Everade

On 4/29/2022 at 7:55 PM, Randy Calvert said:

Given that Yandex is only between 0.5 and 1 percent of the searches done online… I’m not sure how much of a priority this would be for consideration. 

Honestly you would most likely be better served having a custom plugin created to support this. 

First and foremost, Yandex market share is aprox. 50% in russia. Almost as big as Baidu when talking about world wide searches. Both Baidu and Yandex are a big deal in their respective country.
Bing aprox. 3% worldwide. Still big. There's nothing compared to google when it comes to searches, that doesn't mean we should leave everything other than google in the void.

Secondly, it was a simple example of a much broader issue regarding "not valid RSS elements".
There are many many services that use custom elements and Yandex being one of the bigger ones.

I'm not interested into implemeting Yandex Turbo Pages, i'm interested into having the possiblity to support custom RSS elements/tags as shown in my original post.

Still no possibility to handle Yandex Turbo Pages or any other custom RSS system with the current tool set without modifying the source code.
Please let us create custom tags.

Would be great if we could create stackable achievement groups.
A great example would be "Member for 1, 2, 3, 4, 5, 6.... years"
So they either replace the prior achievement entirely, or they visualy stack/layer on top of each other.

Also, the achievement icons should be enlarged in the "Badges" overview when visiting a profile.
The numbers are way too small to be in the readable spectrum for many human beings.

On 4/29/2019 at 12:37 AM, Michael.J said:

1. This might be possible but may be further down the line due to the complexity of achieving this.
2. Do you need the ability to edit each fields message? Because if not, that text is covered in the language manager and can be changed that way as well.
3. The database check is probably forcing an exact match in that case. The "contains" option would be an alternative but I will have to check if a database check can be done that is in-case sensitive.

From what i can see, there have been no further updates on this plugin since my initial review.

2. I've already answered that in my review, yes each field individualy.

22 hours ago, Lukazuki ❤ said:

show me the name of the forum software 'CMS' that have in the default template your button switch between light and dark 🤨

Someone here is fearing to missout on dark theme sales.

This would be a great addition, and has become standard for any great web service these days.
For example right now i'm almost losing my perceptual vision because i'm on this website.

Allow HTML posting for admins, then add the appropriate HTML code to post a normal image in the source.

It's a chore but seems to be the only workaround we have.

I wouldn't allow this for normal users though. (Security)

There should be an easy way to post as many images as we want, as easy and quickly as possible with or without light box.

Would be great if there's an option to remove the redirect_hits counter that's currently "hard"coded into the default theme (forums->front->index->forumRow)

The counter seems to be a very specific use case, and it would be lovely to be able to easily remove it without modyfing the theme on every update.

14 hours ago, Muddy Boots said:

@Everade Why dont you choose a different platform and leave instead of whinging on about IPS developers/ment

Yea, why would a 16 years customer complain about a product. Everyone should just keep their mouths shut and look how the company runs its product into the ground.
Hell no, i want to see things get better so i actualy can continue using it. Got a proplem with that?

They've been doing many many bad design decisions in the past, and are most of the time working on content no one ever asked for.
And i'm not saying new features are bad, i'm just saying that they never focus on actual practical customer issues that have been reported years ago.
But they don't, they keep ignoring us but continue taking our money.

It's a great product overall, there's no reason to move.

But at the current state, there's a good reason to stop paying the subscription or at least pause it for a longer period.

On 10/28/2021 at 4:36 PM, Matt said:

Yes we do monitor web core vitals and have plans to improve performance. 

You forgot: "In 5 years from now." 
And yes, i mean it.

On 10/27/2021 at 1:05 AM, Egorkin said:

Does the team test the layout according to these indicators?

Maybe they do, but they don't care. There are hundreds of community posts screaming for improvements but they couldn't care less. You're better off preparing for your funeral than waiting for a patch that adresses long standing issues. They simply don't have a big enough team to work on whatever they want + plus community requests.

5.x may address some issues, but you're better off not betting on it. Never take their word, there's still a ~3 years waiting list of other promises.
But what you can expect is another new re-design as usual. Because new looks and marketing sells best. Let's just cross our fingers that at least some of the underlying ancient code gets updated a long the way.

On 10/9/2021 at 8:29 PM, Randy Calvert said:

But again, why? 

Because there are smart people out there who can get easy faeces done in a matter of minutes or even seconds, unlike some other devs around here.
If you've ever worked in an open source git project then you would be aware how well it works. But that's an entirely different story.
 

This has nothing to do with customization, but with getting their faeces together and offering a proper product.
Customizations only require additional overhead for months on end on the customer (or a plugin which is lots of additional work), rather than getting something so trivial done within 1 or 2 patch time-frames by invision.

They have a subscription based model and ask for a substantional amount of money from each customer.
So it's not too much to expect a decent patch policy rather than a "fix it yourself" mindset.

And it can't get any easier than this. (There are many similiar topics that would be benefitial and easy to fix that are sitting ducks for years)
It would be fixed faster than reading through this nonsense topic that would not have been required if it would have been implemented properly in the first place.

I guess i'll miss out on the next year of the subscription.
It's just not worth it looking at the micro patch notes and still missing out on all the community requests.

I assume they fired most of their devs years back, because updates have been stagnating in the past 4 years.

Another update and this simple thing still has not been adressed.
Feels like 1/2 a dev is working on the subscription based IPB these days.

Latest update broke your Legend News system css for dark mode:

On 6/29/2021 at 5:39 PM, Stuart Silvester said:

The enclosure is the related image (it would be the record image in this case), Summary and Thumbnail are not valid elements for an RSS 2.0 feed - https://validator.w3.org/feed/docs/rss2.html

Category is really the only thing that would be 'possible' to add.

