Matt

Do you see broken images in widgets?

If so.. yeah, the widgets are cached. I'll have to invalidate those caches when a file is edited. The old thumbnails are removed, so that's the issue there.

13 minutes ago, Hatsu said:

Upgraded from 6 to 7, this time with PHP 8.2.26. No problem. Should we test 8.3 next time again?

That bug is still open. Had a meeting with Esther yesterday to run through the last of the bugs and we should get this one done for the next beta.

If you see comments no longer showing on Pages articles, see the original post of this topic. 🤭

Beta 7 is here, and I'm digging deep into my Bingo Lingo today!

As always, check out our release notes for a full list of this week's fixes.

Update:

If you notice Pages article comments have vanished, do one of two things:

Edit each database and under display add a value for comments per page.

Wait until after 3pm GMT and apply the patch. 😇

Hi Harry,

I'm sorry to hear that you're having some problems with your community.

In order to help us understand the problem better, can you let me know a few things please?

1) Your OS (Windows/Mac etc)
2) Your browser and version
3) If you choose 'remember me' when you log in
4) If these issues tend to occur for a short period of time and then do not reoccur for a few days.

Thanks!

The upgrader does not remove this data.

The only place we keep the raw file contents in a table is in theme resources, elsewhere it's fine. We chunk uploads and they get sent straight to disk/S3.

Haha, sorry - I've been on planning calls today and I don't think there's time to run through the open PRs, build and test before my end of day. It'll either be a lot later today, or tomorrow morning.

It took a while to figure out, and yeah it was just a few lines of code in the end.

Fixed in the next beta.

Looking now.

Probably tomorrow.

Ok, update your license with your FTP and ACP and we'll have to add some debugging and upload a new version and see what is going on.

I see one broken image, is that normal for when this happens?

We'll probably need FTP and ACP. Can you update your client area with those details?

Thanks.

13 minutes ago, Kirill Gromov said:

Same bug

Is this broken now?

Yeah, we have no idea. @Adriano Faria let us know when its in a broken state again and let us take a look before editing/fixing the broken files.

I think Esther may be onto something locally.

I can't reproduce this either, and I've really tried.

Next time it happens, please let us know ASAP (mmecham@invisionpower.com) or DM and I'll take a look at your site before your restore the screenshots to see what may be going on.

From what I can gather, the template does have an <img> tag, but the image URL no longer exists, yet all the uploaded images are still attached to the record, so I'm really unsure what is going on.

Thanks Day! Let me know where to send the money. I have two briefcases and the combination on both is 1337.

I like the car analogy. I have a Tesla Model Y, and it doesn't have full auto-driving because I didn't pay for it. I don't see that as a missing feature though. I think a lot of that thinking is just down to our legacy of having a single product for all.

And don't worry, I love what I do and I'm genuinely thankful that I get to do it.

The only constant in life is change, and we've had to change and adapt over the years. Some use it as a stick to beat us with but without adapting we'd have vanished years ago.

There's no need for apologies for this topic. It's better to air concerns and have a discussion than let the fester in the dark.

I appreciate your custom and sticking with us.

Our primary aim is to safeguard the future of Invision Community, not just for those that work here but for the thousands of communities that rely on us to keep delivering updates and new features to help keep their communities active. It means making tough decisions at times, and these aren't always popular but they really do help secure a future.

The most profitable model possible was the one we had 15 years ago when we were selling hundreds of licenses a month at $149.99 a go and our only costs were answering support tickets.

Our monthly AWS bill is quite frankly a hate crime.

3 hours ago, Omri Amos said:

I get that, this makes total sense (considering your high prices for the cloud hosting), but your business started as a software, and some customers rely on that software.

I have my own server (by the way, costs less than your cheapest plan..) and I use it to host my Wordpress website. This website also includes the IPS forums, with tight connections between them (custom automatic SSO, articles->topic and comments->messages syncing, etc) - I technically can't and won't switch to cloud hosting just for the forums - it simply won't integrate well with my entire website that I have around the forums.

I believe you should offer your cloud hosting to the customers that are interested in that, and you can even have some cloud-only features if that it technically only possible to make these features under your own cloud platform.

But what I don't understand is why you are limiting and blocking features, that as far as I understand, CAN technically run on the self-hosted software, and it's limited just because you decided it should be a higher-tier only feature - and that tier is unavailable at all for self-hosted customers, for no real reason other them trying to lure them into the cloud hosting option.

Limit features - that's fine.
Create higher tiers - that's fine.
But let us actually choose these tiers, even if we choose to self-host the software on our own servers, for whatever (technical or not) reason.

I think I answered this fairly comprehensively in my post above.

2 hours ago, PanSevence said:

Let’s appreciate the work of the Invision Community team in developing the IC5 software. As a self-hosted user, I fully agree with this approach. The christmas holidays are coming, so let’s enjoy what we’re receiving because the new version of the software is amazing and brings great value. Let’s hope the new business model for self-hosted users meets your expectations, and we’ll see another version in the future. Unfortunately, looking at the issues caused by self-hosted users, I think there’s a high likelihood that you might move away from this model entirely, but that’s just my humble opinion.

Much appreciated. Most customers are fine and topics like these are just concerned customers voicing their thoughts which is fair and reasonable. 

I used an artist on Fiverr, I can DM you the link if you'd like.

Ok, we still have plans in that direction, but not everything will get into 5.0.0.

1 hour ago, EliasM said:

I would like to suggest that all the features that can work on self-hosting be available in the classic version with different plans for the classic version starting from $500 to $1500

I genuinely do not think there is a market for a $1500 self-hosted option outside of enterprise which often have needs way beyond hosting things themselves and the cost to run that size of community means we're probably much more cost-effective just to host it anyway. We have more knowledge and experience with our platform and the resources it needs.

