Blog Yellek

Worse Than Failure has departed from their usual fare of tragic but funny stories from the IT industry to comment on The Great Pyramid of Agile. Agile development has always been one of those oases off in the distance for me: “Real developers do agile and everything else is old fashioned and inefficient”. I have spent more than a little time wishing that a current project was more agile and that we weren’t wasting so much time on obsolete specifications and schedule readjustment.

There are, however, naysayers saying that agile development is just a pipe dream and that we should as developers stick to the methods that we know, even if they are less than perfect. Who is right?

I suspect that a lot of the ideas that the agilists promote like test driven development and continuous integration are valuable contributions to the field of software engineering. I also think that software development is such a complex process that there will never be a definitive answer one way or the other, no matter how the proponents on each side want there to be.

The problem is that humans get involved and what started as a discussion of new ideas becomes a political/religious debate with entrenched views on each side. Developers in the trenches need to be more pragmatic than that. There is software to write and projects to complete and they will use whatever they can to get the job done.

The key thing to remember is never to stop learning.

I was just looking on the FreeMind site (earlier coverage of FreeMind here) when I noticed that one of the features of the upcoming 0.9.0 release was the ability to execute Groovy scripts from mind map nodes. This raises the possibility of all sorts of interesting applications. One that sprang to mind was to enhance the GTD FreeMind template from YekSoon with scripting functionality to provide a better GTD tool.

Well done the FreeMind team, I’m sure that this new feature will see a lot of use and Groovy is certainly a good choice for a scripting language.

I’m currently attempting to implement some of the GTD principles in my life and I’m attempting to use a free hosted version of the Tracks software to do it. I have to say that it is rough round the edges so here are the 10 things I hate about tracks (as hosted at http://tracks.tra.in):

1. There is no way to set a user timezone. This means that the UI thinks it is yesterday until the clock ticks over in the timezone where the server is hosted. Perversely the javascript date popups get it correct, probably basing it on my PC. Inconsistent and bad.
2. There is no way to sort next actions by anything other than the data they are due. Sort order within that is always by the order they were entered. This means that there is no way to sort by project within that.
3. There is no priority field. This means that you are ending up with an amorphous collection of actions, especially within large contexts like “Work”
4. The UI often glitches up meaning that completed tasks often end up in the to do list meaning that you have to “complete” them all over again until the data gets saved correctly.
5. There is no way to attach files like mind maps to projects, only text.
6. Projects can’t have due dates so the UI can’t display them or check the next action’s due dates against them.
7. The CSS has only been designed with Firefox in mind, there are significant errors in IE6.
8. There is no support for the agenda functionality mentioned in the GTD book other than through setting up another context. If I need to talk to someone in the work context then I want that listed under work so I can see all of my work items together, not completely separately.
9. There is no sort order on contexts, they appear in the order you enter them.
10. Although there is some AJAX used there are a lot of screens that require a page refresh. The whole UI is poorly laid out and doesn’t look anything other than clunky.

All of these make tracks very difficult to work with but since there are no other alternatives at present I’m using it for now. I only wish that I had the time to write something myself. I have a whole bunch of ideas for features but I can’t see myself getting to any of them any time soon.

The article Fear and Loathing in Typeland from the O’Reilly OnJava Blog expresses really well my reason for preferring Groovy to Ruby. One of the problems with starting something new as a reaction to something old is that whatever has gone before is automatically bad. In this case I think that Java does have some advantages, as Robert Cooper points out, and that the Ruby guys just need a little maturity.

The Groovy crowd on the other hand seem to me to have a healthy dose of humility in all they do so that when it makes sense to use Java for something they say so. Admittedly there is still a bit of “We are better than Ruby” which needs to be watched but when it comes to the community I think they are right.

Ruby is certainly more popular but I for one, perhaps irrationally, am not interested in spending my time learning it because of the negative vibe I get from Ruby people.

I just started working in a different department of my daytime source of income generation and of course made a point of checking out the state of local wiki’s. It turns out that there is a Media Wiki installed which is something I haven’t used before but SnipSnap was getting a little long in the tooth so what the heck.

