Doing Nothing to Save Time

I noted recently with my ELFquake project that, at least with a slow brain like mine, there’s a definite efficiency trade-off between procrastinating and just Doing Stuff. In the time it took me to get around to implementation, research and plain old thought had led me to much better, quicker ways of doing things. I’ve just had another episode of this in a shorter timescale.

I’m in the process of setting up a local server on an old laptop, mostly running Web-based apps that in the past I’d have simply put on a remote host. I’ve been rubbish at getting the funds in to pay for hosting. But it’s occurred to me that virtually all the apps could be run locally, with the material being uploaded to a static host such as (One exception being WebBeeps – that really needs to be live online, here’s the archived site has the docs).

A lot of the things I want to run are backed by a Fuseki SPARQL server. I thought a good place to start would be a Wiki I put together, FooWiki. I’ve recently discovered the joy of Docker, and there’s an image of Fuseki available. So I’ve also put together an image containing an nginx Web server and the pages & scripts needed by FooWiki. (Dockerfiles here). Today I got that working (some minor bugs in FooWiki, but it’s basically working).

Now I want to get the Wiki pages and upload them to GitHub. So this afternoon I made tweaked versions of the Wiki page rendering, leaving out all the editing bits, so it would be suitable for a totally static site. But there’s a snag – the pages are generated on the fly in the browser from the results of SPARQL queries. I’ve just spent another maybe 3 hours putting to get together a Selenium-based setup to crawl the Wiki, as rendered. And it’s just occurred to me that it would be much, much easier to process the SPARQL results directly – lose the browser entirely.

So I’ve spent 3 out of maybe 4 total hours on solving a coding problem, only for me to think of a different approach that will be much better and probably only take about 1/2hr. I should have watched some art documentaries instead.

Nah, wasn’t wasted time. Making the necessary tweaks to the FooWiki code has made me familiar with it again, and I haven’t played with Selenium before, will probably want to use it before long.



Leave a Reply

Your email address will not be published. Required fields are marked *

Post comment