Erik Petersen's blog

A new take on GPS voice bugs

Mon, 10 May 2010 10:01:40 +0100

It is a type of testing I had never imagined, testing pronounciation and even patience. If you were recording voiceovers with a celebrity voice who hadn’t done it before, how would you correct them if they mispronounced something? You could send them a checklist perhaps beforehand.

What if they were intimidating and known for their short temper and meglomaniacal tendencies? A famous celebrity recently was recording a GPS voice over, and was saying “roundabound” not “roundabout”. Luckily a tape of the incident has been released and is now being watched around the galaxy…..

They may have even more problems later when they try the same thing with a linguistically challenged little green man ...

De-smelling by refactor or rewrite: some thoughts

Sun, 17 Jan 2010 01:14:39 +0000

Is there a tipping point when rewriting code should be favored over refactoring ? It is often hard to decide. There are some important factors that can help. Code smells cluster, whether as bugs that are detected, or potential failures lurking in immature code that may break during future updates or maintenance. Steve McConnell wrote (in “Rapid Development”) that “Error prone modules tend to be more complex than other modules in the system, less structured and unusually large. They often were developed under excessive schedule pressure and were not fully tested”. He also quotes indicative studies that these modules are much more expensive to develop. Steve’s metric is 10 defect per 1000 lines of code to mark a module for further work. If it appears to be an error prone module, he advises rewriting it. Other indicators are large concentrations of bugs, large number of checkins, or large amounts of fix time spent in comparison to similar modules. While error-prone modules are typical in traditional development, agile practices like well executed TDD should minimize them. High cyclomatic complexity of code is also a good indicator of the need for simplification.

The hard thing is determining the tipping point between refactor and rebuild. The easy thing is that because of clustering, there is often a lot of low hanging fruit, and focused effort on a small portion of the code should have a massive impact on the code quality.

For more on bug clusters and other productivity ideas, see my Google tech talk “Building Software Smarter”, http://www.tinyurl.com/80-20-rules

Severity One outage - on a golf course??!!

Sun, 06 Dec 2009 08:39:09 +0000

I never really realized the parallels between software development and golf course layout. Both have a technical component, and a reliance on usability. The Australian (Golf) Open moves from city to city each year, and has a hiccup every so often. This year they had a five hour outage.

How do you get an outage in golf? While not in the infamous class , this years competition was held at a cliff top course in Sydney overlooking the ocean. Spectacular but windy. A savage wind blowing balls the width of the green had players blowing par just with their putters. A 5.5 hour delay to play was called by which time the wind speed had halved. The outage confused the local players who said those conditions were typical and just “par for the course”. The greens were all very fast, 10.5 based on their stimpometer score, much higher than the 8 recommended by some players before the tournament started. While the golf club staff ignored the advice, they had tried to slow 4 of the faster holes. The rest of the tournament was played sucessfully in windy conditions that kept the flags taut but didn’t move the balls.

If you could only follow one agile practice, what would it be?

Wed, 25 Nov 2009 03:44:17 +0000

At the MXUG meeting this week, Steve Hayes was volunteered into giving a lightning talk. He asked for a suggested topic, and someone said “If you could only follow one agile practice, what would it be?” My instant comment was “Standups”. Later on Steve spoke about Kent Beck’s revised agile values condensed down to the trio of Accountability, Responsibility, and Transparency (which has good and bad aspects he said). Any practice that supported this would be good, but Steve chose retrospectives. I added my 2 cents worth that I would choose standups if quality was a major goal, as studies have shown that lack of communication in groups creates bugs particularly in larger teams. Linda Hayes has written about this, as has Mike Cohn in Chapter 10 of his new book . Standups also function as mini-retrospectives as well. They start conversations that continue after the meeting to help team members get better understanding of what they are delivering.

Standups have now moved beyond software development and are infiltrating other industries; I recall reading about the new HQ of an advertising company that has meeting rooms without chairs and bar counter tables to only allow standups (and cocktail parties!)

Cool Tools - for customer data etc

Thu, 05 Nov 2009 13:24:07 +0000

Test data is often a bottleneck on testing projects. Security and privacy issues can require complex massaging of existing information, Today, there are tools to make it easy to create it from scratch! I find two in particular quite useful.

I’ve been telling people about generatedata.com for several years. It is an amazing tool for generating data for customers or other items, including items chosen from lists and data ranges. It is free, customizable and can even be run off a USB (after it is customized for the computer and database etc).

Lewis Hamilton, agile tester?

Tue, 28 Jul 2009 02:00:44 +0100

Lewis Hamilton became the youngest Formula One racing car champion last year, in spectacular circumstances with a last gasp fifth place in the last race. This year, as tabloid headlines screamed “hero to zero”, his success has been measured by trying to avoid being lapped by the leaders and ending a five race drought without winning a point.

