4 ms·
edit: via Ford's twitter account: https://twitter.com/ftrain/status/609705511651528704 https://twitter.com/ftrain/status/609705511651528704 > the original was
by danso 11y ago
edit: via Ford's twitter account: https://twitter.com/ftrain/status/609705511651528704 https://twitter.com/ftrain/status/609705511651528704
> the original was org-mode but it had to become word
I'm surprised -- almost annoyed, honestly -- to see that the story body is just one giant honking HTML file:
https://github.com/BloombergMedia/whatiscode/blob/master/index.html https://github.com/BloombergMedia/whatiscode/blob/master/ind...
Maybe it started life as separate files that were assembled, and is now only being edited ad hoc after the first public commit [1]? Maybe...but there's no reason why the repo couldn't contain the component pieces (by that, I mean the individual sections of the page), and rebuild them via grunt. In fact, that would make it far easier to do ad-hoc fixes...as fixing HTML by hand is always a pain in the ass...nevermind working within a 3,000+ line file.
I guess I'm just "annoyed" because mitigating the problems of working with large, raw HTML -- and other parts of the textfile-to-publishing process -- is one of those things that is so wonderfully the domain of the programmer. For non-programming journalists, there's just no concept of being able to create a one-command deterministic "build" of a story...so separating the story into multiple files is very impractical...because it means you have to reassemble them by copying-pasting when it's publication time. Also, since the average writer tends to work in Word or some other proprietary document format (I've spent a lot of my web production career translating .docx files to HTML to be pasted into a Wordpress body field)...jumping into different files is not as easy as it is when you're working with Markdown and Sublime Text. Also, much harder to do a article-wide search-and-replace across multiple .docx files.
After learning Markdown (specifically, Kramdown) and static content systems like Jekyll and Middleman...I honestly could not be paid to go into a job where I had to build/publish content a non-automated process...e.g. a rich text editor, whether Word or WordPress. Being able to produce good writing means, well, first being a good writer, of course...but it's far easier to focus on being a good writer when the production process is made frictionless.
[1] https://github.com/BloombergMedia/whatiscode/commit/837bc712583afa43af92c310f1e3a85817c9585e#diff-eacf331f0ffc35d4b482f1d15a887d3b https://github.com/BloombergMedia/whatiscode/commit/837bc712...
- moron4hire 11y ago> as fixing HTML by hand is always a pain in the ass you have a really low bar for what constitutes a pain in the ass.
- danso 11y agoYou have no idea :)