Age of Conan descends upon us.

Age of Conan is about to come crashing down upon us. It is an MMO game that just happens to have a 14 gigabyte client download. This is around 5 or 6 gigabytes more than the previous record holder, making it a pretty significant bolus of data that people are going to want to squeeze down their tubes.

For those that are unfamiliar with file sizes, that is a staggering amount of content. One of the most common downloads plans here in Australia is 12 gigabytes per month. This means it would be very, very, very difficult for people to download - unless it is hosted locally on an 'unmetered' site, meaning it would be considered 'free' content and not count towards their monthly total.

As with every gaming download, we want to host it on various services to make it easier for Australians to get. This is one of the main things we have been doing for the last, oh, nine years - mirroring gaming-related software so Australian users can get it quickly, without the pain of having to:

- compete for limited international bandwidth
- sign up for expensive or complicated subscription services
- download at a fraction of their maximum speed
- burn through their limited monthly download quotas

Keeping Australian gamers from having to pull bits from the other side of the planet has been a Big Deal for us across the board. I would like to think that, through our efforts, we have made it significantly easier for gamers to get the files they want, when they want.

Ultimately, I'd like to think that, when it comes to distributing gaming files (or any sort of digital content), we have some experience in this field.

We have been dealing with Funcom for the last couple weeks in order to secure permission to mirror the Age of Conan client when it shipped. While they've been hugely cheerful and happy to respond, it is clear that they're simply completely unaware of the problems plaguing Australians - at first they were a little confused as to why we'd offer.

We're still waiting for approval to mirror the full client, which I'm still hoping we'll be able to get - they've been pretty receptive to our request so far. Please note that this article is by no means a dig at Funcom - they've been nothing but helpful throughout the entire process. It is a reflection upon the state of affairs of the distribution of files as it relates to the gaming world, and a look at how Australians specifically are affected by the various issues.

What's so painful about mirroring, anyway?

Contrary to what you might think, there are many things that complicate the mirroring process for us. The problems are vary rarely technical, as you might expect. Those most significant show-stopper for us is usually the result of business decisions made by the developer or publisher of a piece of software that prevent it from being able to be formally and explicitly labeled as "redistributable".

One theory of mine is that this just happens incidentally - the legal departments of these various companies like to include a standard "... will not copy, distribute"-type clause as part of the EULA, blissfully ignorant to the fact that redistribution is something that will happen anyway, almost regardless of what they put in there.

If it is thought about by the business side at all, it is probably in the context of authority - making sure that there's some legal distance that separates them from any responsibility in the event that a user downloads a trojan- or virus-infected client from some dodgy site somewhere. Now that is a very real risk that needs to be addressed - but don't worry, because it's actually a pretty simple problem to solve (as we'll see later).

Why does this even matter, you might ask? Surely if they're distributing it freely anyway, there's no harm in just grabbing it and mirror it? Indeed, that is what is done very frequently. We do it all the time as well, under the implication that when a publisher sends us a file to mirror it, we can do so. However, it is a legal conundrum that I'm not qualified to answer. I suspect at the end of the day it would come down to the enforceability of various EULAs and/or Terms and Conditions of the source website - and it's generally not something we're prepared to challenge on a legal basis.

Does it even matter? Why do we need to mirror things now that we have BitTorrent?

This - unfortunately - appears to be the attitude of many games developers and publishers. BitTorrent (and other p2p file sharing mechanisms) are often seen as the magical cure for all distribution needs. Why bother paying expensive hosting fees, or stress out trying to organise mirrors with other partners when all you need to do is whip up a torrent and use the sheer awesomesauce of getting users to share the bandwidth load?

On the surface this seems like a great idea, but there's a few problems:

• This isn't solving the problem of distribution. It's just moving the burden of distribution from the publisher to the user. Given the users are the one paying for the game, this seems a little unfair.


• Users generally don't like it, because it seems slower. I've lost count of the number of requests we've had to mirror some World of Warcraft patch because "the Blizzard downloader sucks".


• ISPs - particularly some of the big US ISPs like Comcast - hate BitTorrent. There's very strong evidence to support the claim that they are interfering with BitTorrent traffic in order to reduce their bandwidth bills.


• It is utterly, utterly, useless for people on limited download plans who rely on their ISPs or local mirrors to provide them with unmetered downloads (such as the entire country of Australia)

BitTorrent is an excellent piece of software, and it has redefined a lot of how the Internet works in the last few years, and its presence in the world continues to shape a lot of things. We're certainly better off with it than we are without it.

However, for this sort of purpose I really see it only as an excellent supplement for software developers and publishers to use in addition to a regular distribution model, but it shouldn't be the only available system.

What's the solution?

The solution appears simple:

Developers and publishers should release their software as redistributable - not just silently allowing redistribution, but actively acknowledging that it's a part of their software distribution strategy. They should work with content partners (yes, like us) to make sure their content is as widely and readily available as possible.

Further, the status of these files as redistributable should be unequivocal - there should be no doubt that these files are able to be freely redistributed, in the exact form that they are released in the first instance (not repackaged as a .zip file, and not RAR'ed as if saving a few thousand bytes makes it worth having to extract it twice), and the lack of doubt should stem from a clear policy in both the EULA of the software and the Terms and Conditions of the website from whence it came.

We will help absorb the distribution cost - saving uncountable dollars that otherwise would have had to have been paid to hosting companies - and we'll do it with the same professionalism and skill that we have done for almost nine years - and so will all the other players in the market.

This business model has been tried and tested - remember the word "shareware"? It works even better for MMOs where the risk of piracy is practically negligible because of their subscription-based nature.

What about the aforementioned security risks? As I said, they are very real, and they need to be addressed so that everyone can rest comfortably. Fortunately, computer science and mathematics has your back here - there are many good ways to help you figure out if the file you downloaded is actually the file you thought you were downloading.

File hashing (md5sums, sha1sums - which we have been carrying on AusGamers for years for this purpose, being one of the first - if not the first - gaming site to display hashes on file download pages) is one such way, cryptographic signing (using PGP to digitally sign a distribution file, for example) is another. I'd incline towards the former as it is a little easier for everyone and (until people make more dents in SHA-1) still pretty reliable.

A conclusion

I have little doubt that the Age of Conan client will be mirrored on many sites - including, hopefully, many in Australia - regardless of whether or not they encourage or even tolerate redistribution. This is just something that will happen, as it happens for every collection of bytes that appears online for even a moment - it just ends up splayed all over the Internet on mirror sites, personal web pages, and torrent/p2p networks.

Any fuss that is made about it, the more likely it is to be prominently available from the sorts of nefarious outlets that content owners rail against the loudest. Funcom know the score; they won't bother asking people to take it down - but they might not explicitly grant permission to do it, either.

However, hopefully the time will come when planned redistribution of content will become the norm, and more and more content creators will take it on board during the production of their works. Weird fringe cases like Australia will certainly benefit - and the way the US ISP market is heading with their plans to introduce monthly download limits, this may become a bigger issue than ever before.