computer to the network port). And so on and so on, throughout your computer and its connected devices (a few hardware manufacturers being a dishonourable exception 
) all the way up to the operating system.
Until we get to interchanging human-level data. Then we (collectively) seem to think standardisation is about having the same software version as the majority. It isn't. We need to standardise the way we interchange data and as consumers, whether professionally or domestically, ensure that those providing the tools with which we create and share data use and comply with those standards. We need interoperability. This will have much the same benefits to us all as shipping containers have had and that the other standards which IT already depends upon have had. Real interoperability. That means I use my computer to create a file and store it and you use your computer to retrieve it and read or edit it. Without issue. Without you having to use the same software, operating system or even device type as me.
I want to be able to create a document using OpenOffice on a Kubuntu PC, store it on my document management system and have you able to access it with your Symbian or other smartphone and read and edit the document with whatever software you have on it and store it again, then someone else retrieve and open it to check and comment on something in the document using their Apple Mac using whatever software they have, and so on. Without anyone having to fiddle with formatting or install some extra stuff so they can access the document management system. And then me retrieve the document again, open it with Lotus Symphony on a Moblin Netbook, accept or reject the various comments etc. and PDF the document and email it off.
This means the format I save it in has to be a standard and all the editing packages have to properly, really support that standard. Luckily we have one that we can use for that, ODF (OASIS Open Document Format). It means that all technology involved has to be able to speak the same network protocols. Luckily we have some that work great, TCP/IP and HTTP. It means that that the devices all need to be able to authenticate to the document management system. More difficult depending on the strength of the authentication required. But we trust SSL/HTTPS for many online financial transactions so that should do. Awesome, I have what I want!
Except one software manufacturer seems to think I don't want interoperability on my terms, only on theirs. They call it "true interoperability". A phrase which as someone pointed out has all the Orwellian overtones one could possibly want. Especially when you find out that interoperability on their terms doesn't seem to mean the interoperability outlined above but something far more like the "interoperability through everyone using the same software" situation that we already have. Annoying isn't it? Even more annoying when you find out that company, having avoided being involved in development of ODF, are now involved in numbers and trying to prevent it becoming more interoperable.
Interoperability. We want it. It will be good for all of us, make life easier, make IT more inclusive, drive down costs, etc., etc.. Apparently however we must be clear we want real interoperability, proper interoperability, interoperable interoperability. Not that pesky Newspeak "true interoperability"...
[1] Bearing in mind my employer's "not for profit" status. Those in the private sector can think more money for shareholders and maybe the public sector can focus savings on reducing taxes 
Posted at 02:35AM Feb 05, 2009 by Chris Puttick in Open Standards |