Software licensing is a tricky art - too little security and you leak revenue, too much and you leak customers. I worked on several license management systems at Ç-Dilla and Macrovision so I've spent far more time than is healthy thinking about this problem.

In general I think the best system is one that helps keep honest users honest. A speed bump that itches the conscience but doesn't get in the way of legitimate customers getting their job done.

I just migrated to a new computer and have finished several days of installing software and drivers. This is never a fun task, but I've been through it a few times and keep all my license information on a NAS drive to reduce the pain.

WinZip has a great trial model. The product is fully functional and nags you just enough that you'll eventually pay. I've been a customer for years, and as I moved to Windows 7 I emailed to ask about upgrade pricing. I got a prompt response and was up and running in no time.

I didn't need to upgrade. The old version of WinZip installed just fine using an existing license key.

ScanToPDF from O Imaging was a different story. The license is locked to a PC, and there's no way to move it automatically. You have to email them. It then gets worse - there's an "administration charge" to move a license. So as a paying customer I have to wait for the UK office to respond to email and even then I can't continue to use the product I've paid for.

I'm sure it's in the small print somewhere.

But the impact is that ScanToPDF has lost a customer, an advocate (I've suggested the product to others in the past, never again) and infuriated me enough to throw up a negative blog post. Is the administration charge really worth it?

At Catfood I use very simple license keys locked just to an email address. I'll refund any purchase with no questions asked. I'll issue new licenses as needed to keep customers happy. I have an online service to retrieve lost keys.

The products get pirated immediately, and finding a key generator doesn't take a lot of sleuthing. I don't care about this at all, because happy customers recommend products to their friends. Pain-free licensing is absolutely key to happy customers. Don't fall into the trap of putting your energy into complicated licensing and enforcing procedures. Add a new feature instead.

I was a micro-ISV (µISV) for years before I heard of the term. It was coined by Eric Sink to describe a one man software shop, and is now generally used for any small software company.

There isn't much market data available this far down the long tail so I've spent some time analyzing PAD files to see if I could answer a few questions.

PAD is the Portable Application Description specification from the Association of Shareware Professionals. It's used to describe software for submission to download sites. How useful these sites are is another question — read Scott Kane on this if you haven't already.

I spidered all the PAD files listed in the ASP directory, downloading data on 76,066 products from 39,861 µISVs (companies / people / publishers). It's not a perfect data set as there are PADs that aren't software and µISVs that don't use PAD. I've also heard that some people are developing web apps these days. But here goes…

Where are the µISVs?

Overwhelmingly in the US. Other countries with more than a thousand listed are the UK, Russia, China, India, Canada, Germany and Australia (in descending order). Most countries have at least one µISV but the numbers fall off pretty quickly.

How much do µISVs charge for their products?

$29.95.

About a third of products are free and a third fall into 9 price points (all ending in 5). I found over a thousand different US Dollar price points overall.

The most expensive product was a $150,000 Green Living site license from South Beach Software (an order of magnitude more expensive than the runner up).

How large are µISVs products?

There's not much action past 20 MB. Most downloads are between 1-2 MB. There's an interesting little spike around 14 MB. I guess this is a popular framework, possibly Java? The largest download was almost 1.5 GB.

Are µISVs still releasing downloadable software?

This is a tough one to get at because PAD files just tell you about the most recent version, not the release history. The chart really shows a last update distribution for the products in the PAD catalog.

There's a large number of products last updated in mid-2008 with nothing comparable in 2009. Could this be a drop-off in PAD usage? A shift to web apps? Maybe final releases before the recession hit leading to less spare cycles for side projects (my µISV certainly pays for beers rather than mortgages).

How many products do µISVs publish?

This final chart shows that most µISVs have just one product. Of course in some cases there might be a brand per product and still a single entity — it's impossible to separate this out from the PAD data. The largest number of products from a single µISV is 616.

I've been busy updating a couple of Catfood products.

Catfood Earth updates your desktop wallpaper to show day, night and cloud cover. It can also display earthquakes, time zones, places and US weather radar:

Catfood Follower is a Twitter tool that automatically follows people who follow you and/or removes friends who stop following you. I think it's most useful for business accounts:

With these two out the door my next project is overhauling the Catfood website. I'm kind of dreading it, but the current architecture is over six years old and it's definitely showing its age.

I've just released a .NET library for parsing ESRI shapefiles - see ESRI Shapefile Reader on Codeplex. The library and source code are available under the Microsoft Public License.

A Shapefile is actually at least three files: a main file containing shape data (*.shp), an index file for locating shape records in the main file (*.shx) and a database of metadata for each shape (*.dbf) in dBASE format.

I ended up writing the library in order to convert Eric Muller's time zone shapefile into a format I could use with Catfood Earth. I found other libraries that could read shape data but not metadata, or provided a very thin .NET wrapper on top of unmanaged code and so I decided that a fully managed library could be useful.

The library - Catfood.Shapefile.dll - provides read-only, forward-only access to shapes and shape metadata. Currently all 2D shapes are supported: Null, Point, MultiPoint, PolyLine and Polygon. I might add additional types in the future, or if you have a pressing need it would be easy to extend the library by looking at an existing shape subclass and the shapefile specification (PDF).

See the Codeplex project for sample code and documentation.

This is a joke metric that I first proudly displayed on Catfood Magazine back in 2007 1997 (it's broken on the archive of the site). Everyone had a hit counter back then, but as far as I know we were the first site with a non-hit counter.

The dirty secret was that the counter just showed the world population. The readership was a rounding error.

