Registers an unload listener

[Adlago]

A new version of Lighthouse 6.3.0 reports this issue for best practice

Please comment.

[bfarber]

We will review this, thanks for bringing it up.

[Adlago]

On 9/9/2020 at 5:15 PM, bfarber said:

We will review this, thanks for bringing it up.

Did you consider this?

Thanks

[bfarber]

We have a general SEO audit/review planned and this is on the list.

[Adlago]

This is still not solved in version 4.6.0.

From this site

[SeNioR-]

Unfortunately, bfarber is no longer in the team and the error has probably been forgotten.

However, it is worth taking a look at.

Edited June 2, 2021 by SeNioR-

[Adlago]

1 minute ago, SeNioR- said:

Unfortunately, bfarber is no longer in the team and the error has probably been forgotten.

However, it is worth taking a look at.

I noticed this, so I brought up the topic, and it's still available.

[Bocar]

Hello.

 

To fix the unload listener, find and replace "unload" by "pagehide" or "visibilitychange" on your source file.

[Adlago]

27 minutes ago, Bocar said:

Hello.

 

To fix the unload listener, find and replace "unload" by "pagehide" or "visibilitychange" on your source file.

This "unload" is contained in a root_library java script  - and should be edited by IPS developers.

[Everade]

Looking forward to this.

[Adlago]

Hello, IPS team, are you planning a solution for this?

[Rikki]

It isn't currently high on our priority list. It's called in a third-party library we use.

[Adlago]

On 9/22/2021 at 4:20 PM, Rikki said:

It isn't currently high on our priority list. It's called in a third-party library we use.

Are you planning for the next release?

[sadams101]

My site continues to get warnings for this which reflects poorly on my site's SEO. It appears to be a simple fix, so why not fix it?

Edited April 19, 2022 by sadams101

[Stuart Silvester]

It isn't something we can easily "fix". It's part of the jQuery library we use. You're welcome to start an issue on their Github repo if it's an important thing that you think should be improved in the library.

[sadams101]

So it seems to me that IPB should be the ones to do this, as several of your customers have already chimed in here to let you know about this issue, and most of your users likely don't know that it is an issue, but it is affecting their sites nevertheless.

 

[Randy Calvert]

2 hours ago, sadams101 said:

So it seems to me that IPB should be the ones to do this, as several of your customers have already chimed in here to let you know about this issue, and most of your users likely don't know that it is an issue, but it is affecting their sites nevertheless.

 

Personally while it may be an issue, it’s not a critical one and anywhere near the top of the list of things that are important to fix. 

If this really is important to you, make your voice heard to the group that is actually writing it. While IPB can make a post asking for it, if tons of other people do it as well, it’s more likely to happen. Just like y’all are trying to say here. 😉

This has existed since 2020 for jQuery which has a MUUUUCH bigger user base than even all of IPS.  People are obviously not clamoring to them to fix it. So it needs more attention. 🙂 

[Adlago]

On 4/19/2022 at 8:08 PM, Stuart Silvester said:

It isn't something we can easily "fix". It's part of the jQuery library we use. You're welcome to start an issue on their Github repo if it's an important thing that you think should be improved in the library.

The unload() method was deprecated in jQuery version 1.8 and removed in version 3.0.

Do you plan to change the jQuery version of the root_library?

 

[Stuart Silvester]

21 minutes ago, Adlago said:

The unload() method was deprecated in jQuery version 1.8 and removed in version 3.0.

Do you plan to change the jQuery version of the root_library?

 

We use jQuery 3.5.1, which hasn't changed for a few years. I'm not at my desk right now, but it's likely in one of the other libraries we use, such as history.

[Adlago]

7 hours ago, Stuart Silvester said:

We use jQuery 3.5.1, which hasn't changed for a few years. I'm not at my desk right now, but it's likely in one of the other libraries we use, such as history.

Yes, sorry - I took the trouble to check. Although many supposedly responsible sites state what I wrote in the first sentence.
Ok - you are using a 3rd party library - this library introduces delays that create poor quality in your product. Don't you have any programmers to eliminate this issue in browsers? I think there are very few improvements you can make to take your product much further.
With a few simple (for me of course) revisions of some templates - without touching the CSS, I achieve, surprisingly for me, good results in tests from various well-known analysis sites. I.e. instead of putting effort into the CSS, just remove all the javascript annotations... I can share everything I do absolutely free - for me this loading speed game was and is just a hobby...

[sound]

below gives an interesting view on all this

https://web.dev/bfcache/?utm_source=lighthouse&utm_medium=lr#never-use-the-unload-event

 

Never use the unload event #

The most important way to optimize for bfcache in all browsers is to never use the unload event. Ever!

The unload event is problematic for browsers because it predates bfcache and many pages on the internet operate under the (reasonable) assumption that a page will not continue to exist after the unload event has fired. This presents a challenge because many of those pages were also built with the assumption that the unload event would fire any time a user is navigating away, which is no longer true (and hasn't been true for a long time).

So browsers are faced with a dilemma, they have to choose between something that can improve the user experience—but might also risk breaking the page.

On desktop, Chrome and Firefox have chosen to make pages ineligible for bfcache if they add an unload listener, which is less risky but also disqualifies a lot of pages. Safari will attempt to cache some pages with an unload event listener, but to reduce potential breakage it will not run the unload event when a user is navigating away, which makes the event very unreliable.

[Adlago]

As the picture from my first post in this topic, this issue again concerns a single function from line 2 column 40996, see pictures below.
If I'm raising this topic, it's because with the new changes to the Lighthouse test, this issue has been moved to the Good Practices section, and its existence drops 8 points for that section.
Isn't there a developer here in the community to review this feature and why should it actually be there after being reported for bad practice?

 

[sadams101]

So we are now over 3 years after this thread was started, and the Unload Listener issue is still present in IPB with penalties on PageSpeed Insights. Is there any plan to fix this issue?

[Jim M]

9 minutes ago, sadams101 said:

So we are now over 3 years after this thread was started, and the Unload Listener issue is still present in IPB with penalties on PageSpeed Insights. Is there any plan to fix this issue?

I am not a developer, but as Stuart mentioned above, this is likely contained in one of the JavaScript libraries which we heavily rely on (such as jQuery) and can't rip out for version 4. With that said, our software is very fast and at it's core, meets Google PageSpeed as satisfactory. Remember page speed or arbitrary scores alone will not cause your site to not be ranked, it may be the deciding factor between 2 sites but not altogether. 

Version 5 will have some major changes to JavaScript coming to it though. You will want to watch that information as more and more information about version 5 becomes available.

[Adlago]

1 minute ago, Jim M said:

I am not a developer, but as Stuart mentioned above, this is likely contained in one of the JavaScript libraries which we heavily rely on (such as jQuery) and can't rip out for version 4. With that said, our software is very fast and at it's core, meets Google PageSpeed as satisfactory.

Version 5 will have some major changes to JavaScript coming to it though. You will want to watch that information as more and more information about version 5 becomes available.

After versions 3.x.x for the "future" 4.x.x there were similar promises... Alas, many of the promises remained an empty hope... And whether version 5.x.x will meet expectations - is forthcoming, but the pre-ad is just an ad…