Some stats going into the new year:
Seems that all 3 of the more modern platforms tracked by the system have continued to gain ground at VBs expense. Although I'm sure there's probably a fair few migrations going to other forum/social/community platforms that unfortunately aren't tracked by this system, such as Discourse and Flarum.
Here's some interesting info about a recent transition for Blizzard and their gaming forum from their old in house forum solution, to Discourse, and why they made that decision instead of moving to a more traditional forum solutions like the ones we're following in this thread.
Kaelon on The Admin Zone said:
This is probably old news by now, but for the past year, Blizzard has been steadily migrating all of its own in-house custom-designed forums to Discourse, and it has done a beautiful job of customizing the Discourse stack for its community of gamers and video game players.
https://us.forums.blizzard.com/en/wow/
Kaelon on The Admin Zone said:
So, full disclosure, while I have a lot of colleagues and professional contacts at Blizzard and in the gaming industry in general, I am not a Blizzard employee and was not intimately involved in the forum transition. That said, I know a lot about what went on and also evaluated Discourse for use on my other gaming sites.
In my view, Discourse is not an appropriate solution for most classic discussion-based forum communities. It is, however, a great solution for faster-paced social experiences. There are several reasons for this:
- Conversations are not threaded, they are linear with threading-like references.
- Discourse is on a Rails stack, and is notoriously resource-hungry. Self-hosters will find themselves spending ~60 min a day managing and maintaining the installation and optimizing their servers.
- Though it is open source, managed hosting for this offering is expensive since Discourse.org is in the hosting business, and other third-party hosts offer cookie-cutter installations.
- Blizzard had a dedicated web team spend over a year custom forking Discourse and then slowly migrating boards over (and they made the decision to just migrate ~2 weeks worth of posts, and then archiving all of their older content).
- UI is streamlined and highly responsive, and has a mobile-first approach which will appeal to current web users across platforms.
For people who typically post less than ~150 words per post, Discourse will work just fine. But for admins who are used to tweaking installations and who, they themselves, are not developers, customizing Discourse is not for the faint-hearted. Other forum platforms, like Flarum, seem to mimic the major UI innovations that Discourse advanced (such as in-thread summarising) while maintaining a more accessible code-base. But this largely glosses over a single glaring problem afflicting all of us these days:
Forums are essentially a dead medium, and outside of niche discussion communities, will not capture significant new audiences at the pre-2008 activity levels. For a growth-based experience, we should be investing in micro-blogging and hyper-discussion platforms, more similar to Twitter, Instagram, and Snapchat -- though these, too, are now archaic compared to the media-rich video-centric experiences that are driving over a third of content creation on the web today. For networks of users creating content, going mobile-first with an emphasis on video would be a better investment of time rather than trying to use a platform like Discourse to catapult your content engine -- which it will not.
Source
Here's the version breakdown for the big 3. Still a shame that we can't see a more detailed 4.x versioning breakdown for IPS.
I think it's very interesting to note that XF now stands at 29% on the 2.x branch a year after release. I'll be keen to see how quickly that increases once 2.1 is released seeing how widely anticipated it has been and it sounds like many big boards have been waiting to see what was in store before planning their 2.x upgrade path.
This graph shows that in terms of conversions from one platform to another, IPS to XF still seems roughly double that of XF to IPB. Although this has dropped marginally over the last 18 months in favour of IPB. I do wonder how accurate these number are though - are 4.2% of forums still converting from IPS or XF back to VB? Perhaps the cases are migrations to a more modern platform gone wrong and the admins succumbing to users demand to get back to the old site.
The slight increase in conversions of XF to IPS does make you wonder what is drawing people to that platform. Is the prospect of a full community suite getting more and more desirable and perceived as a necessity by some in the competitive social landscape of today?
Will XF ever address that in the core? Or will features such as the new API and 3rd party apps allowing bridging with apps like WordPress make this less and less important as admins are more easily able to add these features via dedicated 3rd party integration?
There are some interesting stats
here from an
IPS forum that recently converted from XF showing traffic breakdown on various aspects of the full website within the IPS system:
Systems is classed as more other core pages within the IPS app such as:
- Profiles
- Search
- activity streams
- Account Settings
Going by this graph, only a 1/3 of the site traffic is the core forum, which I assume is primarily reading and writing on threads themselves. That seems pretty low and makes you wonder where visitors are spending all that other time on the site. It's worth noting that clubs isn't broken down on this graph. I'd be keen to see the exact combined percentage of clubs and member map for this site, as it's a very good example of a community with lots of geographic sub communities where both clubs and a member map with both members and local clubs can be displayed and browed by users.
There are some very big XF sites that could do with core features like this, such as Tesla Motors Club, who I believe had to commission their own
groups /
clubs platform. Plus lots of smaller communities who don't have the funds to build their own system, and so have to resort to using some mediocre 3rd party XF solutions or managing networks of facebook groups instead.
Another really interesting takeaway from the above graph is the gallery, and how small a slice it takes up. I think this highlights a core UI and UX posting and new gallery content discovery problem inherent to all forum software suites at the moment. It seems most forum users prefer still prefer to post photos and videos in threads. Probably because they get the most traffic there, resulting in better engagement for the content creator.
IPS recently upgraded their gallery and the new lightbox, it's a great upgrade, but discovery and getting users into and using the gallery can still be a huge issue for many admins. They've tried to fix this with their activity feed, but it's still rather poor IMO and not easy or enjoyable to use. All forum platforms still need to fix that.
On XF1.x I found that Sonnb's gallery and use of masonry effect displaying uncropped thumbnails made it more enjoyable to browse albums and new content. The lightbox he developed was excellent too, and it seems a shame that XF seems to have fallen quite far behind on that front with keeping their solutions as nice to use.
The addition of video upload to posts is a big media sharing hurdle addressed in 2.1, really please about that and it will help in making the platform more competitive with Facebook groups.
To wrap up, here are some more interesting stats monitoring PHP major versions:
18 Months in June 2017 ago vs today in December 2018.
7.x is now at 47.8% adoption rate, it would be interesting to see how this compares to the new stats collected from the XF 2.x installs.
PS. Loving that you can resize images in the editor now! Great for posts like this