Comments for Semantic Werks

Comment on DeMarco and “Cannot control what you cannot measure” by John Hunter

John Hunter — Sun, 29 Apr 2012 03:56:43 +0000

When metrics are used to aid learning they can be beneficial. When they are used to set goals tied to bonuses or indirectly tied to money via performance appraisals… they mess things up. The focus turns to meeting the number not doing the job. http://management.curiouscatblog.net/2004/08/29/dangers-of-forgetting-proxy-nature-of-data/

Even when metrics are used for learning they can lead to all sorts of trouble when there is not an understanding of variation. Thankfully software developers are more likely to have a basic idea of variation than MBAs, but it is still a big problem.

http://management.curiouscatblog.net/2006/05/09/understanding-data/

I very much like “Let us stipulate that there are endless examples of low-maturity teams out there whom no technique will help.”

Comment on Using GitHub for 3rd Year Software Engineering by Neil

Neil — Thu, 26 Apr 2012 21:35:55 +0000

No, no data to hand, although the standard approach (Google web search comparison) shows 245M results for Git and 3M for “IBM RTC” (yeah, 7 years of PhD and this is the best I can do . I feel the same as you re: OSS. In general, my hunch is that employers would like someone who at least understands version control principles. Specific tool practices can be taught on the job, I think.

Comment on Using GitHub for 3rd Year Software Engineering by planetpolly

planetpolly — Thu, 26 Apr 2012 19:50:16 +0000

Our software team transitioned about a year ago from SVN to Git / Github. Thinking about it from a teaching perspective, I would way rather have a separate version control tool, outside the IDE, to start with – it makes it clear what the tool is doing, and then when a student is later shown IDE integration, they ‘get’ that is is making their lives easier, but they better understand the underlying mechanism. Otherwise it’s easy to memorize the ‘steps’ to get code checked in without really understanding the underlying version control model…
Do you have any info on the industry adoption of these tools? Teaching students how to use git and the underlying models seems like a valuable skill. In general, I support teaching students using open-source tools and leaving the big expensive tools for when / if they get that kind of firepower in a workplace

Comment on Requirements tools and tasks by Kambria

Kambria — Wed, 28 Mar 2012 21:27:37 +0000

Isn’t the difficult thing for humans, the unknown? Can you look at the evolution of code and see what percentage changes or is add/deleted in coming releases to identify a baseline of expectations. If people understood and could factor in how much the anticipated cost will be, knowing that there will be changes that could be huge.

Comment on Writing Complex Latex Documents with Scrivener 2.1 and MultiMarkDown 3 by Neil

Neil — Thu, 15 Mar 2012 19:17:02 +0000

I passed them through with comment tags. But if you use the Markdown reference syntax, I think it will translate into the correct Latex: see http://daringfireball.net/projects/markdown/syntax#link.

Comment on Writing Complex Latex Documents with Scrivener 2.1 and MultiMarkDown 3 by Weldon Bonnell

Weldon Bonnell — Thu, 15 Mar 2012 19:06:03 +0000

Thanks for writing about your experience.

Still confused about the use of the \ref command.

Is there some syntax for MMD3 to generate the \ref in LaTeX automatically or after all this are you still passing them through with the ⟨!– –⟩ ?

Comment on Writing Complex Latex Documents with Scrivener 2.1 and MultiMarkDown 3 by Tim Brandes

Tim Brandes — Tue, 28 Feb 2012 22:51:31 +0000

Thanks for the article. Thanks jan gerben for the MMD3 explanation.
Based on this, I’ve written an article covering the topic, too:
http://timbrandes.com/blog/2012/02/28/howto-write-your-thesis-in-latex-using-scrivener-2-multimarkdown-3-and-bibdesk/

Comment on Case studies and grounded theory in software engineering by Neil

Neil — Fri, 24 Feb 2012 04:40:12 +0000

Thanks Jorge.

I agree with your critique. I suspect at least one reason is simply information overload. I find it difficult enough to stay on top of my own field; trying to determine how other disciplines might be of use seems overwhelming. Doesn’t it come back to the fact that true inter/trans-disciplinarity is incredibly difficult to organize and succeed at?

Comment on Case studies and grounded theory in software engineering by Jorge Aranda

Jorge Aranda — Fri, 24 Feb 2012 04:02:00 +0000

This is a very nice post, Neil.

Incidentally, one of the problems I have with grounded theory is its disregard for current and valid theory. You’re supposed to start with a blank slate, and let a theory arise from your data. But for most of the questions we’re interested in, there’s already plenty of theories (sociological, psychological, organizational) that, with slight modifications, should be applicable in our field. Why ignore them?

Comment on Elsevier Journals for Software Engineering by Neil

Neil — Tue, 21 Feb 2012 19:43:28 +0000

Thank-you, $spam_commenter. I see your scripting skillz are somewhat lacking.