A recent post on Signals vs. Noise was reminicent of something I wrote last month.

In this post, they review a number of customer complaints and categorize them. They also make the following astute observation:

Every feature that’s missing is essential, a must-have, and the fact that it’s missing is killing someone. Yet the #1 thing that people like about our software is how simple it is.

Here they’re inferring that if they were to add the features people want, the software would become unusable and “the #1 thing people like” would be no longer. Their mantra is simplicty, and these pesky customers keep asking for stuff that might make their product less simple.

Welcome to the real world of post-release software. People want features. People use the software in different ways for different things and want different features. If you care about your customers, you will do your best to give them what they want while trying not to harm those customers who aren’t interested. If you continue pleasing customers, as say Microsoft did with Office, you may end up in a situation where “most people don’t need the full Microsoft Office collection”. You might even have a product on your hands that accounts for $10 billion in annual revenue… and that would be terrible.

Of course, it comes at a cost. Things do become more complicated, and you have to start innovating with your UI. The screenshots of the upcoming version of Office certainly reveal how much work they’ve had to do with organizing all of the features they have. When OSX was being designed, I imagine the fisheye menu at the bottom of the screen came about at least in part because they wanted to preserve screen space without making the icons tiny and hard to click on (that and to make it look sexy). It’s not a matter of limiting the product’s capabilities, it’s a matter of enhancing the product’s capabilities in response to customer demands, and trying your best not to harm usability.

For every “essential, a must-have” feature a customer asks for and 37signals refuses, there is another opportunity for a competitor to make an entrance. There are only so many opportunities you can give to competitors. There are only so many times you can tell a customer you’re not interested in helping them before they start to believe you.

We were having a debate on this at work, so I decided to look it up. I also projected Vista’s hardware requirements based on past trends, but they haven’t been released yet… we’ll see how close they are. Between Windows 2000 and Windows XP, everything pretty much doubled. All numbers are minimums. If Vista hits the marks below as minimum requirements, I would be very surprised.

I just read this bit about the new PayPal Mobile, and it sounds really cool.

Like any new technology of this nature, what matters less is the technology around it, and more the chicken-and-egg problem of getting people to sign up for an account when there are no retailers using it, and retailers to start using it even though there are no people signed up. I’d love to see something like this take off though. Eventually a cell-phone based payment system could evolve to replace credit cards, and could probably be designed to be faster and more secure. This system is comparitively clunky, but would be a nice first step.

Plus if every print ad could effectively be a storefront, that might make print ads more profitable, which presumably would mean newspapers/magazines could charge more for them, and require less of them. Some magazines these days have 3-4 pages of ads before you even get to the table of contents.

I was browsing through some Windows Vista screenshots, and this one was particularly interesting. It was the new Windows Performance Rating tool. First of all, I think a tool like this is a great idea. It gives people a fairly simple way of figuring out if their computer is up to the task of running Windows, and if not, where they should best spend their money on upgrades.

The part about this screenshot though is, presuming the data displayed are real, it basically says that an Athlon XP1800 (supposed equivalent to a Pentium 4 running at 1.8Ghz) with a gigabyte of RAM is underpowered for Vista. Without making any judgements on that fact (I’m sure plenty of blogs will), the reality is that using Vista won’t mean “pop in a CD and upgrade from XP” so much as it will mean “time to go shopping for a new computer.” I know at this point I wouldn’t spend a few hundred dollars on memory and video card upgrades… never mind a CPU (which generally means motherboard too). It’s just “buy a new computer” time.

Continue reading →

Google confirmed yesterday that it acquired Writely, a pretty cool online word processor that I use on occasion. There is more than enough speculation out there that people think Google is going to try and build an Office suite, and at 37signals they are guessing that Google is

“…building up half an office suite. Google knows that most people don’t need the full Microsoft Office collection. They don’t even need most of it. They don’t even need half of each product. They just need a few things (like creating a quick, simply formatted document and sharing it with someone).”

Continue reading →

Myth2Ipod lets you set your recorded TV programs to be transcoded for viewing on the iPod. The best part is that it also builds a php page which you can point iTunes at, and it treats your TV shows like a Video RSS feed. So, iTunes goes off, downloads new ones, keeps them labeled properly, etc. – just like a regular podcast. That way you can just drop your iPod in, download all of the shows you’ve recorded since you last sync’d, and go.

Right now it’s not making use of MythTV’s commercial flags, so your recordings still have commercials, but I wouldn’t be surprised if that was in the works. Very cool. Also good if you’re about to get on a long plane ride (as I am) and have a bunch of recordings sitting on your PVR that you hadn’t gotten around to watching.

With Ruby on Rails there are a number of options on what server to use. Supposedly Lighttpd is “fast as lightning” but FastCGI on Apache is the “standard” method for production use.

Out of curiosity, I went to a site which I knew ran Ruby on Rails and gets a lot of traffic, and inspected the HTTP headers.

Unsurprisingly, the response when requesting the main page included this header:
Server: Apache/1.3.33 (Unix) mod_deflate/1.0.21 mod_fastcgi/2.4.2

However, as the browser started requesting images, they came back with this header:
Server: WEBrick/1.3.1 (Ruby/1.8.2/2004-12-25)

It seems like they’re using the so-called not-for-production webserver to serve up static content. That’s interesting, given that some have described the usage of that server to be reserved for cases when you have low concurrent users and mostly dynamic content. This qualifies as neither.

If you’re a software nerd, chances are you’ve heard of FogBugz and probably read the Joel on Software blog. I’ve never used FogBugz, but I’ve used a number of bug tracking systems and have learned a lot of lessons about how bug tracking “should” work in a large organization. I walked through the FogBugz tutorial, assuming I’d see some of my common troubles solved. Since Joel is an outspoken expert in the software community, I presumed that the bug tracking software would be very well thought out. It certainly was appealing to look at.

But, one of the first questions I had, and couldn’t figure out from the documentation and tutorial is – how do you handle a bug that spans multiple releases? It’s a pretty simple question. If you have a bug, and it exists in release 3.0 of your product and onward, how do you mark it? Let’s say you decide to fix it in release 4.2 and 5.0 of your product, but you’re not going to fix it in 3.x or 4.0. How do you indicate that?

Continue reading →