Beating Facebook

Hey, there's something exciting happening! We're making a huge leap that'll help UNA-powered communities catch up with or even outdo the world's top social networks. I'll share more info later this month, but I just had to let you in on the results of this cool test I did.

So, I made the same post on both Facebook and UNA, keeping everything the same, like privacy settings, comments, and what you see on the screen. Then I ran a Lighthouse test on both, using a super-fast 1Gbps fiber connection from Australia. Now, Facebook uses this fancy CDN/Edge network thing, but our setup is simpler with a server in one place and the UI in another. We're not using a CDN or server cache, and the local cache doesn't matter because Lighthouse clears it for the test.

I did the test in Incognito mode, so Chrome extensions wouldn't mess with the results.

Facebook result:

After a few attempts, the best score is 1.7s overall speed index. Most if it in FCP (nothing happens for 1.7s).

UNA result:

The speed index is at 1.5 seconds, with most of it in the LCP, which means the page actually starts loading way sooner – the FCP is only 0.3 seconds. The 1.4-second LCP is mostly due to about 0.5 seconds of server response time, where the backend does authentication and sends data to the front-end. Some of it is because of the transport time between two data centers in totally different places, and some due to API performance. Oh, and there's no CDN. With more improvements, we're aiming for a sub-1-second overall speed index in a full production setup. In fact, for simple pages, we should make it twice as fast as Facebook. And, hey, our Accessibility, Best Practices, and SEO scores are perfect now.

Why does it matter?

A modern social network must provide fast and seamless experience. For a software platform like UNA this is the fundamental requirement. You can add more functionality and customise various features, but the core performance must be same or better than established market leaders. The rest comes down to creative ideas and correct implementation.

How and when?

This is a new, evolutionary development we've been actively working on for the past 6 months. I will share more details by the end of the month with a way to test and provide feedback.

Need it now?

General availability is still a bit off, but we're doing custom builds that include this new development. If you're interested in having our team create something just for you, book a meeting, and we'll show you what we've got up our sleeves.

Email me directly via ay@unacms.com or book a meeting

  • 3141
  • More
Replies (11)
    • when browsing one, every time we go to another page the logo is reloaded and the same static resources are reloaded with each new page even though 80-90% have not changed, how do you want to beat Facebook with such an architecture? Imagine what this means when you have a billion users. All the electricity on the planet is not enough to feed what UNA will consume to feed the necessary servers. I don't know about the new version, but the current version leaves a lot to be desired. this problem has been unresolved since 2018 https://una.io/d/header-reload

      • That's exactly what we are working on ✊and I'm showing the results of what we're achieving thus far. Will share more details when it's a bit more ready for live testing.

        • Let's hope that this will be resolved, if this happens, then all the UNA sites will fly... until then, I honestly do not have the courage to invest in a site that is going bankrupt. What matters in a business is the difference between the resources consumed and those gained. In a market where advertising is the only hope for profit, the price per display and per click is competitive. We will never be able to beat a big site if the site's code is not clean and light. To work with large numbers, any critical point means a hole in the pockets, and bankruptcy.

          • Hello do not forget about https://validator.w3.org this thing have important too and little microdata to be #1. Thx.

            • My experience with UNA has already shown me, ways in which UNA is more advanced than Facebook or Twitter. The modules or plugins extend the functionality beyond other platforms. And the Admin can add pages with HTML or I frames, to add anything the Admin wants. I like turning the Post section into an Online magazine, accessable by anyone on the web. It's a better vision of what social media can be.

              • Speed is great but alot of us on UNA Cloud are still waiting to get access to streaming servers and Jitsi servers. Many ideas revolving around those have been sitting on the back burner for years and now other social media sites have used those ideas and advanced them. UNA Cloud members need streaming/jitsi server access and many 3rd vendors cannot help make our sites unique because they do not have access to Cloud Servers. This has stunted the growth of many sites who cannot show off a site because features aren't available for Cloud Members. If my site had access to streaming/jitsi servers and UNA Developers helped Cloud members implement mods that need access to to servers then my site would be growing astronomically. Without ways to make money my site has been in the red for years and does not stand out compared to Facebook and other social media sites. The question you have to ask is what can UNA do that would make users want to stay on UNA sites more than Facebook, Instagram, etc..

                • UNA can do all these things already. Streaming/Jitsi server, etc. We have it at www.SafeSocialOnline.com. Just waiting for investors to take it to the next level of safety with AI Moderation of images, video, and audio.

                  • I hear you. The performance post is not just about performance. We will need a bit more time before we can share details, but the whole point of this development is exactly about looking at what makes the leading social networks so engaging. It’s not about adding new features at this stage, rather it’s about streamlining user experience.

                    • People on UNA Cloud servers cannot do streaming yet or personalize Jitsi video chat/calls because UNA Clouds are not configured for streaming/video chat. Yes I could use another server for the streaming but then I would have to juggle two different servers and if a bug is found on one server that effects the rest of the site on another it would just cause more and more headaches

                      • Facebook sucks. I am only using Facebook because I have to for some reason. It is now full of virus ads and shit reels.

                        • I agree. It does suck and yet Facebook entities still account for solid 70% of all social networking traffic. Moreover, they are a good reference for performance, which is what we are mostly concerned about. Eventually, I hope, an UNA-powered site, or a few of them will kick them out of existence.

                          Login or Join to comment.