A series of articles about the development of this site? Surely that's possibly the most self-indulgent article that I could ever post?

That's undoubtedly true, of course; still, the first incarnation of Noise To Signal is notable for one important fact: it failed. Utterly. In a way that made everyone involved lose interest in even posting on the site. And I've never done a design that just didn't work, before. Sure, early G&T isn't great (well, neither is the current design - but I'll get onto that in another article), but it works. The first NTS design just didn't. So, what valuable lessons can be learned from this?

NTS officially started life on Monday 11th June, 2005. That all important first email, as sent to my fellow ODers:

Date: Mon, 11 Jul 2005 20:50:36 +0100
From: John Hoare
Reply-To: John Hoare
To: Observation Dome Admin
Subject: Media website thingy

A website covering film/tv/newspapers/radio/any media at all.

A front page, which will be a cross between OD and G&T.

Left hand side being a group blog type thing.

Right hand side being a load of cool articles/interviews/reviews.

Won't be done until after waaay DJ, and probably after a G&T redesign,
but you've got time to think if you want to be involved or not.

For what it's worth, originally I was going to do something that just
I contributed to, but I thought I'd open it up and make something a
bit bigger. But I suspect a lot of you either won't have time to do
anything for it, or would prefer to keep anything you do on your own
sites, or other sites that you contribute to. But the offer's there if
you want it.

--
John Hoare
www.ofla.info | www.ganymede-titan.info

In a rare example of me not fucking about, I came up with a test design of the front page on August 1st. Here it is, exactly as it was done at the time. It was a deliberate reaction against the design of Observation Dome, as I wanted the two to be completely different from each other.

Moral 1: Sometimes, reacting against a design is not the right way to go about things. Think about what you need for this design; don't worry too much about your last one.

Nonetheless, at the time, everyone liked it. A lot. So, I went ahead with that as the basis of development. Different people have different ways of dealing with test designs; some people do Photoshop mockups, some people implement the designs straight into the CMS of their choice. I prefer to do HTML mockups - there's no worrying about how to actually implement the CMS side of things, but you still get to figure out the kind of markup and CSS you want to use, which you'd have to figure out seperately if you do a Photoshop mock. Still, there's no right way to do these things; whatever you find easiest, really.

It was at this point that discussions about the name came up.. We had loads of ideas; as well as Media Whores (rejected due to concerns about the suitability of the name on people's CVs), there was Media Darlings; Lazy Media Slags (a personal favourite, thought up by Cappsy - but rejected for the same reason that Whores was); and Status 5 (sounded cool, but rejected for being completely irrelevant). At this point, we weren't overly-concerned, however - it'd come eventually...

After a lazy August, I restarted work back in September. Now we had a basis for the design, it was time to decide which CMS to use. For something like Observation Dome, which is essentially just a group blog, Movable Type was more than adequate. The media site (as we were still calling it) was clearly a far more complex proposition; covering all media ever clearly required an advanced categorisation system, which MT simply doesn't have. Moreover, I wanted to add more community portal features, such as polls - the kind of thing that MT just wasn't designed for.

Until two weeks before launch, I considered using Drupal, which would have been ideal. Unfortunately, impatience got the better of me; I know how to use MT, and whilst using Drupal with existing templates would be easy enough, designing new ones would take quite a lot of effort to learn. So, I decided to go with MT as a stopgap, and transfer it across to Drupal later.

Moral 2: Use the tools that are appropriate for the job. If you need to spend more time learning them, fine; you save yourself a lot of time, effort, and pain later on.

Discussions continued. And, as so often happens when designing, you have to compromise. The original idea for the front page of the site was the two equally-spaced columns you see in the test design - the blog on the left, and the articles/reviews on the right. Unfortunately, this left no room for something I think is really important on the front page of a site like this: a Latest Comments list. How many times do you go to a site with one of these, or refresh the page, just to see what comments people have added? It increases interaction hugely - because it's obvious where the new comments are, rather than hiding them away.

Besides this, I wanted the archive lists on the front page too. And if polls and stuff had been added, we would have needed a third column anyway. I wondered about a complete redesign, but didn't quite have the heart to get rid of one that everybody had liked. So, unwisely, I added it. And the design went ever-so-slightly out of the window. (The reason this is the case will become apparent in Part Two.)

Moral 3: If you need to completely redesign... completely redesign. Don't be afraid of starting again if there's something that doesn't work.

