Ernies Tech Thoughts

I have seen a few list recently of essential applications people install on their PCs, so I thought I would put together my own list:
For web development

• Firefox, with the following extensions:
  • CSS Viewer
  • HTML Validator
  • Web Developer
• Notepad++
• Apache
• PHP

For Windows apps:

• NUnit
• SharpDevelop
• Visual Studio

General Development:

• MySQL
• WinMerge
• Subversion with TortiseSVN

Protection:

• Spybot
• AVG
• Zone Alarm

Miscellaneous:

• FileZilla
• Mozilla Thunderbird
• YAReg for dual booters needing to access Linux partitions
• CDBurnerXP
• 7-Zip
• OpenOffice

I noticed recently that stylegala.com was up for sale. This is one of the sites I would check regularly and I always found it full of good info about web design issues, especially after years away from web programming (I remember the days of Netscape 1 and Lynx…….).

The opening price was $30K but at last look $40K. It started from 1 person but a community grew and grew around the site. So who would buy this site and what could they do with it? I see 2 possibilities:

1. A corporate group buys the site. Naturally their focus will be a return on investment. Can more income be squeezed from the site without alienating its community. I would be suprised. And if the new owner doesn’t share the same passion for the site aims I believe many of the important users will drift away. Possibly 1 or more of these users will make the bold move to start a new community from scratch.
2. A large group of users chipping in small amounts and owning it collectively. Could this ownership structure survive and prosper? I would like to hope so, but a few core users will need to take the lead with out upsetting the other owners.

Even though this site didn’t start out as a business, it has become one. Do businesses survive when the original passionate owner moves on? My guess is this one will be swallowed up, and much of its character will be lost. The upside is that someone else, somewhere else will replace it.

As I talk to clients about their software needs, some are suprised there is another way to get software than shrink wrapped vendor supplied. I sat down and put together a small paper on how free and open source (FOSS) software can be a better alternative in many cases, and especially for smaller organisations. It is in pdf (90Kb) .

Today we released our first version of a bridge for punbb forum and Mantis issue tracking software. Both are GPL and so naturally is this bridge.

Recently builderau ran an article about Australian companies lack of contribution to open source software. I hate to agree, but overall I believe it is the truth.

Over the years I have worked at/with a number of firms (some big, some small) and many have used some free software/open source somewhere. Most of it is in development tools and server applications, hidden away from end users. So the fact the web server is apache doesn’t enter anyones conversations outside the developers/tech staff. Outside these staff and possibly their immediate managers, no one else is likely to know FOSS applications and tools have invaded the company. The manager 2+ levels up probably thinks php is from Microsoft or IBM, but they got a good deal on price. They are only using it for the cost reason.

This leads to the situation of companies not contributing back to the FOSS world, as these managers belive once the software is paid for ($0 is a price), then it is not their problem. There is no perceived benefit of sharing knowledge, skills and expensive staff time, that may have payback in the future. This years budget is the priority, vendor lock in, support and maintenance fees down the track are next years problem.

Government has also played a part overseas, and there have been rumblings from some Australian governments about the benefits of FOSS. OpenSource WA looks interesting, but otherwise it seems to be all talk without real understanding. From opensource NSW the document about Interoperability using open standards is in word format, and the site contents never seem to be updated. I know for a fact there are people in the organisation who are passionate about their goals, but they are surrounded with apathy.

Anyone who has searched through the massive list of projects at Sourceforge and other places will know they are full of interesting ideas that never got off the ground, or stall after a couple of releases. And it is not an outrageous statement that a group of developers with all the required technical skills does not necessarily make a successful project, especially when the target audience are not fellow developers. At the same time FOSS (Free and Open Source Software) has the ability to draw on such a diverse range of people in both background and abilities and the potential is enormous when properly harnessed and directed.

Through a link on a link I came upon the site for the book ‘Producing Open Source Software’: producingoss.com. I looked at the style of the first few pages and the authors name and it immediately made me think Subversion. This turned out to be true as Karl is an integral part of the project team and a few other well known projects. I have been impressed with the Subversion project from the outside, and we use it for all our source control, so in my opinion he is extremely qualified to write such a book.

