It's still easier to make a static site. I think the authoring tools to generate HTML just currently suck, or if they don't suck they have some process that needs to run on the server to serve the site.
It's still easier to make a static site. I think the authoring tools to generate HTML just currently suck, or if they don't suck they have some process that needs to run on the server to serve the site.
My job isn't to ridicule them because it could be better, but make their solution better and ensure my solution for them works as well as theirs, if not better, without hampering the success they already had with what they were doing already.
A lot of developers don't want to admit this—in my experience at least—but tons of these ad-hoc web solutions actually work better than a ton of strategies experienced web developers would implement on their own.
So much of what counts is what the business offers and how they relate and interact with customers. Sometimes all that takes is a word doc exported to HTML. We can use our skills to improve it, but the real magic is in the humans running the business. I love it.
I find something fun is that finding ways to improve these solutions can actually be genuinely challenging. Sure, you can make a better website, you can deploy it with sophisticated infrastructure, etc. but at the end of the day, do their customers prefer it? Does it improve their business? Sometimes that part isn't trivial at all.
It's actually even more charming than any solution we ended up providing because this particular restaurant (sadly closed now) was run by a 60-70ish year old man, who's cropped portrait photo was their logo, `float:left` in the header of their index.html exported from Word. I don't think you can buy authenticity like that.
I of course thought that was hilarious because I was 19, an idiot, with my first tech job thinking I'm such a professional cranking out CodeIgniter-based contact forms and static About Us pages.
Looking back, I'm pretty sure we did them dirty. Whatever solution we sold them on (IIRC, they got a wordpress site with a custom theme) was probably less useful. Which sucks.
It seems weird to say it since so many people rely on it, but wordpress is overkill for so many things. It frustrates so many clients to no end, requires ongoing maintenance that's quite expensive in some cases, and well, it requires an active server handling server-rendering and form submissions and such. The vast majority of clients simply don't need it. They don't even need themes.
This party I'm helping has something like 7 wordpress sites strewn about, all on hosts that cost way too much, some not even updated or maintained at all, sitting on servers that peak at like 40% utilization and otherwise hum along at near-zero utilization beyond what the wordpress isntance requires.
This is a lot of the internet. My experience with these folks has inspired me to build something that would meet their needs, but it's one of these things where... I don't know, if I build it, I doubt anyone would come. But yeah. Most people's needs for the internet are remarkably simple. Even Wix or similar are way, way too much. Yet those basic blogging engines totally miss the mark too. Most of these business users aren't interesting in blogging. They just want a simple home base where they can dump various types of information and let it hang out forever.
I can think of a few products which aim to fix this, but they aren't quite simple enough for the types of people we're describing. The people who know they need stuff online, but don't want to know much about it and don't want to learn much either.
I see people bifurcated into "people good at time estimates" and "people who think time estimates are always wrong." The former have been keeping track, improved continuously, and have become good. The latter have no clue.
Setting up a website from scratch is not hard, but it is time consuming, especially when you throw in maintenance.
I still try to "engineer" to the best of my ability—separating raw input from derived data from configuration, data normalization, etc. With Lambda functions in Excel now, I kinda just pretend I'm writing Lisp in an FRP editor / runtime environment. The ETL tools with PowerQuery are quite good for the scale that these restaurants operate at.
Hard for me to turn off my brain in my full-time job when I am tasked with poorly recreating a feature that Excel nailed years ago.
Gesundheit? It’s still in Office 365 (subscription) and Office 2024 Pro (standalone).
I have been looking for a good alternative that can create more correct HTML and which gives you more options than Words HTML exporter for a couple years now.
I have tons of little single HTML file sites I make from this [Vue template](https://vue-template.spaghet.me/). I also have sites just published as public Notion docs, inline photo libraries from iCloud, etc. Its amazing how easy and available it is to just connect stuff, and how often I want to over complicate things and build something from scratch when I can just link stuff together.
On an aside I also love things like [mmm.page](https://mmm.page/) for quickly stitching together little micro sites or single use sites if I don't have time to do something else. Its fun exploring all these tools
But some of the pages have "documentation" which is basically a bunch of tables with dates and text, which they handle themselves. And the solution we landed on was very similar - they just have a Word doc with this table, they give it to us, we export it to html/css, and dump it into the appropriate place.
It's not elegant and not a scalable solution, but for the use cases where it's relevant, it's the easiest way, hands down.
In a parallel universe, web browsers are called webitors, and they can edit websites as well as view them. People can suggest changes. People can publish annotations. Web hosting is like email--pick (or build) any service you like and pick (or build) any client you like. The protocols will sort it out.
Though making the web this big publicly editable pile of documents would create a lot of spam. You already have to filter a lot if you have a form on a website, imagine if instead of filtering structured data, you had to filter diffs.
I think maybe a good half measure is just bringing web authoring tools BACK to web browsers! Netscape Navigator had one if I remember correctly. This would just allow you to write HTML files to disk though.
Bonus point for some standard protocol where you login with HTTP Basic Auth, and POST/PUT/PATCH/DELETE HTML directly to a page. You'd need some special server that understands that, but ideally it would be open for anyone to implement. You point your browser to https://yoursite.com/my/page.html and assuming you've logged in with HTTP Basic Auth, your browser suggests creating the page, rather than rendering the responded error page.
Edit: Protocol is the wrong word, someone correct me what this is really called. HTTP based... editing? A standard API? It's now something I want to make so I'll need to figure that out haha.
Edit 2: CamperBob2 in a sibling comment mentioned wikis. Just realized I'm describing a wiki lol
There was an effort many years ago to provide browser extensions that allowed users to mark up, edit, and annotate existing web pages in a way that would be visible to the community of users. It never went anywhere, unfortunately, probably because it would have gored way too many sacred oxen. (And/or it would have been co-opted by spammers, turning it into a social networking/reputation management project.)
And of course any such project that became popular would either ruin the creators as they try to keep up with moderation, or devolve into an unusable mess if they don't moderate.
https://en.wikipedia.org/wiki/User_agent
What you describe came and went, subsumed under the homogenizing effect of silos like Facebook. Mashups were the big new thing. There have been various attempts to bring it back, but the inertia of silos makes it hard to find anyone interested.
Nowadays it's more important than ever to curate content well.
The job of the browser is to _render_. That's agnostic on whether you're creating content or just reading it.
Doesn't produce particularly modern HTML, but if you want to author HTML like it's 1999, it's out there.