Disqus - Latest Comments for kjbuckley

Re: SOL[I]D - Interface segregation principle

kjbuckley — Thu, 06 Jul 2017 12:46:12 -0000

Of course, SRP might lead you to ask why a calculator is being asked to handle saving something in the first place!

Re: Blog question | Scott Adams Blog

kjbuckley — Wed, 24 Jun 2015 12:43:12 -0000

Looks fine (18px font size) to me in Chrome 43.0.2357.124 (64-bit) on Mac OS X 10.10.3 (14D136).

Re: What I miss about HackDays · Betta Coding

kjbuckley — Sun, 29 Apr 2012 13:40:03 -0000

FWIW, yesterday's Ipswich Hack day was great. Very casual, no vendors pushing their products (although Tom had managed to get a few to hand over cash and prizes). Just a bunch of people having fun hacking on whatever they fancied, using whichever APIs they thought looked interesting or useful.

Re: Guide to Agile Practices

kjbuckley — Thu, 05 Apr 2012 09:56:16 -0000

"the code modifications corresponding to such entries can be verified to be behaviour-neutral: no new unit tests or functional tests are introduced, for example"

Functional tests I can understand, but surely some refactorings – extract class, say – would be expected to introduce new unit tests?

Re: Redirect permanent

kjbuckley — Fri, 05 Nov 2010 04:42:18 -0000

Does that mean I have to simplify http://www.halfarsedagilema... too now?

Re: Tom Morris: 2010-04-19

kjbuckley — Mon, 19 Apr 2010 06:51:52 -0000

The Prags get it right too.

Re: From TextMate to VIM for Rails Coders

kjbuckley — Sat, 20 Feb 2010 10:26:34 -0000

But in VIM, I have to fall back to the shell command like find . -type f|grep rb$|xargs grep xxx. Do you have any suggestion for it?

Install ack.vim (and make sure you have ack itself on your system), then just type :Ack xxx **/*.rb (or simply :Ack xxx to search the whole project).

Re: Testing Third Party HTTP APIs - Richard Livsey and Bartoz Blimke - Ruby Manor

kjbuckley — Tue, 15 Dec 2009 04:40:28 -0000

A while ago I had a similar problem, and cobbled together FakeTTP. It's a Sinatra app that you can post chunks of Ruby code to, to determine how it should respond to requests. Because it's a standalone app it's aimed at acceptance/integration testing rather than unit testing.

It's pretty basic, and for 99% of cases something like WebMock will be much more suitable, but FakeTTP might be useful if the code that makes the request isn't run directly from the code under test (maybe it's on the back of a queue, or for integration tests it might be another app altogether).

Re: Ruby Manor - Ruby

kjbuckley — Mon, 14 Dec 2009 13:22:57 -0000

I'm probably being an idiot, but when I try to either follow a 'continue reading' link or go directly to eg http://effectif.com/ruby/ma..., I get 301ed back to the main article page.

Re: Has Twitter finally mainstreamed syndication?

kjbuckley — Thu, 30 Jul 2009 11:00:09 -0000

I guess the difference with Twitter is that it's just a website (or it can be if you want it to).

There's no messing around installing feedreaders or setting up Google Reader or whatever -- if you have a Twitter account already you just hit 'follow' and you're away.

Re: Homeworking builds teamwork – and communities

kjbuckley — Sun, 28 Jun 2009 14:51:34 -0000

As you say, it depends on the type of work you're doing. I have a bad habit of forgetting that there are parts of the company (like, say, 99.9% of it or so) that aren't involved in writing software, but for those that are (or who have another team-based creative role), I think co-location is hugely beneficial.

If your job doesn't involve working as part of a team (an actual team doing something together, not just a bunch of people under the same manager), then I expect home working is at least as effective as working in an office (and cheaper), as long as you don't end up isolated.