Wanting to create a table I read the Media Wiki table help and was absolutely horrified. Really this is no better than typing HTML table tags. There seems to be a 1 to 1 correspondence between HTML markup and wiki markup. Contrast this to the expressive and elegant Twiki table markup which achieves much the same functionality with a lot less code.

Even with the javascript syntax assistants in the media Wiki edit box this whole syntax is really just a waste of time and Twiki has had the same syntax for at least 5 years that I can remember.

powered by performancing firefox

Dan’s Data has an interesting piece on how video games will bring about the end of the world where he talks about a future where the virtual world is so compelling that people will become totally absorbed in it to the exclusion of participating in the real world. He then postulates dire consequences because only religious extremists will be able to resist the allure of these games and he doesn’t want to live in a world where the extremists are in charge.

It’s an interesting idea. I’ve talked before about my decision to give up games and maybe I’m fulfilling what Dan is talking about. Christianity, after all, is about dealing with the world as it is rather than retreating into a virtual world. Sin and the pain associated with this fallen world is very real. If people are retreating into their virtual worlds to avoid dealing with it there is no resolution. I feel the allure of that retreat very strongly but I know that the only way of dealing with my sorry state is not to deny that it exists but rather to acknowledge it before Jesus and allow Him to forgive me. If p[eople are retreating into virtual worlds how much more difficult will it be to reach them for the gospel?

It’s an interesting thought.

Looks like Redmonk have employed a new analyst: Anne Zelenka who has a new Redmonk Blog: Tech Decentral In one of her initial posts she mentions What I Will Cover which is an interesting list given that it covers a lot of technologies that pique my interest. I do like both James and Cote and I read their blogs but I find them a little dry for my taste at times so it will be interesting to see what Anne brings to the table. Certainly Redmonk Radio should be more interesting.

I notice in her list of technologies that she will be covering scripting languages but I see no mention of Groovy or Grails. One day I hope to see these two technologies included in such lists as a matter of course but until then I can only hope.

Les Ey has put up an interesting piece on the relationship between software development and evolution (How Complex Information Systems Point to the Existence of God ) and, while I agree with most of what he says, the following caught my attention:

While I was studying Computing, the lecturers would emphasize the importance of a stage in developing computer programs that is called the “Design Phase”, they would emphasize the point that time spent designing a system more than paid for itself when you went to “code” or build the system. I’m yet to hear anyone suggest that I should not worry about designing a system but just go for it and let it evolve.

I’m wondering if agile development is really “just letting software evolve” or whether there is design involved but just during the process. Agile development certainly eschews the need for a “design phase” as such with design occurring continuously throughout the software development process. Is this “evolution” of software or not?

I was reading a post over at Steve Gillmor’s Gesture Lab today and I found myself making comparisons between the goings on on the Gillmor Gang and things I remember from primary school. Mike Arrington and Steve are constantly on the edge of leaving. People don’t turn up and don’t say why. Steve wonders at the significance of who didn’t turn up when and whether he will get an e-mail about it. Jason Calacanis is starting his own show and Steve wonders if it will be more popular than his, the insecurity is a mile wide. Hugh Mcleod gets banned for misbehaving.
When the gang do turn up they bicker and I for one don’t enjoy it when they do. I find it so much more listenable where they disagree in an amicable manner rather than acting like spoilt children. I don’t want them to agree, I just want a more pleasant listening experience. I don’t enjoy conflict in my life, however necessary it is, and I don’t want to listen to other’s conflict when I’m trying to be entertained and informed. Don’t get me wrong, the content, when it happens, is top rate and often really funny. I just wish they all would stop acting like little kids sometimes and more like the brilliant insightful creative visionary adults that they show themselves to be in between.

In my daytime income producing activity I get to edit a lot of documents. Most of those documents are based on templates of one form or another. Over time I have developed a list of the things that I hate about various templates and I thought I’d share them here for template designers everywhere.

(more…)