This content originally appeared on daverupert.com and was authored by daverupert.com
Matt Haughey wrote a blueprint for his “Dream CMS” and we had him on ShopTalk to talk about it. That got me thinking about what features I’d want in my dream CMS. It’s fun to think of what a modern CMS might have like inline editing, asset serving, monetization/membership functionality, and more imaginative comment moderation.
Most of my dream CMS features center around improving my writing and content surfacing. All these ideas seem like a pretty good fit for LLMs or AI, even though I don’t drink from that particular well often.
1. On-the-fly fact-checking
As I think more about verifying outputs, I’d love a feedback tool that did a quick fact check before I hit publish. Did I not cite a source? Did I make a false assertion that ruins the credibility of the whole post? Did I not consider some perspective I should have? Am I just bitching and moaning? CMS, help me, please! I’m making an idiot out of myself.
2. Auto-tagging
I write a lot of posts but I’m shit at tagging them. I wish I could tell my CMS “Hey, here’s a general list of topics I write about, please recommend tags before I hit publish.” Then I think it’d be cool if the little robot butler could crawl old posts suggest tags for those or find themes emanating from my subconscious that I hadn’t picked up on.
3. Topic bundling
À la auto-tagging, it would be neat if a CMS observed my content at a high-level and suggested opportunities to bundle posts of a similar theme into an online “zine” to offer new topic-based entry points for new readers. My lists of reverse chronological posts are thematically scattershot and I think some readers might find this more useful.
This idea gets into listicle territory pretty quick, but I often pull on thought threads over the course of months or years but that is entirely unsurfaced on my blog. I’d also love it if the machine could bundle links I share on my Mastodon feed to pull in other perspectives as well.
4. Idea validation
I’d love to know what content is outperforming other content. I’m sure I could piece this together myself with a half-dozen spreadsheets from Google Analytics, but I’d love to log into my CMS and know “Your CSS posts tend to have 10x engagement over your longform thoughtvomit” or “Your shitposts are driving traffic but are negatively impacting the quality of your site”. I don’t play the engagement game much but I’d love some data-driven validation nudging me in a positive direction when I’m discouraged, perhaps it could even suggest a handful of topics.
There’s a gross content farm machine version of this idea that identifies popular topics, suggests titles, an LLM writes posts based on those titles, and Dall-E plops an unsplashy image at the top… but that gives me the ick. At that point, you’re not blogging your thoughts or offering anyone any human value, you’re just clicking ✨ Sparkle Buttons.
For what it’s worth, baking these features into a CMS has little material benefit for me. This site is statically generated from markdown so it’s practically the opposite of a full-featured CMS. Or… perhaps it’s the perfect situation to start exploring some of these ideas.
This content originally appeared on daverupert.com and was authored by daverupert.com
daverupert.com | Sciencx (2024-04-25T20:03:00+00:00) Ideas for my dream CMS. Retrieved from https://www.scien.cx/2024/04/25/ideas-for-my-dream-cms/
Please log in to upload a file.
There are no updates yet.
Click the Upload button above to add an update.