Meanwhile, discussions about the name were by this point beyond tedious, with That Media Site becoming a distinct possibility at one point. In a set of emails on September 12th, the following were suggested: Media Review, Watch Media, Mary Whitehouse Takes It Up The Arse - And She's Dead, Too, Media League, Media Life, RED DWARD PPL TALK ABOUT TELLY AND FINGS, Media-ocre,Media Cited, Media Polytechnic, Something Without Fucking Media In The Title, For Fuck's Sake., The Media Medium, Media, Shit and Cuntflaps, The Media: Yeah, No, Yeah, No, Media Around You, Thanks Media. Thedia, A Terribly Sensible Media Site For People Reading Our CVs, Egos Gone Mad, We Were Better When We Did The Thing With Beards And The Penguin, Look and Read, Stop, Look and citesten (and Watch), Happy Media, Mean, Media and Mode, Maid Media and Her Merry Men, Maid Marian and Her Media Men, Maid Marian and Her Merry Media, and (wait for it) - Media Marian and Her Merry Men.

Things were not going well.

A week later, we were nowhere nearer, with Media Focus Group, The Focus Group, Media Focus, The Media Circus, Square-Eyes, and Square-Eyed all considered and rejected. Until Tanya hit on the perfect name: Static. Everyone liked it; it was just perfect as a name.

Unfortunately: all the domain names were taken. Literally, all of them that we could have used. We considered staticmedia.org, but it sounded too... corporate, and rather clunky in any case. We considered Static Interference, but it was too much of a mouthful. I'm usually slightly sniffy about people demanding good domain names - I don't tend to think they're quite as important to the success of a site as some people think - but on this occasion, we really were stuck. There was always the possibility of using a subdomain - I could easily have set up static.ofla.info - but the sheer scope of the site demanded that it have its own domain, really. And for the first time, a name I really wanted to use for something couldn't be used - purely because of the domain name situation.

The next suggestion was Pink Noise, but that was rejected as being GAY. Then there was High Definition, Standards Conversion, White Noise, Signal To Noise... and finally, Noise To Signal. The idea behind the name being pretty simple; if signal to noise indicates the ratio of useful information to waffle, then Noise To Signal indicates that we, erm, write a load of shit.

There was resistance to the name at first, as people didn't understand it - but once I explained... well, it was less a case of people being more enthusiastic, and more a case of people being completely fed-up and bored - as indeed, you are now by the whole discussion - and seizing on a name that was just about acceptable. Four days before launch, and we'd got somewhere.

The domain name was next. It's an issue that I feel strongly about - because people can infer so much from a domain name, however unconciously. A .co.uk indicates a business (or, at the very least, some form of commercial activity) set in the UK; a .org is supposed to be for a worldwide organisation, etc etc etc. My preference would really have been for a .info - a relatively new domain that really indicates what most sites on the web are all about - information. NTS isn't really an organisation - it's a group of people giving out information about various types of media.

Nonetheless, the group preferred .org - I think mainly because it's more well established, and also because four letter TLDs still look slightly weird to some people. The interesting thing is, for this type of site, using .org is pretty well established - because .info was only launched in 2001. It seems amazing to me that there wasn't a suitable TLD for a lot of sites until that time. So, although my preference would have been for noisetosignal.info, I wasn't too fussed about going with .org instead - and besides, NTS, as a group operation, can claim a certain status as an 'organisation', however loose.

At any rate, it was important not to give it a .uk domain, as that restricts the scope of the site to UK things only. We want to encourage international contributions.

Moral 4: The name of your site is important as a mission statement... but so is its domain.

Meanwhile, work on the site was continuing apace. The actual article pages were easy enough to design; they tend to design themselves once you've got your front page, as it's simply a matter of carrying certain design elements across. What took more work was the logo; we'd hit on the idea of changing the logo depending on what media the article was about; filling it with a different bitmap image. These were excellently done by Seb, and were the best thing about the entire design.

The test site was now implemented in Movable Type; we'd added categories down the side navbar, a little about section was written (always important in defining the feel of a site), people were getting ready their articles - everything was going beautifully. There was real excitement in the air about the site, and what we could achieve with it.

Unfortunately, at this point, we hit a snag. A snag that effectively, completely fucked the site. And it was so easily avoidable. Believe me, my rather simplistic and obvious morals are about to become rather more important.

More next time.