Category: Uncategorized

Embeddable image links for asciicasts

A couple years ago I stumbled across ascii.io (which is no asciinema.org), a phenomenal tool for recording terminal sessions. Command-line mechanics is one of the biggest barriers of entry for scientists learning informatics, and I’ve found this tool to be much more useful than screenshots or copy/paste text for communicating terminal behavior to students and colleagues.

The ability to embed asciicasts has been around for a while, but because this option relies on JavaScript it has its limits. For example, any script tags I paste into a blog post on WordPress will be stripped out, and the same is true for Markdown files hosted on Github. So up until now I’ve simply been using hyperlinks to refer to asciicasts on pages where I cannot embed JavaScript.

A new option became available recently: you can now embed image links for your asciicasts. Because these are simply images (probably dynamically generated and cached on the asciinema server), they can be placed pretty much anywhere. And although it’s still not as nice as a truly embedded player, the image link looks and behaves almost like a player.

Now that this feature is available, I have gone and added an asciicast link to one of my Github repos. I’m quite pleased with this feature and look forward to putting it to good use in various other ways!

Validating genome annotations revisited: gt speck

I have written previously (here and here) about my frustrations with the lack of validation tools for GFF3 data. I had even begun prototyping my own system for writing schemas that could be used to validate GFF3 files (much like XML schemas used in validating XML files). I hadn’t gotten very far, though, before the talented folks behind GenomeTools released the gt speck tool. The speck tool uses a domain-specific language where individuals, users, groups, or communities of practice can explicitly enforce a set of rules regarding how annotations should be formatted, how features should be related, and which attributes are required. Most scientists and software have requirements beyond what is detailed in the GFF3 specification, but in most cases these are implicit and inconsistently enforced. I look forward to integrating the speck tool into my annotation quality control workflow, and I have great faith that the genomics community has a lot of benefit to gain from wide adoption of tools like this.