Chris’ Corner: Write Once, Read Anywhere

Ryan Mulligan has an interesting exploration of a somewhat niche CSS Grid behavior in CSS Grid Gap Behavior with Hidden Elements. When creating a grid, I find it’s common to create pretty specific column behavior, and let rows auto-generate, but of course you can be specific about row sizes as well. If you do that, say, grid-template-rows: auto 1fr auto; or the like, and you have a gap that puts spaces between grid items, you’re going to have vertical space between those rows whether there are items in them or not. I think this is great-to-know information because WHERE THE HELL IS THAT SPACE COMING FROM?! is everyone’s least favorite kind of CSS debugging.

Mitosis looks like a cool project to me.

Write components once, run everywhere. Compiles to Vue, React, Solid, Angular, Svelte, and more.

If I was creating componentry that needed to run in multiple frameworks (I imagine design systems are the #1 use case here), I’d give it a shot in a heartbeat.

I think if you asked a bunch of people (I stopped writing this for a sec to ask a bunch of people) if they would use a real non-Safari browser on iOS if they would, they would say yes. As a reminder, you can download browsers like Chrome and Firefox on iOS, but they are just Safari under the hood. Mastodon is probably a pretty weird audience to poll here, but we’re somewhere around half of people saying they would.

I imagine instantly losing half the users of a software product you make is at least a part of the reason that Apple doesn’t want to allow it. Then again, I dunno. Safari is free and iPhones are not. You’d think they’d make choices more centered around making their phones desirable. I’ve often wondered if Apple chose to change this policy, or were forced to, if Google would have a version of Chrome immediately ready to go. Some news reports that… maybe? I sort of doubt it, especially in the case of Mozilla who I don’t think can afford to do entirely speculative intense development work.

I find it fascinating how the world has all this strong, solid, database technology. You wanna store data in a smart way, use tried-and-true tech like MySQL. That’s what WordPress uses and it’s half the web, right? Right! Well, unless you want to search that data. MySQL doesn’t have very good full text search. Slow and not good at ranking. By extension, WordPress’ built-in search is pretty bad, which helps paid offerings like Jetpack fill the void.

For big companies with their own data storage plans, since it’s known that you can’t use your main MySQL database for good search, often reach for something else. Another database in which to replicate and sync data into that is actually good at searching. That’s basically the premise that makes tools like Elasticsearch a thing. But hot dang if that doesn’t explode the technical debt you’re responsible for.

Chalk all that up as a reason that Postgres is nice. MySQL, but better (debatable, I know, but let’s say better), and Full Text Search is Awesome, says Montana Low.

Do you think you could have guessed top 10 most-visited websites worldwide? Wikipedia has a list. As of last month:

1. Google Search
2. YouTube
3. Facebook
4. Twitter
5. Instagram
6. Baidu
7. Wikipedia
8. Yandex
9. Yahoo
10. Whatsapp

I would have done OK. I would have biffed Baidu and Yandex, as while I’m aware of them, they aren’t on my brain much as a U.S.-based person. Yahoo blows my mind, and I didn’t even know Whatsapp worked on the web. But I certainly would have put Wikipedia higher! Higher than Twitter, anyway.

I was just thinking about this while reading Alex Hollender’s Design notes on the 2023 Wikipedia redesign. Always worth taking note when we get to read a little behind-the-scenes on a redesign of a site in the top 10. I liked the insight into the goals:

Olga, the product manager, and I started by articulating a few goals (which are interconnected and overlapping, as most sets of goals are):

• Make the website familiar & welcoming to anyone who visits (thinking especially about younger people, in other parts of the world, who have not yet discovered Wikipedia)
• Improve the experience of reading, navigating long articles, and knowledge discovery
• Better accommodate divergent needs (reading vs. editing)
• Develop a more flexible interface, with an eye towards future features

Olga also picked key metrics to monitor: pageviews, edit rates, account creation, and session length.

Me, I give it a thumbs up just for fixing that line-length issue. It used to be whack on desktop.

New look is classy I think. Not so different it sets off any weird alams but addresses many issues.

Just in case they are reading, if you could fix how tables render on mobile so I don’t have to side swipe reading TV episode entires, that’d be great thanks.

There are tons of examples of responsive tables on CodePen.