The subtitle ‘How to Run a Successful Free Software Project’ gives the real crux of the book. Karl gives a lot of very useful ideas and tips on how to give a project the best chance of success in areas such as:

• Internal and external communication
• Project management style
• Release possibilities
• Licensing issues

From my experience with FOSS projects, the chapter on Communications is the minimum that anyone involved in such projects should read and he states ‘…[the] ability to write clearly is perhaps the most important skill one can have in an open source environment.’ (p. 84). He goes on to discuss writing style, content and how to keep discussion as productive as possible. The subsection ‘Avoid Holy Wars’ will ring bells from too many FOSS projects.

Chapter 8 on ‘Managing Volunteers’ is another essential topic, and an area where a project can quickly fall down if the team are strictly developers without a management bent. It is easier to lose a volunteer than get them in the first place, so treating them well in task assignment, appropriate permissions and sub system control is essential. He then moves on to the always thorny issue of forks both from the viewpoint of one from your project or being the person to fork an existing project.

All up there are 181 pages over 9 chapters and appendixes with links to various ‘free software’ version control systems and issue/bug trackers, and example bug reports. The writing style is easy to read and he draws on real world examples (with an expected bias to Subversion), so you can see how the ideas fit into a real life and sizeable project. He doesn’t get dogmatic or preach, it is jusr a good collection of advice from someone who has plenty of experience in the area. Criticisms - maybe an extra chapter on what has made some of the better known FOSS projects the success they are could have rounded off the book, but you can’t have everything.

I recommend this book to anyone already involved or considering becoming part of a FOSS project to give this book a read. The FOSS community needs this kind of book alongside all the code that is being written. A little reading now can only help smooth the process, and give any project a greater chance of success.

The web site has versions in html, pdf, rocketbook and XML source and a link to purchase the hard copy version from the publishers O’Reilly.

1. Successful Dutch football team
2. Ancient Greek hero
3. Brand of cleaning product
4. Acronym for over hyped repackaged technology

It is a quiet time work wise at the moment, so I decided to clean out some of the accumulated gunk that has has gathered on the hard drive. Apps, articles etc that looked good, get downloaded and not seen again for 6 months.

Saved away was a pdf of the book “Free Software, Free Society: Selected Essays of Richard M.Stallman”. I must say I didn’t read every essay to the last line, but it was an interesting read, if a little dogmatic.

I admit that I came to start using free software / open source software, for “…purely practical reasons…” as he believes most recent users have. To me, the freedom aspect was a bonus, which I have recently begun to better understand. The first time I installed a GNU/Linux distribution, I had no interest in diving into the source code and extending or modifying and redistributing. I wanted a solid OS, and that is what I got for a very good price. The same when I installed any of a number of other apps such as Firefox, Apache, MySQL (the list goes on). But recently I have been looking at extending/forking an existing app which is no longer being maintained. If this had been a proprietary application I wouldn’t be in a position to do so. Fortunately the license is GPL so I have the freedom to do so. Maybe I am wasting my time and the Darwinian ’survival of the fittest’ nature of the marketplace, had already weeded out an inferior product, but I don’t think this is true. So build on the work of others and reinvent less of the wheel. If it fails it doesn’t matter at least there was the opportunity. More details to follow once the alpha is ready.

Back to the essays. The term “free software” has lost the battle with the term “open source”. This is my belief, not his admission, but I see his frustration in that what he has long advocated, has been narrowed from his vision, and become the dominant position in the wider community. “Free Software” has just conveyed the wrong message in its title for most people. It has a long history, but is it to late to modify the term “free software”? To me the word liberty makes more sense, but was apparently rejected long ago. To me a change may reinvigorate the discussion on the real freedom of software, rather than the marketing use of the term “open source”.

With OpenOfffice 2.01 being released last week, I have installed it, and removed both OpenOffice 1.1 and MS Office. So far it has all been positive with just a minor issue with a complex graph from an Excel spreadsheet.

One thing I did notice, and slightly concerned about is the need for the java runtime to have OpenOffice run [I don't remember this from v1]. It doesn’t cause me a technical issue as I already had the runtime on the PC, but is it really necessary. If Java was not a product of Sun I may have never given it a second thought, but it is. Maybe there was some outcry of this coupling, but I didn’t hear it. Possibly MS should take them to court.

Streber (a web based project management app) has made its latest release. There is still more work needed before a production release, but so far it looks very good. I have started using it internally for project tracking. Streber was originally to be the next version of netoffice (which I had been using), but all the code has been written from scratch.

Give it a look if you need a web based project management tool.