My new count of non-visitors uses the US Census Bureau's world population estimate and subtracts unique visitors from the Google Analytics API. The count is cached for an hour so it doesn't slow the page down too much.

Them: Hello, my name is Phil, how can I help you today?

You: My DSL connection is slow.

Them: Okay, I can help you with that, have you...

You: My first thought was that the Linksys router that's been working perfectly for five years has gone wrong. So I connected my computer directly to your off-brand modem with the same result. I then thought that the problem must be with the computer, so I reinstalled it from the manufacturer discs and rebooted about seven times.

Them: Let me connect you to my supervisor...

Them: Tap, tap, tap, oh, we seem to have switched your service back to the basic package. Tap, tap, tap, fixed.

Sometimes it's fun to argue with support. Sometimes you just need to short-circuit the idiot script to get through to the person who can fix the problem.

I think it's time for CAPTGUAs or Completely Automated Public Turing test to tell Geeks and Users Apart. A quick puzzle or two that bypasses the first couple of levels of support.

I'm a Joel Spolsky stalker at the moment - after Stack Overflow DevDays last month I spent three days this week at the Business of Software conference in San Francisco.

It was an incredibly high value conference, in terms of both speakers and attendees. Next year it will be back in Boston, which sucks for me, but I'll make every effort to attend.

I was really excited to see Geoffrey Moore speak. An old boss once bought a crate of Crossing the Chasm for everyone in the division to read. It's still the best business book I've ever read. At the conference Moore spoke about innovation - specifically differentiation (get out of the competitive set), neutralization (get back in to the competitive set) and optimization (productivity gains). All three are essential, but you're shooting yourself in the foot if you spend too much time on neutralization - "Best of breed is a suckers game". His thesis was to do the bare minimum needed to stay competitive and then pour resources back into differentiation.

A theme of the conference was on motivating yourself and others - how to build a great company/culture. Several speakers talked about carving out time for creativity and fun. Carsonified evidently operates on a four day week. I've spent the last couple of years on a six day week... lots of food for thought here.

I convinced myself to attend this year after watching some of the videos from the 2008 conference. These are available on the Business of Software Ning - I'd recommend joining and checking them out. Hopefully videos from this year's conference will be posted soon.

An alternative to our normal loop is another three mile walk starting from JFK and Transverse. Instead of heading towards the beach walk up under the bridge and take the first right to Stow Lake.

Stow Lake surrounds Strawberry Hill. There's a path around the lake with two bridges leading to the hill. I normally walk to one bridge, cross and then up to the top of the hill. It's generally pretty quiet with great city and ocean views in between the trees.

View from the top of Strawberry Hill. Rudy is in there somewhere, click for a larger version.

After the hill I walk over to the Museum Concourse. The de Young rises through the trees like an implausibly stranded copper aircraft carrier. It hosts a range of paintings, decorative art, sculpture and modern art. You're bound to find something you like here.

The copper cladding will turn green eventually, probably before the new Bay Bridge span is finished.

Opposite the de Young is the California Academy of Sciences. The Academy features an outstanding digital planetarium, a three story rainforest and an aquarium. It reopened last year after a major rebuild and it's insanely popular - get there first thing to snag the first come first served planetarium tickets.

On the way back I detoured up to Prayerbook Cross, donated in 1894 to commemorate the first English sermon on the pacific coast after Sir Francis Drake discovered Drake's Bay (a name which didn't stick).

(2.97 miles, total elevation gain 103 feet, 1 hours, 17 minutes, average 2.29 mph, view in Google Earth, view in Google Maps.)

Hike starts at: 37.770433, -122.479941.

Just got back from the StackOverflow DevDays event in San Francisco. I was a bit worried that this would be overly focused on marketing FogBugz and StackOverflow. There were brief pitches on each (and I learned that FogCreek is launching hosted source control and code reviews called Kiln, now in beta and looking pretty nifty with tight integration with FogBugz) but this wasn't the focus.

I was also a little concerned that I'd be the only one there without a StackOverflow profile t-shirt. Luckily I didn't see any reputation toting pod people at the conference.

Happily the day was very code oriented, and very diverse. Spell checking in Python, smartphone development for iPhone, Nokia (via Qt) and Android, ASP.NET MVC and jQuery. I spend most of my time at the moment in C#/.NET and it was really valuable to spend a day briefly diving into different stacks and platforms.

Joel said that they'd be back next year and I'm hoping that it offers a similar diverse range of topics.

Zumbox is trying to take the paper out of the postal system. It's a laudable goal, if it takes off it would stop me from feeling that I need to do this:

Signing up is easy. Enter your mailing address and Zumbox send you a letter with a verification code. Once verified you can start sending and receiving mail online. You can mail a few people for free, bulk mail is five cents per recipient.

Of course you'll only receive mail that has been sent to you via Zumbox. It's not a mail scanning service (like Earth Class Mail) so you end up with yet another mailbox to check.

Zumbox is trying to help businesses go paperless. This includes bills and other necessary communication. It also includes junk mail.

My experience so far is mainly junk mail. I did get a circular about recycling from Gavin Newsom but otherwise just a stream of special offers.

This is a big problem because Zumbox provides very little control over email notifications:

It's all or nothing. Either I get a daily email reminding me to go look at junk mail, or I get no notification at all (and might miss the next thrilling update from Gavin).

Most of my bills and statements are already paperless via email. This isn't as secure as Zumbox, but I'm not sure how much of an advantage this is as I really just need notification.

I really want to like Zumbox, but right now it's just another source of spam.