Apollo 11 glitches: the alarming 1202, the 4 sec fastforward and the soft landing bug

Mon, 27 Jul 2009 06:38:37 +0100

It's forty years on from the moon landing, the culmination of the work of 400,000 people. I can remember as a kid seeing the hazy picture on TV (One of the few bits of TV I saw as a kid, as I grew up on the Pacific island of Nauru but must have been in Australia on holiday).

Sluggish starts slow subsequently successful Service screening streams

Thu, 09 Jul 2009 07:24:08 +0100

That title could be from a page in Graeme Base's book Animalia. It is an amazing picture book, with a small boy hiding in the inventive illustrations on each page. It has recently been turned into a cartoon and is used in creative writing classes in schools as well. I don't know if Michael Jackson's children know of the book but their late father would have loved it, like he loved Peter Pan.

Breaking News breaks records across the net

Fri, 03 Jul 2009 10:00:30 +0100

The sad passing of Michael Jackson became a breaking news story that sent a shockwave spike through the infrastructure of the internet. Google saw a huge spike in traffic that security threshold detectors interpreted as a hacker attack, resulting in a 25 minute captcha confirmation or service refusal from Google News. Yahoo had the MJ report headlined in their news aggregator, then Google had it further down, and Microsoft's Bing had it way way down. A Bing blog post seemed to imply that it was ranked down because it did not have images with it, which breaking news typically wouldn't.

The website TMZ.com broke the story 20 minutes after his passing, then went down briefly under the load. The LA Times was the first news site, and it fell over soon after. Online news services saw a 20% increase in traffic, but the availability of popular web news sites dropped from 100% to 86% and didn't return for another 4 hours.

Asia Pacific CITCON does Brisvegas!

Thu, 04 Jun 2009 01:58:48 +0100

After a great conference in Sydney in 2007 and Melbourne in 2008, The Continuous Integration and Testing conference (CITCON) is coming to Brisbane in 2009, on June 27th. It’s an open space conference where we all meet the night before (June 26th) to propose and vote for sessions (and arrange the session schedule) then we socialize, then we come back on the Saturday and enjoy sessions of demonstrations, discussions, BOF, etc interspersed with eating and socializing again. Participants said it was an amazing experience both years, so may be scintillating in the Brisvegas sunshine.

This is a free conference, running on sponsorship (Special thanks to this years sponsors!). Check out the details including registration at http://www.citconf.com/brisbane2009/

We’ll be at the Acacia Ridge Conference Center, a short distance south of the Brisbane CBD. On a slightly serious note: It may be free to attend, but it still costs money to run so if you do register but can’t come, please tell us so we don’t overcommit on the catering etc, and someone else can take your place (and your t-shirt and show bag!). Folk from the US, NZ, India and around Australia are already registered. Join us for some great CITCONning!

Short and long game thinking, tests driving design and CRAP metrics

Thu, 28 May 2009 05:59:56 +0100

Kent Beck recently posted To Test or Not to Test? That’s a Good Question on the complex “theory versus practice” issue of always automating tests, where he states,”Then a cult of [agile] dogmatism sprang up around testing–if you can conceivably write a test you must”. By classifying projects into long game and short game, he argues that ROI becomes a major issue on whether a test stays manual. He says “Not writing the test for the second defect gave me time to try a new feature”, but several people commented that this was a technical debt tradeoff, and Guilherme Chapiewski noted he had done the same thing with a Proof of Concept that went live then he had to rewrite major chunks later. It is interesting that this ROI discussion is reflecting the experiences of the pre-agile functional automation community. Back in November 2001 (Wow! Long time ago!!), I posted to the Agile Testing list some considerations for not automating . While many of these were from the context of two separate development teams and the automaters using expensive test tools, the risks of incomplete automation and insufficient ROI dominate. The benefits of having the same people both develop the code and the tests are great, and beyond my experience when I wrote that post.

I think the ROI issue for code-based tests will go away over time. Much of the creation of code-based tests is mechanical. Just as programming languages replaced assembler and took care of fiddly details (what registers to use, low level comparisons etc) and build utilities replaced simple text file include statements, I think that soon it will be standard practice to have tool-created unit testing to handle mocking, dependency injection and assert-based testing. Mocking was originally very manual, then tools were developed. Dependency Injection was very manual,then tools were developed. For assert-based testing, we’ve already seen Agitar’s tools , zentest and now pex amongst others. I think these tools will become standard, just as coverage tools are now standard in IDEs when they originally were luxuries costing tens of thousands of dollars. Another variation of this is tools like Celerity recently blogged about by Jeffrey Frederick. Celerity is a fast way to run GUI web tests, but could be handled as a mechanical translation not a manual one. Some meta language could generate Celerity and selected browser tests in a single step.

