Shell zip command that makes an de-subversioned fileset

I wanted to give somebody a zipped chunk of a local copy of an SVN repository without handing them subversion info in the form of all the .svn directories and files that were lurking within.

After a little lazy googling for a preboiled answer to this admittedly trivial task I read the zip man page and came up with

zip -r directory_zipfile directory -x \*.svn/*

• The -r option zips the directory and its contents. Note: you don't need the .zip suffix on directory_zipfile.
  
• The -x supplies a pattern to match filenames that should be excluded from the zip. All the SVN info lives in hidden directories that match the pattern I used.

As it happens this is also a quick way to make a version-free copy of a versioned file tree: it's pretty simple to run the command and you can move the copy about the place easily before unzipping it.

You could say that people using SVN are clever enough to know this and indeed to write their own fiendishly complicated scripts to accomplish the same task. But this would not be particularly truthful or helpful.

UK E-voting isn’t ready at all

Open Rights Group’s report on the May 2007 e-voting trials in Scotland and England is now published. I started to find the report hair-raising when I read that ‘problems included: use of desktop productivity software, machines in public areas with open ports, informal transfer of files using personal devices, and single-factor authentication’. In other words, bumbling amateurishness.

It’s pretty clear from the report that the people who run elections – from the the polling stations up to the Government’s Department of Constitutional Affairs – are generally not especially well trained in technological matters and are vulnerable to being blinded by science. They’re proud of their reputation for impartiality and feel threatened when people ask questions that they don’t understand about laptops and servers, internet connections and the like. Hence they have harmed their reputations by providing demonstrably incorrect answers to ORG’s observers rather than admitting that they cannot answer for themselves. Their lack of candour is regrettable because it undermines our confidence in them and therefore our confidence in the democratic process of the UK. I hope that these people are better trained in future to work with the new systems (which I am sure will prevail).

There were also shortcomings in the design of the ballot papers and vote counting interfaces – the items that voters and adjudicators were interacting with. Forms and interface design are both things in which, as an information designer, I have a professional interest. If any of the companies contracted to supply the e-voting systems are feeling a little shamefaced about what they have come up with so far, then why not consider asking the advice of one of my colleagues in information design?

Also see the Ministry of Justice’s announcement of the trials: http://www.justice.gov.uk/whatwedo/may2007electoralmodernisation.htm

Looking back on the sixth Friends of St Bride Library conference

A really enjoyable affair, even for muggins who was pretending to be a sort of technician type looking after the speakers – a making sure they had their presentations in the right way up kind of job.

There's been some very positive feedback to the Library about the conference. That makes it all even more worthwhile. Also see enthusiasm at We Made This.

Talks I enjoyed the most included those by Tim Fendley, Morag Myerscough, Paul Rennie and Max Gadney. Programme

**

7 June: and on PrintWeek