the Design Experience Weblog Archive

I coverted a graphic into a favicon.ico file for openacs.org. This shows up next to the URL in the location bar in IE and Mozilla, and next to the page title in the Favorites menu in IE. I used IconCool Editor to create it. This is a nice little pixel editor with most of the tools you would expect if a graphics editor, plus quite a few pixel level editing tools. Too bad it is focused primarily on icons. The tools would work great for bitmap textures used in 3D rendering.

Favicon.ico files are a collection of icons at different sizes. The most important is 16x16 pixels using the 16 standard Windows colors. This is a very difficult medium to work in. Without this size and color-depth IE will ignore the icon.

Of course designing a CMS is hard. It is not the difficulty of the programming so much as the decisions you have to make. Do you want it easy to use or full of features?

I am working on the edit-this-page package for OpenACS 4. Originally designed and programmed by Luke Pond, it started out as a simple way to quickly add content and build custom content types. Myself and a few others ran into limitations with the original, so we started work on version 2. Now it is hard decision time. I like the idea of a system that allows you to create a new content type right in the browser, without getting into the code. This adds complexity, and can limit the possibilities. You can never imagine every use it will have. That was the problem with the original version. I had some different ideas about how pages would be structured. On the other hand, OpenACS is a toolkit for programmers, so a little programming to get the results you want is assumed. The question then is what tools to provide the programmer to make his job easier.

So we decided to go back to basics, and build an application that provided very simple support for content and made very few assumptions on how you would use it. If we can build a good assortment of sample applications, it should help to show the possibilities

The strength of OpenACS is in the ability to combine application packages and service packages to build new things. I like the idea of smaller simple service packages that are combined in different ways to create a larger application. This reminds me of the UNIX philosophy of small tools that do one thing only. This allows for the greatest flexibility. Designing the interfaces to one of these tools is the tricky part. You have to think of the possible uses of the tool, and make the API as flexible as possible. If a programer can't build something you had never thought of, it is not flexible enough.

Later I will write some more about my ideas for the types of services I want to provide to application builders, myself included. I want to be able to mix and match features as required.

As OpenACS 4.5 nears release, I am thinking about where it can go. Now that the core is stable, what sort of applications can be built on top of it? Already there is dotlrn, a course management system build on the OpenACS toolkit. Certainly many web sites have been built using the toolkit, but the developers all seem to busy to share their success stories with the community. Of course, being too busy with paying work is not a bad thing.

So I want to start collecting information on how OpenACS can be customized and extended. Is it a great platform to work on. The core services of users, groups, permissions, and content repository make a great foundation to build on. Standard add-on packages such as bboard (discussion forums), content management, faq, news, press, and ecommerce are available, with more being developed every day.

I am most interested in expanding OpenACS to support knowledge management type applications, allowing the collection, categorization, and search of user-submitted content. These basic tools can be used by the largest corporation to the smallest community group. I especially think local schools and libraries would benefit from the ability to gather and share information.

If you are interested in building a community focused web application email me: dave@thedesignexperience.org

Special thanks to Duncan Smeeed, Jenny Levine, David-Carter Tod,Philip Greenspun, and the entire OpenACS Team for the inspiration.

How Can You Start?: an updated several step plan to becoming a programmer. I agree that Viscual Basic for Applications is a real programming language, and can learn how to program with it. Here is a list of books I recommend to learn to hack MS Access.

I keep seeing books for programming web sites using MySQL and PHP and I wonder where the OpenACS book is. My guess is that the OpenACS hackers are busy building web sites.

I still think a "book" of some sort might be a good idea. Not just a rehash of the technical documentation, but a narrative of how to build something new with the OpenACS toolkit. Now I just need to enlist some hackers who have actually build something on OpenACS 4 to submit a paragraph or two.

I should post this idea to the bboards, but I need to think about it a little more and focus on what kinds of information I should collect. Of course, possibly I should just see what people can write up, and then figure out how to make it into a "book".

I should stop saying book, what I really mean, is a long document, seperated into chapters. Most likely it would be in HTML, PDF, or some other electronic format. I have no delusions about producing and actual print volume. I just want to gather up all the knowledge trapped in hackers heads, and distributed in the bboards and private email, and organize it so someone who is unfamilar with the toolkit will understand the potential.

I need to make titles optional.

I am glad Karl Martino is planning to update his How I Got a Career that I linked to earlier. I like the way it was written. Sharing my path to a career is what this web site is all about. I keep track of my wanderings and very slow learning process so hopefully someone else can learn from this.

I am sure in 2 years everything will have changed again. It seems like software development is changing under my feet. I see alot of interesting projects that I might like to participate it, yet I am also trying to increase my general understanding of basic software development concepts. I think that some of it I understand, but just do not know the formal names of everything. Philip Greenspun said SICP was good to learn to program, or to learn to talk like a programmer if you already knew how to program. I think that explains the stage I am at right now. I understand many of the concepts, but I don't even know why or when I need to use them.

More Later.

Noodling around on the internet, I found How I Got a Career, by Karl Martino, an interesting story of one man's path to becoming a programmer. It lets you know that anyone can do it, if they have the desire to learn. A couple of points I found that I would expand on a little:

Buy a PC if you don't have one. I don't care if your school has one... you need one at home to practice on. It doesn't matter as far as learning goes what type of PC, although there is more support out there for Windows programmers.

I would say: Buy a PC. It doesn't matter what type of PC, although there is more free support out there for Linux, FreeBSD, or other variations of UNIX.

Buy a programming language compiler...

Download a free language.

You have probably heard of these popular ones by now - C, Pascal, Visual Basic and Java. You can buy these compilers from great software makers like Microsoft, Borland or Symantic, right from CompUSA. Each one has lessons and tutorials to get you started.

You have probably heard of these popular ones by now - C (gcc), Perl, Python, Tcl, Java (yeah I know, its only free in $$), Ruby, Lisp, Scheme, and on and on.

I guess my main point here is, any computer will do. There are free programming languages for virtually every computer available. The most important point, is to just do it.

As Karl says: Write a program. It will suck. Don't worry, most software sucks. Just keep going. The second most important thing is to read other people's programs.

This point Karl did not make, but it might actually be more important. I have learned more reading others code than I ever did from a book. That is why open source software is so great. Sharing code is sharing knowledge and learning.

Another benefit of open source is the opportunity to work with other programmers. On the OpenACS project, I started out with basically no skills. From this I have progressed to being a useful member of the team. I am still learning, or course, but the opportunity to learn from more experienced hackers is invaluable.

I should write more on this topic, but my brain isn't up to it right now. Check back later.

The Rise of the Creative Class, Richard Florida: an interesting article that suggests the percentage of creative workers in a community has an effect on the economy, and states that the factors to attract these workers are not what you would expect.

This is an interesting article. I just skimmed it. I did notice that Albany, NY is #2 for creativity among metropolitan areas between 500,000 and 1 million people. There is also a book and consulting/speech business behind it.

I updated the RSS link tags to conform to the new style. This is supposed to be the last change, and the format will be frozen. Now tools can read this tag from a URL and get the RSS version of the page. Pretty neat.