I would like to add that there are many great and well known RSS feed markups that ask for custom elements.
One example is the Yandex Turbo Pages
https://yandex.com/dev/turbo/doc/rss/markup.html

We're lacking the flexibility to do anything at all with IPB's RSS system.

One single line is all it needs my fellow devs:

"purpose": "any maskable"

Could you increase the space between the day/night button and the menu? (mobile only)
It's very close on mobile.

On 8/5/2021 at 11:28 PM, Danny Michel said:

Looks like it's still happening

I used SEO tools to identify all pages with broken images and fixed them manually.
IPB has removed image fixing tools a long time ago. I doubt they bother giving us these back in a timely manner.

On 7/29/2021 at 5:06 PM, Dll said:

The thing is, unless the image sizes are specified on the page, it doesn't matter which method you use, the page will still jump around as it loads. So, instead of height: auto, as it seems to be at the moment, the height of image thumbnails really should be set to the correct number of pixels. 

Yes, layer shifting still happens without specifing the size, that's why you should always specify the size. But layer shift is a whole different story. 😉

However even without specifications, it would still gain a big performance boost because the important content can be loaded before images are starting to show up.
Which is the main purpose of lazy loading.

On top of that, the current system blocks the entire text to be loaded when scrolling, which is most likely the worst case scenario for a forum.
 

On 7/29/2021 at 11:47 PM, Ibai said:

I would not recommend enabling lazy loading for all the images. It makes the forum way slower in my experience.

Test it and see it for yourself.
It's technically not possible to make it slower, that's not how any of this works.
You don't want to enable lazy load in every case, critical images should be loaded directly of course, but that's basically not existant in a topic and/or post.
 

Lazy loading reduces initial load time due to reduced page weight, which is one of the most important things for a website.
It's also a bandwith conservation as it only loads content on request.
And a system resource conservation, as it conserves both server and client resources.
Which is similiar to what we have right now, but WITHOUT the mentioned drawbacks in my original post.

As long as you're not running it on Netscape with a last century potato on dial up that doesn't support lazy loading of course.

1 hour ago, 13. said:

It would be better to use the native method everywhere and fallback to the JS-based method only when needed, so the majority of users of modern browsers won't experience problems just because of a minority of technically backward Apple browsers. It's very easy because native lazy loading works perfectly fine when used with JS-based method at the same time.

Getting school flashbacks where you have to wait to move on because the stupid kids still don't get 1 + 1.

2 hours ago, Stuart Silvester said:

We added it to profile photos.

We can't completely replace the Javascript solution yet, Safari & iOS do not support the native method. It is on our radar though and we're keeping an eye on support.

Fallback like @13. mentioned please.
Your product's main feature should be prioritzed above 2 bad browsers that form a minority.

Altough yes, safari is the 2nd most used browser, it's still only 18% of the world market (and declining).
So a fallback for these should do until it can be removed eventualy.

https://gs.statcounter.com/

It's not just a simple scrolling issue but an overall performance killer.

This is what scrolling topics with images in IPB looks like:
(Tested with 1gbit/s internet connection <10ms latency)

Why is this bad?
Text + Images + Scrolling = Forum

And this affects 3/3 of your product. (quickmath 😛)

What i'm trying to say is that, new features are cool, we love them.
But don't forget your roots, they need some love too.

Reason for the scrolling/loading issues:
IPB preloads the spacer.png file instead of the actual image, then it goes for the image itself.

This blocks all content from being rendered, because that's how browsers handle it, which results in a truly awful scrolling experience.

It's normal for community members to sometimes come up with a great piece of topic or post, which consists of many images to make it look nice and clean.
So it's not uncommon, i see this problem on my board all the time.

Todays browsers all feature native lazy loading.
 

Spacers' death shall be imminent. 

On 7/19/2021 at 1:58 PM, Square Wheels said:

I didn't like the badges from when I first saw them.  Now that they've been in place for a while, they look even worse.  Plus, I have no idea what their purpose is.

I've heard of tools that allow designers to design badges.
Could be a solution to fit your taste, whatever that might be.

Once upon a time, you threw a SEO improvements article at us.
It specificaly claimed, that it fixes this SEO specific issues related to duplicate pages due to the GET parameter usage.

However there's either more that's missing, or you've re-introduces new issues along the way.

Yandex has been screaming at us for quite some time, that there are duplicate pages based on GET tags.

Yandex is specificaly pointing at sortby and tab to be the main issue:
/?sortby=date
/?tab=comments

The appropriate solution (according to Yandex):
Yandex asks for Clean-Param Directive if the site page URLs contain GET parameters (such as session IDs, user IDs) or tags (such as UTM) that don't affect their contents.
Not sure if Google is going for the same, but they're usualy pretty close.
https://yandex.com/support/webmaster/robot-workings/clean-param.html

Looking forward to further SEO improvements.

The dark blue buttons are barely readable.

These can be found in the Statistics pages.

I did some more tests and came to the following conclusion:

Baidu doesn't accept the following generated IPB sitemaps:
sitemap.php
sitemap.php?file=sitemap_applications

While these work just fine:
sitemap.php?file=1_sitemap_database_categories
sitemap.php?file=sitemap_content_forums_Topic_1
sitemap.php?file=1_sitemap_database_records_1
sitemap.php?file=sitemap_content_forums_Forum

Here's an example from Baidu:

I'm not sure if Baidu generally rejects Indexes or if it's an issue with the delivered format.

One more dark theme issue are answer ratings: