Matt

Changed Beta Release to Yes

👀 From the first commit (mostly v4 at that point) to the first real alpha: 2059 commits in just over a year.

ETA 2 minutes...

Who wants another blog entry today? 😮

Little update: bugs are being squashed slightly faster than they are being reported, which is a good sign. We've got two blogs queued up, one for tomorrow and one for next week. I've just seen the video Ehren has put together and it's his best yet. It's fitting that he opened the v5 video blogs, and has created the last. That should be out next week. We will have a bunch of other blogs after, but this is the last major feature reveal. As for alpha testing v5, don't worry because... (After we've got things a little more stable or the bug report forum will be like Harry Potter's living room but instead of trying to stop Harry opening an invitation to Hogwarts, I'll be trying to stop our devs from having a breakdown).

Around 45 bugs fixed so far, with another 39 in the queue.

How's your Thursday going?

It is no longer flawless.

Well, it appears that after 4 minutes, it's entirely flawless so we may as well just release final?

I've invited about 25 people in for the first few days. This is just to control the flow of reports in the early days. We expect there to be a lot, from the new demo system through to the product. Once things are a bit more stable, we'll invite everyone else in that has requested to test. Thanks all!

Only 2? That's amazing as you built the deployment system. It means it only broke once?

Who is ready for an alpha site to play with?

No, but it might be worth considering a one-off script/tool that does this pre-upgrade.

Update: The alpha system has been built, and all the initial teething problems using Invision Community v5 without IN_DEV have been fixed. As mentioned above, currently all JS/themes and CSS is built on demand, and rebuilt on demand (often quite a lot). In v5 it's built once per version and stored in a new location, so there were some teething issues with that. We'll test this new system internally today and if all goes well, we can start getting people in tomorrow.

I mean, that's a work of beauty. I recently did something similar by collecting all the cables, small electronics (USB hubs, etc) and put them in a single plastic crate. Now I know where to go when I need something, and if it's not there, I can buy a new one. 😆 @Esther E. has been busy listening to your feedback, and she's implemented two changes (I'll update the blog shortly). The first, is that we're going to convert all your open tags over on upgrade, be that 10 or 10,000. You can then decide if you want to delete any from the AdminCP post conversion. The second is a bulk tag add/remove tool via multi-moderation.

There's plenty of time to think about how you want tags to work in v5 and when you are ready to upgrade you can tweak tags on your v4 prior to upgrading.

Yes, all these, regardless of how many, will be converted.

See my post above. I think there's a little confusion over the tag conversion process. All 'closed' tags admins have created are converted over. We then offer to convert the top 25 user created tags over, as well as any per forum/category tags.

To be clear, all "closed" tags created are converted. These are the tags you set up as an admin. The 25 number is based on the top "open tags" that members have created.

That's correct. That's why we're here talking about it. We made decisions, and then adjust as needed based on feedback. However, I'm not sure that use of tags is mainstream. It's unusual to have 1000 tags, and the application is niche. But this is an open conversation.

We picked a number, if people need more then we can convert more. What we don't want to do though is set people up to fail by converting over hundreds of tags because it's likely you'll never get round to setting them up and managing them properly.

1) Each tag must be created in the AdminCP (Content Discovery > Tags). It's a bit like the current "closed" system where you have to add the tags manually if you want your community to use them. 2) No, this was a strong part in the new system; if it is going to work, it must be universal. To start omitting tags from some apps, some forums, etc weakens the entire system. This is a way to bring content across your community into one place. If you need tags like status labels (open, fixed, etc) then I would recommend a Pages database and custom fields. Think of it like organising your house. You have a lot of things that can be hard to find (I know I had a charging cable here somewhere...) so you buy 10 large plastic crates and organise everything into those boxes (electronics, clothes, toys, cables, etc). You label each crate and it's now easy to find what you want. Need a charging cable? Open the cables box. Need to find your Gameboy (it's 1999 in my head) then go to the electronics box. It's easy. If you then added a bunch of rules on top (blue cables can't go into any box, orange toys can't go into the toy box) then you break that simple and strong organisation. Likewise, if you chose instead to buy 500 small lunchbox sized boxes and put one item in each, then you have changed nothing, objects are still impossible to find and are not in the same place.

A lot of decisions to make. Why do you have 60,000 tags? How do they benefit your community?

1) The top three items are from the normal "Featured Content". Oh, actually, we haven't spoken about that much have we. So we merged "featured" and "Our picks" into one thing. The old "feature" didn't really do much apart from allow you to filter in a widget and show a featured block above the content. This new page is much simplified. You can still feature replies and show the post at the top of the topic as you can now. I should probably do a blog on all the other stuff we did but haven't spoken about. 🫠 2) Right now, all tags but I think it would be nice to have a toggle in the AdminCP form to determine this. 3) Yeah, nice idea. We'd need to be careful with Pages collisions. 3) Um. We can add a bunch of hook points ads can use on that page. Absolutely. It offers another way to categorise content across all apps within the suite which was the primary point - we wanted a way to bring all that content into one place and have it all look universal. It will require some thought and it works best with fewer tags, not more. If you have 100+ tags then it's not as effective. Ideally you want broad tags rather than narrow tags. It uses the standard autocomplete form widget which will offer a search box if there's more than N items (can't recall offhand what that is) I think we will need to add some bulk tag management in some form. Right @Esther E.? 😅 It still lives in another branch. I'll be updating the alpha site soon.