For bloggers
The following info is relevent only to people who maintain a blog.
How to claim a webspace
The following content assumes you have a basic understanding of HTML and web technology in general.
Webspace is a directory in a URL: For example, “https://google.com/path/“ is a webspace, which covers “https://google.com/path/a.html” and “https://google.com/path/b.html,” but not “https://google.com/path/c/d.html”, unless “https://google.com/path/c/“ is not yet claimed. We assume that web pages within the same webspace are related, often from the same author.
If you own a blog, you can utilize Roastidio.us as a commenting platform. Add a hyperlink to: https://roastidio.us/roast in each of your blog post, then you are all set. The visitor that clicks the link will be directed to the correct topic page. For the curious minds, Roastidio.us looks at the referer header of the incoming HTTP request to figure out the topic. One thing to keep in mind is you have to make sure the referer header is intact. In Firefox 87+ the default of referrer policy is very strict, it will strip off path and query string when crossing origins. So you would need to add the proper referrerpolicy attribute in the link:
<a href="https://roastidio.us/roast" referrerpolicy="no-referrer-when-downgrade">
Roast me at Roastidio.us!
</a>You can claim the webspaces that you owned. After doing that, all roasts received on those pages will be emailed to you and be censored by you so that you will be in the loop, and the roasts are free of SPAMs. To claim a webspace, you need to demonstrate to Roastidio.us that you control the content by performing a little trick, which is a simple challenge and response:
- You claim a webspace
- Roastidio.us gives you a unique and time-limited token
- You put the token in the index page of your webspace as a meta tag
- Roastidio.us checks the existence of the token by visiting the index page
The index page is usually called “index.html” but could be something else, especially for dynamically generated pages. If your blog or site has hierarchical directories that hold toastable page, you need to claim the top most webspace only. It is strongly recommended that you also have a RSS feed link in the form of rel=alternate links in your index page, however it is not strictly necessary.
Anyone can put forth a claim; however, you and you alone can perform the trick in your webspace. Once Roastidio.us verified that the tag is there, is the original and unique tag it gave you, and does match the claim you originally made, All new topics from there would be yours. The checking is done only once; you can remove the tag afterward.
Some restrictive blog platforms do not allow adding custom a meta tag to the homepage though. There is a hacky work-around: you can put the very long string in the description meta of your blog asthe following:
roastidious:claim=eyJhbGciOiJIUzUxMiIsInR5cCI...Remember to use the full string, which is very long. Usually the place is called description or “tagline”, or something along that line in your blog setting. Most likely it will show up in your blog and make it ugly; however you can remove it as soon as the check is over.
If you are ready, go to your profile, click on your avatar then click on the CLAIM MY WEBSPACE button.
Metadata that we use
Roastidio.us make use of the metadata you published on your blog. In particular, we make heavy use of 2 kinds of metadata:
We highly recommend that bloggers to expose useful and matching informations in the above mentioned venues:
- Your blog’s home page should advertise a feed via alternate link
- Your feed should link back to the home page
- Your blog items’ URLs should have the URL of your blog homepage as a prefix
- the metadata in your feed and your webpage meta should match each other
The following table summarize the metadata that we check as of now, using the most commonly used standards of RSS2 and Opengrah as examples:
| metadata | RSS2 | Opengraph |
|---|---|---|
| site link | channel > link | N/A |
| site name | channel > title | og:site_name |
| item title | title | og:title |
| item url | link | og:url |
| text description | description | og:description |
| html content | content|encoded | in the html body |
| image thumbnail | enclosure | og:image |
| tags | category | article:tag |
If your metadata is incomplete, or the version from your feed does not match the version from html meta, we can sometimes guess and fall back gracefully, but why take the chance?
We also highly recommend bloggers to include the full html text in the feed. Text is cheap in term of bandwidth; there is no reason to be frugal now.