It seems like the way to get started with p52 is to tell everybody why you are taking part. I skipped that last week but thought I'd take the opportunity this week after some of the discussions on the mailing list. So why am I doing it? Basically my reasons are the same as why I took part in SiteSprint. A nice set of reasonable goals to keep me motivated and progressing.

What I'm going to do

I don't have any particularly special or interesting ideas like many of the other participants have. I simply plan to keep writing about development and programming related subjects as long as I continue to enjoy it.

The biggest difference is that I plan to have a quick whirlwind tour of different interesting programming languages now and then - I think realistically that will mean one post a month and the rest will be my regular writing. So what do I mean by different and interesting? Well basically they just need to be new to me, they wont necessarily be that obscure or that different I just want to broaden my horizons. I want to start with Erlang, Haskell and Go. We'll see how I get on...

Managing Project 52

Now this is something that I find quite interesting and I have been enjoying the mailing list threads. When Anton Peck originally started Project52 he had imagined only a few people signing up. This is reasonable and I don't think anybody could predict the user explosion. He covers this in more depth on his blog but here is a short extract.

My idea was so focused on having about 20 people involved, maybe 50 at the most, that having a 10 to 20 fold increase in reach was just laughable. Boy, was I ever wrong!

With around 700 bloggers signed up its become more of a chore than anything for Anton. Personally I think any manual form of checking is going to be pointless unless we get a small army together but even then its going to become difficult to manage. The simplest solution would really be to make it completely opt in - let people participate in twitter and in the mailing list and leave things as we are.

Do we really need to point out who has failed and who hasn't? With SiteSprint most people didn't finish their sites and actually the status for most is totally unknown. This didn't really matter, those that got involved and into it collaborated how they felt was best. I used Twitter and IRC for this. If blogs didn't have feeds then I'm certain we wouldn't even consider managing it but since they do it seems easy on the surface. Having users manually enter URL's into a bespoke system could work but how do we check they are real? I could just provide a link to techcrunch each week or some other website. Besides, this system would probably be harder to create than...

My Vision of Project52 Backend

A back-end should really take the feeds from blogs¹ and aggregate them ultimately creating a set of aggregated feeds that separate the content by tags/categories and author. If your blog is setup right it will share the tags you use and make this pretty easy to work with.

Ultimately then we could have a web site that allows you to view the posts by author, category or go for the fire hose with all the posts if your a reading fanatic thats interested in everything. There are concerns about noise but you would only need to subscribe if you wished. It would be fairy simple to add functionality that created a feed backed on a subset of users with a url of something like /feeds/users/joe+john+jack/ - with a nice mechanism like that we can use it for tags and categories too.

If you've thought about this like me then you may have realised I am describing a very common system. These are usually known as planet websites and used for many purposes across the web. This style of site is proven to work and scale, we simply need to add a few custom views to a planet. It is something that should really be quite simple to setup and there are a large number of open source solutions that make most of this very easy.

Is noise going to be a problem? Not in the slightest. We don't even need to show the main aggregate to everybody if we don't want too. What we need to take advantage of is the fact that planet websites are just feed aggregators and that is the hardest part about what we need to do. Anybody that has had to deal with the vast and horrible array of feeds different blogging software products produce will understand the pain.

Failing conditions

I don't think people can really fail or be booted out - it really goes against the principle of the full thing. The same with badges or rewards, they don't really mean anything (at least not to me). We should be doing this for ourselves and our blogs, the reward is to have content your happy with and getting into writing.

Hell if somebody does 30, 20 or even 10 weeks out of the year I think thats still a win. It's a start and thats something, OK its not as grand as 52 weeks but they had a go and produced some content. Good on them.

While I don't think people should fail or be struck off I think we should sort them by those that have been keeping to schedule and those that have fallen behind. Think of it like for each week that you complete you get a point and then we sort by how many points you have. It would be a bit like a high score but we need to make sure we don't give a point per posts unless we want to change the name to project infinity :).

Round-up

These are just my thoughts and really Anton needs to make a decision as to where to take P52. I'm confident I could have had this system mostly up and running by now but I don't want to step on toes or start a programming COOP!

However, I'd be happy to help out here I can and I know some other people would too.