2 minutes ago, beats23 said:

True. It takes 100 cents to make a dollar, which means every cent counts, no matter how small. I'm 100% sure your company would make extra income by offering extra features, such as your course app, to classic customers. 

But we risk spending 150 cents to make that dollar by doing that.

16 minutes ago, EliasM said:

I would like to suggest that all the features that can work on self-hosting be available in the classic version with different plans for the classic version starting from $500 to $1500, and we would be free to upgrade or stay in the discounted package,

Yes, it is technically possible but doesn't fit with our business plan.

To do that would mean you'd need to be very rigid with the type of host. We'd also have to completely re-tool those features already written to work with that droplet/docker image. We'd also not really be able to make it profitable given the number of people willing to host things themselves.

There are very few self-hosting only community platforms left. There is a reason for this. 

9 minutes ago, beats23 said:

It was stated in the beginning that IPSv5 would have features like paying to view a forum topic. I don't see any pay-to-view features announced. Are those features still coming in the classic package? 

I don't remember this being discussed as something on our roadmap specifically. There was a notion of doing more to monetise communities around the start of v5.

21 minutes ago, EliasM said:

The only thing that is annoying is that you did not clearly mention during the post that the feature is still not specified for cloud customers

Yes, I addressed that in my post but a shorter version is that discussion around self-hosting licenses and not getting features is toxic and dominates any feedback we get from a small minority. We didn't want to make announcements about that until we were very sure about it. We are still in the early stages of beta testing.

13 hours ago, Marco Junior said:

Why not release these features for us to test in BETA?

Live community features, and the topic summary feature relies on our own cloud platform. The topic summary is actually pretty clever in that our cloud platform pulls data directly from the database outside of the software and runs the whole thing through a custom algorithm written in python to calculate a score for each post. It then updates the database. This isn't run via tasks or cron it's run via our own queuing system on cloud. 

Topic assignments is likely to be added to the business tier and all those above. We are no different from others selling services that restrict some features to higher plans. In the past, plans were based around storage GB and bandwidth but that's rarely the case now. Tiers are now based on feature sets.

I know it can be uncomfortable to talk about the cold realities of running a business, but the landscape has changed for community platforms.

We feel that relaunching self-hosted with the Classic license has allowed self-hosting to remain viable for the life of v5. That doesn't mean it won't be the case for whatever comes after, it just means that no one internally is even thinking about v6, let alone making plans for it. I can't give you a cast iron guarantee of something that may not exist. Who knows where we'll be in five years time. 

In 2024 to date, cloud represents 86% of our income. Without Cloud, Invision Community would not be here, or at least not as we know it now.

By having a profitable model, we can still provide a self-hosted option which just about breaks even or makes a small loss depending on how many bugs us developers add in. We are happy with this. Hopefully you are happy with this. We have committed to this model.

But it does mean that there will be a divergence in feature sets. Partly because it's hard to build modern features with just PHP and MySQL and partly a business decision to entice people up the tiers.

We could create a docker image with node, etc and be incredibly rigid with hosting environments to the point only a few hosts are supported with droplets, etc. But that would decimate our self-hosting customer base with most not being able to run the required servers, or not having the knowledge to set it up and it would decimate our cloud business by offering the same functionality but elsewhere.

That's the business stuff, let's move on to transparency.

Part of the discourse around self-hosting versus cloud (even though we never frame it like this) is toxic. There is a small but incredibly vocal minority that refuse to accept that they cannot get every single feature we write and drag every feature release blog down with negativity, angry posting about cloud's existence, and accusations of 'money grabbing' (you are very welcome to pay our AWS bill). This can be hard to handle in a public forum. We have two options; endlessly have the same conversations or hide/block/ban posts and people which then comes across as thin-skinned and unable to take criticism.

Ahrefs does not have the problem that we have. Their customers are all 'cloud' customers and they're used to not getting all the features, so to say "yeah we'll test it in beta but you might not get it" will not be met with (much) anger and resentment that we seem to get when we mention "cloud only".

This has been ongoing for years and is the reason why we heavily moderate comments on our blog posts. You don't get to see a lot of the rage posting because we never approve it. But it's there. It's always there.

With v5, live topics, live community and topic summary were only ever going to be cloud only due to the underlying technology. Two out of three of those already exist on cloud so should come as no surprise that they remain so.

The other features we were undecided. From our point of view, everything is a tier. We could bring topic assignments to Classic, but then it would mean that we'd need to bring it to all cloud tiers. Which means that there's little incentive to pay for a higher tier level.

We could have said "this feature may be on a higher tier than classic" on the blog announcement but it would only invite more toxicity and more anger from this minority group. We wanted to wait until we had it decided. And part of that decision was based on how it went over during the alpha and beta stages. When we create a feature, we never truly know how it'll shape up, or even if we'll keep it; more true during the early stages.

Now we're in the beta stages, we have a firmer idea of its usefulness and future direction and are more able to make those decisions. That's the honest truth. There is no conspiracy. We have customers who run large enterprise communities testing these same features and offering input too.

We could have done better for sure. But it's a really tight line we're trying to walk on between informing customers about where possible features may end up without instigating another round of self-hosted versus cloud which is draining quite honestly. Not to mention demoralising for our team who proudly present their work and a good portion of the feedback is on why self-hosted licenses aren't getting it.

But moving back to features, if you look at the volume of work that has gone into v5 and all the good stuff that's been added, it really is a very small percentage of features that will not make it to every single product tier we offer.

As always, we may not be perfect, but we're here and talking about it.