Random Thoughts on Software

This isn’t about software per-se, but I want to propose my solution for this whole music industry copyright quagmire.

Record labels have traditionally had several purposes. They find musicians, they advertise/promote on behalf of the musicians, they record music, they put the recorded music on the latest-greatest tangible media (audio tape, CD, DVD, video tape, etc.), and they distribute the tangible media to a store to be sold.

But today, two things are very different. First, tangible media is becoming unnecessary. We are rapidly moving away from obtaining information on tangible media. Tapes, CDs, video tapes, and DVDs are prone to loss, theft, and damage. With hard-drive and internet-based storage becoming so cheap, why would anybody want to deal with tangible media in the next 5-10 years. Second, the market demands the free distribution of digital content. Whether information distributors want to admit it or not, the market views the distribution of content as a commodity. Without selling a tangible product, the record industry cannot expect to make a huge profit from distribution.

It seems to me that the RIAA (and friends) are having a hard time adapting to these changes. Its especially difficult for the record industry because they are used to overcharging people, and now they can’t charge $17.00 USD for $0.03 USD CD with 10 songs on it…. 7 of which suck.

IMHO, the record industry needs to quit its whining and start adapting these changes. They will never (ever) be able to control the distribution aspect of the market again, and they just need to face it. It seems to me that they are wasting time and money by trying to regain something that they will never be able to.

But, just because distribution is a commodity, that doesn’t mean that people are not willing to pay for the content itself. I still think that people are willing to pay for music that they like. How can money be made off of freely distributed music? Commissioning.

In a commissioning model, the record label does everything that it used to do, except it is no longer making a huge markup on the distribution of music. They would promote artists, and the listeners that enjoy the music would (as a community) commission the artist to release more music. The music would be freely distributed to the benefit of everybody. The more people that listen to it, the greater the potential commission for the artists next release.

There are numerous ways one could implement the commissioning model. A popular implementation of this model would probably be releasing songs only after a certain amount of money is collected.

I think the same model would work for digital art, movies, and any other type of digitally transferable art form.

In an earlier blog entry, I pondered the question of whether or not XDoclet was a necessary tool. Norman Richards has a wonderful response to my question.

I use Microsoft Outlook for all my e-mail and contact management, so I’ve been looking for a news aggregator that integrated well with Outlook.

I tried intraVnews and I was fairly happy with it. There are a few problems, but its a pretty good product overall. The main problem was that every news item is pulled in as a seperate item (like an e-mail). This was a little annoying because I subscribe to about 30 different sources and I get a lot of news updates. I wanted to be able to scan all the new headlines from a particular source without having to delete each individual item after I was done with it. Yet, I still wanted the ability to save items that really interested me.

I really like the simplicity of www.bloglines.com. This is a Web-based aggregator that allows me to save items and to scan headlines. And, because it’s Web-based, I can access it from anywhere while still preserving my blogroll, update/session history, saved items, etc.

I have even integrated it into Outlook. Here’s how:

In his Weblog, Perryn Fowler recently advised to “Make it easy to use well, not hard to use badly“. This makes a lot of sense to me and reminds me of another bit of wisdom that I’ve heard from sales people, “cater to your top performers”.

Wisdom from the realm of sales advises sales managers to focus their training budgets on their top sales people. For example, many sales managers will spend their money sending their top sales-people to training seminars, while not training the rest of the sales force. This is beneficial to them because of the 80/20 (aka 70/30 or 90/10) rule: 80 percent of their revenue is produce by the top 20% of their staff. Despite complaints about inequality and unfairness, sales managers realize that it is probably better to send their top performers to a really good week-long sales seminar than to send everybody on the sales team to a mediocre one-day training session. The top performers will get more out of the training, will use more of what they learn, will generate more money for the company, and will “trickle down” the knowledge to the rest of the sales force.

Some discussion has taken place on whether the 80/20 rule applies to software development as well. People like Alistair Cockburn and Fred Brooks talk about similar concepts as well.

Here’s where I think these two concepts merge: make it easy to use well so that your top-developers can be more productive. One could probably make this argument for most aspects of software development: language, tools, API, UI, etc.

There is, however, one case in which I would rephrase Perryn Fowler’s advice. When talking about software that is responsible for human life, I would say that a UI must to be “easy to use well, AND hard to use badly”

I just added comments to my blog using Halo Scan.

Update: I am now using blogger.com’s new commenting capability.

I heard a news headline on the radio this morning about how some people in China are finding ways to circumvent the governments attempts to restrict internet access. I got to thinking about the possibility of a Web-based Web browser. After doing a search, I was quickly reminded that just about 99.99% of any software ideas anybody comes up with, somebody else has already written or pondered about.

The way it would work is that you would have a server that has no filtering/restrictions that exists outside of the restricted network but is not blocked by the restricted network. A computer from within the restricted network would then tunnel its requests through the server. It’s a pretty simple concept really. And, the advantages are more than just access to restricted sites. If the server is setup to be anonymous, you can provide truely unmonitored (private) Web searches.

A browser toolbar that tunnels the page requests through another server seems to be the most elegant design, but the problem with that is that you would have to be able to install software (i.e. the toolbar) on the computer that you are using.

Another approach would be to put the brower into full-screen mode and then actually emulate the browser buttons, address bar, bookmarks, etc. This could actually turn into an interesting service offering. With this model, I can envision the emergence of an ultra-thin Web browser that has no functionality besides connecting you to a Web-based Web browser which renders its own browser interface (using flash or something) along with the content that it displays.

I have two new articles online:

GZIPping with Java

JBoss Meets Eclipse: Introducing the JBoss IDE