jaffathecake

38 ms·

451.
by jaffathecake 14y ago
Nah, have a small inline script that records events and replays them to your full script when it loads. Or, if something only works with js, don't show it until the scripts have loaded, but let plain content load & render in the meantim
452.
by jaffathecake 14y ago
Hah, I used to use that trick, until TV's started prefetching colour-key pages. Pah.
453.
by jaffathecake 14y ago
The initial path part of BBC urls ( http://bbc.co.uk/this-bit/etc/etc ) weren't allowed to contain an underscore, because they can't be displayed on Ceefax. Lovely bit of legacy.
454.
by jaffathecake 14y ago
What's your browser & OS? RAM isn't the bottleneck here, it'll be the CPU via a poor JavaScript engine. Of course, it's open source, any optimisation suggestions will be gratefully received. The calculation of sprite boundaries happens
455.
by jaffathecake 14y ago
Unfortunately Safari doesn't support the FileReader APIs required to get image data out of the file. http://caniuse.com/#feat=filereader - as you can see Safari is lagging behind here
456.
by jaffathecake 14y ago
Unfortunately Safari is really lagging behind on the FileReader still. But yes, it should still show what the site is about, noted.
457.
by jaffathecake 14y ago
Sprites often perform better than data uris. With data uris, you're putting the image data in your css file. This creates two performance differences to images. Firstly, the user will download the image data even if it isn't used on the pag
458.
Using Sass 3.2 to create mobile-first CSS that doesn't leave IE8 behind(jakearchibald.github.com)
2 points by jaffathecake 14y ago|0 comments
459.
by jaffathecake 14y ago
Confirmed! Interesting. Safari on OSX and iOS makes the request regardless. Firefox didn't request the far-future file, but also didn't request one of the must-revalidate files. Not sure what it's doing. Good spot.
460.
by jaffathecake 14y ago
I don't think this is the case, not in Chrome anyway. See http://www.youtube.com/watch?v=q7QL0NAwES8 - All files are "must-revalidate" except the JS file which is far-future cached. When the manifest is changed an update is triggered, all
461.
by jaffathecake 14y ago
Bah. Nice idea though. So far we're using a double-sized sprite image whether you're on a high dpi device or not. Thankfully everything that supports appcache supports background-size, and the 'waste' of serving high-res sprites to phones t
462.
by jaffathecake 14y ago
Yeah, redirects to manifest entries makes the cache fail, but a 302 on a page the user visits (which is covered by a FALLBACK rule) is fine as long as it's to the same domain. As for GOTCHA #9, I thought that was fixed in iOS4. But yeah als
463.
by jaffathecake 14y ago
Not strictly true, most sites rely on the client's cross-domain policy for security. Without it, this site would be able to read your Gmail & facebook pages. If a useragent came out that had a major x-domain hole, I'd expect data-sensit
464.
by jaffathecake 15y ago
My initial worry is the subjectiveness of "offence". As a speaker I like to think I'm putting on a show, I try to make my talks entertaining by throwing in a few jokes. If jokes were discouraged at a conference, I'd neither want to speak or
465.
by jaffathecake 15y ago
Do you have any evidence against progressive enhancement or are you just trolling? We built http://m.lanyrd.com using progressive enhancement, it didn't hold us back, quite the opposite, we were able to support devices old and new. In fac
466.
by jaffathecake 15y ago
We didn't have to do anything special to get offline stuff working in iOS, sounds like we're lucky we didn't hit the problem you're having. Here's a simple demo that works in iOS http://appcache-demo.s3-website-us-east-1.amazonaws.com/xhr/
467.
by jaffathecake 15y ago
You're not far off... To cache everything you're tracking/attending visit http://m.lanyrd.com/calendar/yours/ - the root url redirects here if you're logged in. The data for each conference is versioned, you can see the version number as
468.
by jaffathecake 15y ago
We were hoping to get away with a single double density sprite, but alas, as you point out, not all devices support background-size. Not just old Android, but also old Blackberry (pre webkit), and Opera Mini. We feature detect background-si
469.
by jaffathecake 15y ago
Firefox mobile didn't give us any problems at all, it just did as it was told. Lovely browser. ahem Although we have fallen into the browser sniffing trap once. Opera's ApplicationCache doesn't support 'FALLBACK' which we heavily rely on.
470.
by jaffathecake 15y ago
At the moment, the event you're tracking or attending are available offline. So yeah, you need to log in to trigger offline caching. When you view the 'Your events' page you'll see the events being cached if your device supports it (a reaso
471.
Show HN: A little tool to generate CSS for bits of a spritesheet(spritecow.com)
6 points by jaffathecake 15y ago|0 comments
472.
by jaffathecake 17y ago
It's common for people to see something they don't think directly benefits them and decide it's a waste of the licence fee. Perhaps you're not tied down to older browsers, so Glow has no benefit compared to other libraries, for you personal
473.
by jaffathecake 17y ago
Glow does support Firefox 3.5, Chrome 2, Opera 10 and Safari 4. We're certainly not going to ignore new versions of browsers until they appear on that list. At some points, we've been ahead of jQuery when it comes to new browsers. Basically