mattmecham.com

We always eat our dog food

Eating your own dog food is the imaginative term used within the computing industry to describe the act of using your own products. It's obviously derived from dog food manufacturers who taste test their own products. How else would they know that it's actually chicken and rabbit flavour?

With that vile mental image aside, 'dog food-ing' is a very useful process which often helps mould the development process. We use our own forum system very extensively. A lot of the features in our software have come from the dog food-ing process. Inline moderation and multi-moderation came from spending hours organizing the bug reports forums. I found it a very tedious process replying to each bug report then moving each bug report to the appropriate forum. One day I figured it'd be much easier if I could set up two or three moderation actions and run it with a flex of my mouse. This makes it very easy to reply, close and move a topic. A few months later, I figured it'd be neat to be able to click a bunch of topics and then apply either a single moderation action or a multi-moderation action to batch process those topics. Inline moderation was born.

These two functions would probably not exist in our software if we didn't actually use it on a daily basis. In this case, eating our own dog food was very productive.

As my thoughts turn to content management systems, the need to eat our own dog food is magnified. We obviously have a web site - several in fact - and we need to manage them on a daily basis. We need to share the same 'master' templates across three or four domains and each site has very specific requirements. I'm virtually salivating at the prospect at eating this chunk of dog food. I already have a list of things that I'd love to see in a web management system. Through years of hand editing and watching Charles hand code blobs of perl code to create a rudimentary CMS system we know exactly what we need in order to run a very diverse suite of sites. The last development stage will be to hand each team member a copy of our C(?:W)?MS system and ask them to copy a well known site - as this will be the first step that many of our potential customers will do. This will get us thinking about how to design our import tools and how the system works in converting an existing site and ultimately work out any logic or code flaws.

Having said all of that, meaty chunks in gravy are optional.

Paying to get less email

So, this Monday I began my Monday ritual. I loaded Thunderbird for the first time since Friday afternoon and pulled out a magazine to read while my connection downloaded 1800 emails at full speed. Now and then, I glanced up to watch my Mac's beachball icon spin around merrily as Thunderbird's spam filters went into overdrive trying to chomp and digest the vertible gastronomic feast of spam.

Roughly 4 minutes later, Thuunderbird proudly displayed 180 emails in my inbox it decided weren't spam and moved the rest into my trash. A few minutes of mindless clicking and deleting passed while I filtered out the rest of the not-spam-but-nearly-spam emails. I then read the remaining 3 genuine emails and loaded Mozilla to start the day with the feeling that this is a stupid way to start the week.

A few of my peers at IPS towers have purchased an annual subscription to Mail Blocks to help fight the war against wasting time. This is a good service but it relies on the other party authenticating the email they sent by clicking a link in an email that Mail Blocks sends. To me, this defeats the purpose of reducing unwanted mail as for every spam email you get a validation request bounced back to the sender. Often the sender of the SPAM is unaware that their email address is being used so they end up with all the "Validated your email!" requests.
Other systems rely on you downloading the email so it can be processed on your computer. This is all very well and good, but you're still having to download the email first and with many UK broadband providers now capping downloads, this is a time consuming and often costly exercise.
I remembered a recent article in .net magazine about a service called Email Systems that actually took your POP log in information off of you and gave you a POP account with them. They then download your mail every five minutes to their servers, process the mail, remove and store the spam elsewhere and then leave the 'good' email on their POP server for you to download. This, to me, seemed like an ideal situation. The only downside is that it costs £3.75 a month for every email address. However, I decided that it was worth the 12p a day to free up the time and bandwidth consumed by downloading junk email and signed up.

I've used the system for about 36 hours now and it's pretty effective. It's caught over 700 spam and virus emails since I signed up which is pretty impressive. It's nice to actually download all my mail in a few seconds and only have a few emails to manually clear up. The system isn't perfect and I'd like more configuation options such as a white-list, black-list and some way to manually add your own filters. It's also a little annoying because it sends you an email to tell you it's intercepted a virus. I can see the sense in that so you'd know that you might already have a virus, but it'd be nice to switch that off and I can't seem to do that. The only configuration option it has is the ability to ping the IP address of the sender to ensure it came from a valid source. They do promise more options soon and they would be very welcome. Overall, it's a great low-cost way of filtering out the junk and I no longer dread the first download of every morning.

So, I'm paying a company to read and censor my emails. Take that, you tin-hat wearing GMail hating weirdos!

Ikea creates Ikea hate site

With a flex of its Swedish muscle, Ikea has created its own hate site. It's a fascinating concept - why wait for someone else to do it when you can do one yourself!

It's obviously part of their new marketing campaign "Elite Designers Against Ikea" (http://elitedesigners.org/) which features the terribly chic 'Van Der Puup' and his "crusade against the big stupid blue place, IKEA".

It even features a little store, where you can buy a mug for £370 or a Limousine flag for £420. On attempting to purchase a mug, I was told:

Thank you for your interest in my beautiful creations.

Your order will be examined and processed shortly if we feel you are worthy of posessing our Elite Design items.

Pure farce, but a great example of a viral campaign which is sure to set blogs alight around the world.

Now, if only Microsoft had thought of this. It would have saved their legal department millions.

Prototyping the easy way

When you're at the very first step of developing a new product, it's often helpful to create prototypes of the final application for evaluation. Sometimes a great idea on paper doesn't work out in code or a good idea doesn't gel well with the user interface.

Some developers claim to hide themselves away in a sterile environment when creating their latest project. They claim they do this so they are not influenced by other products and won't be tempted to copy other ideas and thus their product will stand on its own merits.

Ethically sound but also foolish. Part of prototyping is to evaluate what works and what doesn't. If you don't expose yourself to this part of the process the end result will be ineffective as a finished product. The easiest and cheapest way to prototype is to test out your competition. Trial their software as much as possible and draw up a list of things that you do like and what you don't like. This will give you a new starting point of inspiration and should help you knock out those early design issues.

Obviously the aim isn't to simply build a clone of a current product. The aim is to get your creative juices flowing in the right direction and to reduce the need for complex prototyping.

Over the past few months I've tried just about every CMS product I can get my hands on. I've written up a long list of good things and a longer list of bad things and it's got me thinking in a new direction. In using these systems I've yearned for features that don't exist. Through using badly designed content management systems I have a clear idea of what makes a good system. I've also read their user's comments and feedback to see what they like and what they don't like. I feel this gives me a much better picture of what I want to acheive.

Most inventions take an existing product or idea and make it better. Fellow Brit James Dyson re-invented the humble vacuum cleaner. It's highly likely that Dyson had been using vacuum cleaners for years and through discovering the negative points he drew up a mental list of things to change. Had he never used a vacuum cleaner and denied himself the ability to prototype existing products he'd never had the inspiration to make it better.

So, next time a developer polishes his imaginary halo and tells you that he's never used a competing product tell him that's not always a good thing to admit.