Mechanically generated tests are cheap to produce and overcome ROI issues. However, they only reflect the current code. The benefits of test design infusing the coding approach are missing. If tests are not being automated for whatever reason, some analysis of the refactoring risk should be done, at least to know where and what the error-prone code is. One way of doing this is using the Agitar-created CRAP metric , which Bob Martin recently blogged about as a way to keep design clean. While I currently believe all code should be created test first wherever possible, techniques like the CRAP metric can highlight the complicated bits for refactoring where possible. While it may be a great intellectual challenge, there is no need to refactor a complex industry standard algorithm. [Aside: is there an inherent advantage to doing test first design all the time? Perhaps, just as renaissance masters only painted and sculpted hand and faces and left the rest to their workshop staff, we only need to focus on core functions for test first and do the rest test last?]

As Kent says,”By insisting that I always write tests I learned that I can test pretty much anything given enough time.” Time is often a rare commodity, so Kent argues compromises are often needed in short goal projects. As Ron Jeffries said in a comment on Kent’s post, “My long experience suggests that there is a sort of knee in the curve of impact for short-game-focused decisions. Make too many and suddenly reliability and the ability to progress drop substantially.” I hope that advancements in mechanical generation of tests don’t push us into a short game perspective, impacting the use of hand crafting tests to drive design. At the same time, metrics that can be run as part of the build to highlight areas for refactoring on all projects are proving valuable (and I’m looking forward to state coverage ). By any measure, these are interesting times we live in. Long live long game thinking!

Wolfram Logic Bugs

Thu, 09 Jul 2009 07:44:00 +0100

While Wolfram Alpha seems to have been endowed with both a sense of humor and self, there are some curious logic bugs. Try this, enter Name kim stacy terry jan frankie. WA shows the popularity of each name in the US census. It also asks whether you meant the male or female version of each name. This is very intelligent! It also says it assumes it is a list, and offers to do a multiplication instead???!!!! Clicking the multiplication link returns the standard “dont understand” message.

My favorite (so far) relates to two movies based on books. Enter Name tom jones . WA knows this is not just any two names. It assumes “tom jones” is a book (with a movie link option as well) then says it is a title, then the result is Data not available . Mmmm, let’s understand this a little better from the WA point of view, I assume it is a book, Yep, It’s a book title that I know, Nah, forget all that, too hard, no idea at all! It does find it as a movie though.

If we try another entry name color purple we get the same problem: no book data but normal movie data. We can guess what is causing it: there is a record of a movie of the book but no details on the book itself yet. Hopefully the book data will be “made available” soon….. However, if you just type in Color purple , you get information on the color purple, with a link to the book or the movie. Clicking on the book link then show full details of the book. Huh?

OK, let’s just try Tom Jones . Now we find out he is a singer, but we also have a link to a book or a film. Clicking on the book link tells us now that there is some limited data: the author is Henry Fielding. As Alice in Wonderland said, “Curiouser and Curiouser”, I try Alice in Wonderland , and there are 5 movies, but no book!

Finally, I tell WA “Im not sure what to do with your output” to which it replies “WA is not sure what to do with your input”!

PICT rules! (and so does 80:20!)

Wed, 13 May 2009 07:04:36 +0100

I finally got to do some serious combinatorial test design using PICT (download it here ) defining a series of rules covering a multi-stage approval workflow over a group of input queues. What could have taken several thousand tests reduced to under 40 after some experimentation to define the rules, and after several tweaking runs to optimize the mix of tests (including a seed file to get a successful case for each input queue), there were 2 tests for each of the core queues, and one for each of the others, as well as a mix of failure situations. It is an amazing tool that I want to keep investigating further. Thank you Jacek for your work on PICT!

Unlicensed confusion, by design?

Thu, 30 Apr 2009 03:55:58 +0100

Context is King, a picture’s worth a 1000 words & Jazdy Prawo is a Polish master of disguise driving around Ireland. This is a tale of ambiguous requirements, poor design and PICNIC (explanation soon) . In 2007, the Irish Police wrote a confidential internal memo to clear up some confusion- Police trying to get details of traffic offenders with Polish drivers licenses often found they had stopped Jazdy Prawo, and details of more than 50 of them had been recorded with different addresses and even sexes!

The end of printed Next and private blogging

Wed, 08 Apr 2009 12:50:08 +0100

More than 20 years ago, a now defunct magazine called “Computing Australia” had the final column by Graeme Philipson talking about how Mainframes would have a long lifespan. I wrote a letter at the time saying how it was a pity he chose a dead duck for a swansong. Now, I have to acknowledge he was right and I was wrong. A week ago, Graeme wrote another final column that had featured weekly for 10 years in the IT pages of the Tuesday edition of a local Melbourne paper, “The Age”. The section, renamed Next several years ago, was many pages long in its heyday and was the source of several of my jobs over the years. Now not only are the ads on the net, so is the whole mostly-national IT section. As Graeme observed, “IT journalism is moving to the internet”. The strange thing is that the Net seems to have no content from “Computing Australia”.

I wrote many blog posts last year with pen and paper that were good writing practice that I never got round to publishing, I hope to read many more posts from Graeme if he continues , and I will try to post more as well. Graeme’s last topic was the internet and communication. May we both use the former for the latter for many years to come!