Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

This will sound crazy, but I recently deployed a content management workflow for a rails app that mounts a WebDAV drive inside of a staging rails application that's running http://sitepress.cc/ (think Middleman in Rails, without all the dependencies)

If the marketing team wants to edit something, they mount the content drive, make changes using their text editor of choice, and are able to preview when they hit Save.

After they make a bunch of changes and add whatever media an engineer copies the changes off of the drive into a commit and deploys to production.

So far it's been working pretty well. The marketing and CS team are all able to edit markdown copy and run it easier through their content editing processes since it's just files. They don't need to learn or care about git. Designers can throw images in the volume without thinking too much about it.

Engineering is happy because they control their git history and commits. When it's time to optimize PNGs, compress assets, etc. the engineer can deal with and deploy to product with confidence.

We also run rspecs against every single page of content to enforce SEO, branding, etc.

Happy to answer questions or write something more detailed about it if others are interested.



I setup winscp to automatically sync a directory from a coworkers laptop to a webserver so he can edit his html documents in Microsoft Frontpage locally and have it sync with the remote server.


I've done something similar by grabbing a google doc by url, scraping its html markup, and generating a site from it.

Some people have even done a google doc markup like language.


I highly recommend Sheets too for tabular data. The series of /vs pages at (like https://www.polleverywhere.com/vs/turning-technologies) are all generated from one Google Spreadsheet. You can see the spreadsheet data itself if you click "Download Spreadsheet" and see the tables if you click on any of the subpages.

I implemented the "scraping" via a Rake task. When marketing updates a point of comparison, an engineer runs the task which simply downloads a CSV file that is plugged into these pages and a nicely formatted XLXS file that's made available for download for people who want to see all the data in one place.

There's "big data". I call this "tiny data".


As seen on HN 2 days ago: Show HN: HexoPress – a blog that syncs with your Google Docs https://news.ycombinator.com/item?id=13857137


I used to work at a project where I suggested doing EXACTLY this, because the customers were pretty old-school Microsoft Office users and we had to store lots of documents and organize them.

The idea never caught on because the people didn't find it as a good solution. Happy to see that somebody, somewhere out there has built exactly this and is happy with the outcome ;)

Cheered me up a lot, thanks for sharing!


Your welcome! The engineering team got some chuckles out of me deploying WebDAV but it works! I originally wanted to use Dropbox to sync a shared folder, but their Linux client doesn't deploy well in a Docker container. The first attempt at a WebDAV server was nginx, but it had bugs in its WebDAV implementation, so we reached out for good 'ol Apache.

The whole thing feels so retro, but it works pretty well and makes it just that much more fun of a solution.


> I originally wanted to use Dropbox to sync a shared folder

This makes it even more funny, because I thought of that too! Wow, feels like reading an autobiography ;)


Wow, didn't expect this much interest! If you find https://sitepress.cc (or are an avid Middleman/Jeckyll fan) and these workflows interesting we're hiring a marketing engineer to make this pipeline even better. https://www.polleverywhere.com/jobs#marketing-engineer or reach out to me Brad at Poll Everywhere dot Com


Looks like a typo in the Marketing Engineer vacancy copy:

> You understand how HTTP cache headers, HTTP/2, and asset fingerprinting can increase speed page load times

Overall meaning is obvious, but “increase speed page load times”—probably meant either increase loading speed or decrease load times.

Are you open to fully remote, by the way? (As in, someone working with you while on the move traveling in SE Asia)


Thanks for pointing out the typo. I'll fix that.

Yep! We are open to remotes that distant. In fact, we have an engineer working with us from Malaysia.


That's pretty cool. What are the marketing folks using as a WebDAV client?


They way I understood him, they simply mount the WebDAV share using their regular file manager, so Explorer on Windows or Finder on Mac.


On macOS people mount the WebDAV volume as a network drive in Finder. To them it seems like just another folder.

As others point out WebDAV is also integrated into Windows Explorer.

The beauty of it is that people don't need to install additional software and almost everybody knows how to use a file system.


I do it often for my website. The unix command is sshfs remotehost:remote_dir local_dir.

How do you do it with windows 10 ?


> The unix command is sshfs remotehost:remote_dir local_dir.

That's not WebDAV, but using SSH/SFTP. Requires extra software on windows (either an SFTP client, or if you want the filesystem integration specifically there is a version of sshfs for Windows)

Basic WebDAV support is integrated in Windows Explorer, and has been for ages, but there are alternative clients (both paid and free) as well.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: