Twitter | Search | |
Jeff Posnick
🏢 Google's Web DevRel Team (), 🏠 Brooklyn, NY, 💘
4,465
Tweets
790
Following
2,704
Followers
Tweets
Jeff Posnick retweeted
Smashing Magazine Aug 14
✏️ An in-depth analysis of the new Gutenberg Editor and its impact on the WordPress web development. (by )
Reply Retweet Like
Jeff Posnick retweeted
Ben Kelly Aug 14
PSA: Chrome 70 will begin rejecting cache.addAll() when requests would overwrite each other within the same call. This aligns Chome's behavior with the spec and other browsers.
Reply Retweet Like
Jeff Posnick Aug 10
Not specifically, but have you tried just using the default scaffolding you get when adding the PWA template to your vue-cli 3.0 project? I think the idea is that it should "just work".
Reply Retweet Like
Jeff Posnick Aug 10
Congrats on the launch, ! vue-cli 3.0's PWA template features integration, with easy customization:
Reply Retweet Like
Jeff Posnick retweeted
Evan You Aug 10
I just published “Vue CLI 3.0 is here!”
Reply Retweet Like
Jeff Posnick retweeted
Chrome Developers Aug 10
Hey Hey👋 Okay let's stat this week's wrap up with performance article 🚗💨 Read on "Reduce JavaScript Payloads with Code Splitting" by and
Reply Retweet Like
Jeff Posnick Aug 10
Replying to @ladyleet
Why don't functions buy things on sale? Because they pass by value.
Reply Retweet Like
Jeff Posnick Aug 9
Related to alternative precache manifests is required vs. optional payloads, tracked at
Reply Retweet Like
Jeff Posnick Aug 9
Hmm, this would end up being an interesting usage pattern to document for —alternative precache manifests for browsers that support ESM. CC:
Reply Retweet Like
Jeff Posnick Aug 8
Replying to @getify @____lighthouse
I can't speak for the team, but my guess is that they'd appreciate an issue filed at with an example to reproduce what you're seeing.
Reply Retweet Like
Jeff Posnick Aug 7
Replying to @getify
If you're concerned about the metrics detailed in , then there's an important difference between loading a resource because it's used in the current page, vs. loading a resource to populate the SW's cache in order to optimize future navigations.
Reply Retweet Like
Jeff Posnick Aug 7
Replying to @getify
Anyway, that's some more color as to your deeper *why* question about that log message. You can obviously continue doing it! But given that it can be an anti-pattern, it's giving a heads-up to developers in the console so that they realize there's something to think about.
Reply Retweet Like
Jeff Posnick Aug 7
Replying to @getify
"preload" tells the browser that a resource is needed for the current page, and that signal causes the browser's to use bandwidth it could be using for critical resources. Deferring requests until after onload, inside a SW, means less bandwidth contention for critical resources.
Reply Retweet Like
Jeff Posnick Aug 7
Replying to @getify
One goal might be to optimize for the metrics described at High-priority loading of what's needed for the current page, and low-priority loading of things needed for future pages, is in line with that goal.
Reply Retweet Like
Jeff Posnick Aug 6
Replying to @getify
(I don't want to jump in with unsolicited advice about alternative loading strategies unless you're unhappy with your current approach.)
Reply Retweet Like
Jeff Posnick Aug 6
Replying to @getify
Apologies, but I'm not really sure what kind of response or help to offer. I don't know if you still have an open question about why DevTools logs what it logs, or whether you're looking for feedback about your loading strategy.
Reply Retweet Like
Jeff Posnick Aug 6
Replying to @getify
Getting back to the *why* of your logging question, the idea is to prevent folks from using rel="preload" purely to prime caches or for optional resources. rel="prefetch" might be more appropriate there.
Reply Retweet Like
Jeff Posnick Aug 6
Replying to @getify
I'd discourage you from optimizing for SW caching in that manner. A best practice is to delay SW caching requests until after page load, so that it doesn't interfere with the critical subresource requests needed to display your page.
Reply Retweet Like
Jeff Posnick Aug 6
Replying to @getify
The practical answer re: logging is that I assume the code which logs it doesn't consider a SW fetch() "usage". Taking a step back: is your intention to use preload purely to prime the HTTP cache? Or does your page use the subresource eventually, just not soon enough?
Reply Retweet Like
Jeff Posnick Aug 6
Reply Retweet Like