News for hackers. Stuff that matters.
Re:The impact of removing jQuery on our web performance by Frontend_devs on Tuesday August 16, 2022 @07:00AM (#32481304)
At the end he said a certain cohort of users, but how many users were actually in that cohort? JQuery is an older technology. I find it hard to believe something made to work on older hardware / internet is affecting performance that much. We have 5G and stupid fast chips in phones now. I’m skeptical. Performance issues are usually database queries or code compiling / “waking up” some server function.
Re:The impact of removing jQuery on our web performance by basilgohar on Tuesday August 16, 2022 @08:39AM (#32482006)
There is a massive difference between using a library like jQuery and using vanilla JS for similar purposes. I would not be surprised if a lot of frontend devs aren't fully aware that a lot of functionality is now available in modern, vanilla JS vs. using a library.

The website is a bit of a parody, but the numbers are real: http://vanilla-js.com/

For similar functionality, modern, native JS is orders of magnitude faster than almost all JS libraries. The libraries obviously do more, but they also account for older browsers and just by their nature have a lot more overhead.

The point about edge case devices also should not be missed – the assumption that "we have 5G" should be a target is really quite surprising. There are always cases for a more optimum website. When traveling and there is low signal, or people on pay-per-MB plans, and more – there are all kinds of people that can benefit from a slimmer and lighter (compute-wise) website. The stupid fast chips are not used by everyone, in fact, not even the majority.

Re:The impact of removing jQuery on our web performance by martin_a on Tuesday August 16, 2022 @07:08AM (#32481357)
> We have 5G and stupid fast chips in phones now. I’m skeptical.

I'm not. Rural areas often have unbelievably bad internet connections. In my home town we have around 2 MBit/s for broadband connections and 5G is not there either.

You lose nothing by optimizing your sites for the worst possible case and not for what you'd like everybody to have.

Re:The impact of removing jQuery on our web performance by chrisweekly on Tuesday August 16, 2022 @07:13AM (#32481382)
"Performance issues are usually database queries or code compiling / “waking up” some server function."

That's not true. The overwhelming majority of user-perceived latency for mobile devices accessing typical websites occurs after the initial HTML response has been received. See eg https://wpostats.com or https://httparchive.org for data.

Re:The impact of removing jQuery on our web performance by robin_reala on Tuesday August 16, 2022 @08:29AM (#32481935)
Can I point you to this blog post from edent, on the importance of making government websites work for everyone: https://shkspr.mobi/blog/2021/01/the-unreasonable-effectiven...
Re:The impact of removing jQuery on our web performance by wjdp on Tuesday August 16, 2022 @09:37AM (#32482533)
This quote really hammers the authors point home, I remembered this article on this alone:

> But the GOV.UK pages are written in simple HTML. They are designed to be lightweight and will work even on rubbish browsers. They have to. This is for everyone.

Re:The impact of removing jQuery on our web performance by solar-ice on Tuesday August 16, 2022 @08:17AM (#32481807)
gov.uk serves all of the UK - including the person who lives in the middle of the mobile dead spot, and the person who can't even get consistently working ADSL. 4G/5G doesn't cover the UK by a very long way, and there's plenty of the UK who cannot afford anything but the cheapest smartphone.

It's much less common in the UK than the US to buy a smartphone on credit - cheap phones with cheap pre-paid plans are normal - so low-end models with very slow chips are more common. The sort of thing where even scrolling through your contact menu will induce lag.

Re:The impact of removing jQuery on our web performance by david_allison on Tuesday August 16, 2022 @07:07AM (#32481354)
> We have 5G and stupid fast chips in phones now.

The people who need government services the most won't have this

Re:The impact of removing jQuery on our web performance by atonse on Tuesday August 16, 2022 @08:20AM (#32481833)
Do we have hard data on this?

Just want to make sure we aren’t just making assumptions.

I would think most folks of any income would replace their phones even every 5 years, simply because the batteries would’ve died. And I’d think any phone in the past 5 years would be reasonably decent. Even the ultra cheap ones.

Plus with networks, cell carriers are at least incentivized in general to get folks off 3g into lte since it’s cheaper for them to not have to maintain N systems.

Re:The impact of removing jQuery on our web performance by mejutoco on Tuesday August 16, 2022 @08:27AM (#32481920)
IMO the key concept is that the government does not serve most folks. It serves all folks, so it is good to be slightly conservative in terms of bandwidth.

For data maybe Android share in Europe could be a start, or people with prepaid plans. In my experience both are much more prevalent in Europe than in the US.

Re:The impact of removing jQuery on our web performance by youngtaff on Tuesday August 16, 2022 @09:06AM (#32482240)
GDS have it (or at least proxies for some of it) as part of their RUM data

Device memory is a good proxy for how powerful the phone is

Effective connection type will give some data on network performance

Re:The impact of removing jQuery on our web performance by kwhitefoot on Tuesday August 16, 2022 @07:02AM (#32481321)
You have all those things, I have some of them, many people do not.