CSS Curiosities of the Past(vale.rocks) |
CSS Curiosities of the Past(vale.rocks) |
Windows has kinda become that now, a low quality user-hostile product forced on people through inertia and monopoly tactics.
There came a point where if you wrote good HTML/CSS that passed the W3C validator, you were golden on pretty much any device. This still rings true today.
Unfortunately that means everyone targets chrome.
Then you end up with shit performance or broken sites in Firefox [1].
You'd think with "AI" they could generate js/css that works in both browsers...
[1] Which is the only browser that can still run proper ublock origin...
Edit: someone pointed below that the compatibility problem's magnitude is much smaller. Seems flagged and dead for unknown reasons.
The magnitude is probably much smaller, but that makes it worse, because people will just forget about FF.
I work for a large corporation. A few years ago, they removed FF as a browser people could use at the company and removed it from the internal app store. Then, if you had a version, they forcibly uninstalled it.
As someone who's a developer, I needed access to be able to test on. Never forget going into the tech support office and saying I need this installed to be able to test my code. Support guy says, "Yeah sure mate, let's get that reinstalled for you." Tech guy downloaded it, started the install, and suddenly got a pop-up warning him it was against company policy to have this installed on your machine. It then auto-removed the app files he just downloaded. We sat there dumbfounded. "I guess we're going to have to escalate this mate." he said with a confused look on his face.
Almost four months later, support finally relented, but put in so many barriers to get installed that only a handful of devs have it on their machines now. I get emails almost daily, "Mate, I need you to test this in FF." or "Is this breaking in FF?"
I guess the reason given was FF lags in updating their certs, making a prime target for malware attacks so they just killed it so they wouldn't have to deal with it.
Just kind of strange that my entire career I was constantly told I needed to test this on any and all browsers available because they all have their nuances and you never know what might break something. To moving to a time where I'm testing on a single browser and calling it good? Man, times they are a changing.
Also I agree the thing about Meerita's post being flagged and dead seems wrong. There is evidently nothing wrong with the post and it should not be.
There was a time when we routinely did browser sniffing, maintained IE-specific code paths, used vendor prefixes everywhere, and sometimes shipped completely different fixes depending on the browser. That largely disappeared as browsers converged on standards.
Chrome-first development can still cause bugs elsewhere, but that's usually a "testing/implementation problem" rather than the same compatibility mess we had back then.
I wonder how many collective years of developers’ lifespans were dedicated just to addressing its quirks.
Here's an archive of their momentous blog post:
https://forums.anandtech.com/threads/ie-update-to-debut-ahea...
They shipped a beta later that year, but the full release arrived in late 2006. That was 4 years since their prior release, an eternity in those days!
Either way, the end result was that everyone started using box-sizing: border-box when it was added, and now even the folks at the W3C regret the old box sizing model being used at all.
Surprised how many of these I didn't remember though. Filters, yes. Syntax hacks to trick IE into using one CSS value while compliant browsers used another, sure. But I don't think I ever came across 'HTML Components' before reading this article.
https://developer.mozilla.org/en-US/docs/Web/CSS/Reference/V...
I used these CSS styles in the late 90s to make internal browser-based apps for our call centre agents that looked & felt more like native Windows desktop apps.
---
Shameless plug:
In neutrino (my webview wrapper), we grab those and inject them as prefixed CSS variables (--neutrino-NameOfSystemColor).
Works on windows/macos/gtk/qt, and except for qt it works live (if the user changes the accent tone, the CSS variable updates automatically).
https://alganet.dev/neutrino/ Screenshot of Mint: https://imgur.com/a/WyjpwsR
These were very handy in simpler times but I definitely see why we don't want them around these days.
I remember having to use Javascript to generate the image path (I cannot quite remember why absolute or root-relative URLs were not an option for us though...)
border-radius: 16px;
https://www.webdesignmuseum.org/web-design-history/sifr-2004
...and that was a grand improvement over having to make a "tic-tac-toe" table where you'd be putting things in the particular surrounding cells to fake these fancy border-types.
Boss enters, enthusiastic hand gestures
dev: Web 2.0? what the hell are you on about?
Boss: evasive mumbling
dev: Oh... css border-radius Sure, I will get right on it.
i was seen as a genius at work for understanding quirks mode and how JScript differed from javascript/ecmascript
Some of those hacks were specifically for IE, and there was a debate on staying quirks vs going standards mode. Each had its own drawbacks and choices.
> Internet Explorer had a number of iffy behaviours and lagged behind other browsers for many visual effects, even regarding basic properties like opacity. Thus, Microsoft brought in bespoke filter functionality.
The DirectX filters weren't used pretty much anywhere until it turned out it could do things that were specced and implemented by others years later.
However, the extend phase wasn't just IE4. They stopped piling on all sorts of non-standard vendor lock-in stuff after IE6. By the time IE7 came out, it was clear the web is here to stay.
Anyway:
cursor: hand !indispensable;
cursor: pointer !important;
Dangerous but appealing idea.
Seems like every web dev (well, those working on server-rendered markup and without AI, at least) would feel a desire for this at some point, but it was doomed at the time.
In some ways, reminds me of Houdini regarding the CSS->JS coupling (the other direction is already normal, sure).
But the intent behind it was probably just to allow maximum flexibility... well
div.title { background-color: expression( (new Date()).getHours()%2 ? "#B8D4FF" : "#F08A00" ); }
I'd love to hear more from people who actually used this.
I still hate Microsoft for 'winning' the browser war and then letting IE stagnate for so many years. So much time and money wasted because of their behaviour. Arseholes.
Interoperability is hard. Plus there are network effects — and so that's why lots of ecosystems tend to see adoption of a single implementation as the de-facto standard. Remember when we had to be aware of character encodings that are not UTF-8?
[1] https://technology.blog.gov.uk/2016/10/21/gaining-understand...
... i haven't been able to download my invoices off the Orange Romania site for months in Firefox. They work perfectly in Chrome.
Somehow I bet the ones that fucked up the functionality got a raise instead of being fired.
Lots of doctors offices and others were still running it. But when the virus slashed our staff, we finally got permission to drop the extra work of supporting IE.
That's also why many browsers also don't allow popups without an URL bar now.
For some time, Internet Explorer even had chromeless popups (no OS decorations, title bar or anything). Absolutely nasty, you could draw anything.
IE6 also stuck around a lot longer than previous versions. It was released in 2001; IE7 didn't show up until 2006, and was an optional update which many users skipped.
I jumped ship when ME came out. KDE had a good browser (Konqueror), the engine of which is now known as WebKit and powers Safari. When I wasn't using Konqueror I ran NN or whatever Mozilla version they had in the day. It worked great but ran slow and was a huge memory hog.