Profile cover photo
Profile photo
Tom Wiltzius
3,189 followers -
finding tongues in trees, books in the running brooks, sermons in stones, and good in every thing
finding tongues in trees, books in the running brooks, sermons in stones, and good in every thing

3,189 followers
About
Posts

Post has attachment
Extremely proud of +Elaine Chang and +Daniel MacDougall for taking on SuperPACs -- Social Teeth is launching today. Think Kickstarter for political media. Go vote on political ads that you care about, which could be the first of the SocialTeeth crop!

https://www.socialteeth.org/
Add a comment...

Post has shared content
Build your own SuperPAC. +Elaine Chang & friends kicking ass.
Add a comment...

Post has shared content
Talked today with +Nat Duca at Google I/O about building smooth web apps. Video coming soon, for now slides & some other resources on fluid Chrome apps up on http://www.jankfree.com
Want to speed up your website? Check out a few techniques for isolating jank at http://www.jankfree.com/. Great Job +Tom Wiltzius on presenting this information at #IO12  
Add a comment...

Post has attachment
Of interest for web designers: fixed position elements will create their own stacking contexts in Chrome 22 and later, which subtly changes some layout. You don’t want a broken site, so check your layouts with a new flag in Chrome.

Long version:

The Chrome team will soon be making a change to the way fixed position elements are handled to match a recent spec change -- these elements will create their own stacking contexts which means that stacking order (think z-index) may be affected in certain layouts.

Breakage should be rare, but if you're doing things like setting z-indexes of children of a fixed position element but have these z-indexes interleaved with content outside of the fixed element, you should worry (see below on how to easily check). You should be able to achieve almost all of the same effects with minimal CSS changes.

To check whether your layouts continue to work as you expect them to in Chrome, please test pages with the "Fixed position elements create stacking contexts" flag turned on. For the uninitiated, flags are easy to access via the special about:flags URL.

Why are we doing this? Mobile WebKit browsers already behave this way because it allows for significant scrolling speed optimizations. So if your site looks correct in e.g. iOS Safari, you needn't worry (provided you're serving identical layout on mobile, which you probably aren't). The upshot is that we're going to be able to make pages with fixed position elements scroll much faster, which is good for your web app's users, and we're going to unify layout behavior on mobile and desktop Chrome.

If you’re curious, the relevant spec is here: http://www.w3.org/TR/CSS21/zindex.html but it's, ah, a little dense.

Cheers!
Add a comment...

Post has shared content
See also the original Chrome Blog post:

http://chrome.blogspot.com/2012/02/snazzier-graphics-for-more-users.html#links

Snazz hands

Seriously though, SwiftShader is great for filling gaps in the web platform like the uneven support of WebGL. It's an imperfect solution (software rasterizers aren't nearly as performance as real GPUs), but a big step in the right direction while we wait for everyone's GPU drivers to improve.

Hardware accelerated canvas represents a big shift in the performance profile of HTML5 Canvas2D -- many operations are waaaay faster when they're done on the GPU, but there are some operations that CPUs are better at. If you see any performance slowdowns with Canvas2D, we want to know about it -- file a bug at http://crbug.com/new and tag it with Feature-GPU-Canvas2D.
The latest Chrome Beta release -- rolling out today -- introduces a few improvements to help ensure that all users can have smooth and seamless experiences on the web, especially while using the more modern (and graphics-intensive) applications out there.

If you’d like to dig a bit more into the nuts, bolts, and GPU-accelerated-details of this release, you can find out more on the Chromium Blog: http://goo.gl/NS2x6
Add a comment...

Post has attachment
Anyone have recommendations for a super-solid, powerful wireless-N router? Spending all of Sunday fighting with my current one, I'm fed up enough to look for a new one. I'm fine paying plenty (up to maybe $200) for something that works; these cheap boxes keep failing on me.


My latest problems are with my Buffalo WZR-HP-G300NH, which is otherwise excellent and powered by DD-WRT, but despite trying many different firmwares drops its wifi interface with increasing regularly. I think the problem traces to the shitty Atheros drivers (why oh why do hardware makers fight OSS so hard? We'd fix it for them. See http://en.wikipedia.org/wiki/Qualcomm_Atheros#History).

But I've also had multiple Linksys WRT-54G's burn out on me, and some shitty D-Link I had implemented port forwarding incorrectly in their closed-source firmware which cost me days of frustration before realizing that the problem wasn't configuration but a firmware bug.

In short, I'm ready to move on to a more robust situation. Is there anything in the space between $50 pieces of junk and $600 enterprise boxes?
Add a comment...

Post has attachment
Dinner at casa folsom. May have wrecked the cast iron pan in the process, all that cheese is stuck like jam on polar fleece.
Photo
Add a comment...

Post has attachment
Add a comment...

Post has attachment
Add a comment...

Post has attachment
The bay looks nice this morning.
Photo
Add a comment...
Wait while more posts are being loaded