Ever since the early days of Steem it has been obvious that a main feature that could explode Steem (and now Hive), maybe even creating a 'Facebook killer', would the enabling of easy Hive forum/community integration into existing websites and to be able to easily create entirely new websites with Hive too. Years later, we still don't have these essential features easily available for anyone to use.. but we are closer now than ever, so let's take a look at how we can improve the situation...
Background
Back before Hive was born, Smart Media Tokens (SMTs) were touted as being part of Steem's answer to this ongoing challenge. SMTs were layer 1, native - yet custom - tokens that would enable anyone to create their own branded token and use it to power a Steem/Hive backed website too. The potential for this was/is huge. Imagine anyone being able to create their own crypto powered version of Reddit, for their own community and even have their tokens accessible via major exchanges - easily and effortlessly.. Arguably this would have exploded this chain to a household name.
Unfortunately, the events of the hostile takeover of Steem that led to Hive being born, left SMTs in the dirt and they were not continued on Hive. However, the idea did not die completely, Hive Engine launched layer 2 tokens on Hive and enabled custom sites to utilise them. However, the fees being charged for this are prohibitively high and the options for integration are limited.
Excitingly, in 2024, we have two active projects on Hive that are moving to create the code needed to deliver these features to us all. SPK has been working alongside Ecency to deliver their 'breakaway communities' project for some time and recently we learned that the core Hive team are planning to work on creating their own solution for this to extend the system that runs Hive.blog too.
Testing Breakaway Communities On SPK/Ecency
The most promising option today that can enable easy, mass adoption of Hive by niche communities seems to be the 'Breakaway Communities' sub-project run by the @spknetwork, who also run @threespeak, Hive's primary video sharing site. This system uses a fork of the open source code that runs the main Ecency website, but adjusted so that the resulting custom website will only show posts and content from a single community. While the system is currently quite basic, it is already functional and people can already create Hive powered sites using it (see the breakaway communities website for some basic info).
We recently tested out the current version of this system (on ureka.org, the domain I used for my social network long ago) and were asked to provide some feedback, so that's part of the purpose of this blog. However, we are planning to go much further than this and to produce a full analysis/design report that can be used to optimise the User Experience (UX) of the breakaway communities system in order to speed up it's use and adoption.
Before we get to that though, here's a summary of points gleaned from the setup and use of the Breakaway Community system:
Setup
- Install Options: It is possible both to self-host this system on a private webserver and also to opt to have a version hosted by the SPK team. This is similar to how Wordpress operates. It turns out, though, that the system for setting up these communities is incomplete and does not yet support the creation of sites for pre-existing Hive communities. Since I was initially wanting to test against a pre-existing community, I opted to self-host the code.
Suggestion: Update the breakaway communities site to support creating sites from pre-existing Hive communities.
- Instructions: Self hosting is as simple as installing Docker on a webserver and making some basic adjustments to a configuration text file, then running some basic commands on the server. This is straight forward for experienced server admins and developers, but complete noobs will probably need to do some research before reaching success.
Suggestion: Update the breakaway communities site so that the instructions for self hosting are more obvious - they are currently listed on a page called 'breakaway'. This should probably be renamed to something like 'self-host'.
- Server Configuration: One issue I found is that the Docker template auto installs the Nginx webserver in order to host the new site, but I already have Nginx installed and needed to run the site through an SSL encrypted port (https: instead of http:), so I had to manually disable Nginx in the docker configuration file before I could launch the site.
Suggestion: Provide more detailed instructions for the docker configuration file, including how to disable the included Nginx server and also to configure it to support SSL connections.
Initial Impressions & Optimisation Suggestions
- Appearance: The User Interface is essentially the same as the one on Ecency.com. I use Ecency on my phone because I find it to be the most usable option on mobile devices - it has it's own app. However, Ecency on desktop needs some UI improvements. The functions are mostly fine, but the styling and resulting User Experience are not optimal. Links and text do not contrast sufficiently for users with limited eyesight, the colourscheme is uninspiring and generally the site needs to be sexier! The good news is that the functionality of the site seems to work well and also with minimal delay/lag.
Suggestion: Carry out a full UX analysis and redesign of the User Interface to ensure the system is competitive with other social sites right out of the box.
- Branding/Design: I was hoping that it would be at least possible to set some custom colours for the new site, but alas, that is not currently possible. There are some default themes that can be chosen, but they cannot currently be edited. I can remember using Ning.com to run my first social network - long ago - before they kicked everyone off that wasn't a paying organisation! Their sites are are/were a good example of how to allow users to customise a social network's theme.
Suggestion: Site admins are going to need a page in their admin area that enables them to set common colours, fonts and other basic details in the site. It's also important to provide a way for custom CSS to be added (ideally without needing to edit files on the server), to allow other layout changes to be made.
- Initial Site Loadup: When I load the site initially, I see a rather unattractive loading screen that counts down to zero, as if content is being loaded. However, this process occurs even after the site has been loaded and I revisit the homepage. This leads me to think that this countdown is for no practical reason in relation to content loading. This is generally not a good UX decision as it wastes the user's time for no benefit.
Suggestion: Remove this!
- User Login/Signup: This system will continue to benefit from the ongoing improvements being made to the SPK login process (see video here: https://twitter.com/3speaktv/status/1772797118056640538).
Suggestion: **
- SEO: Metadata: Another standard feature that is missing is the ability to edit the metadata for each page. This is essential in order to be able to optimise the site for search engines and to compete with other sites in the same niche.
Suggestion: Add the ability to edit SEO metadata, page titles and page thumbnails (including metadata for all web 2 platform schemas) to the site's admin area.
- SEO: Page Text Content: It is standard and necessary to allow site operators to edit the text content on the pages of their site. This is needed to both help educate new/existing users and to cultivate a unique feeling for the site. There may be unique aspects which only apply to a specific community, such as rules or other requirements - these may need to be conveyed in various places. Additionally, websites MUST have unique text on them in order to prevent them from being penalised by search engines.
Suggestion: Allow page content to be edited, either in the settings area for admins or at least somehow via configuration files exposed on the server. Often this is done by having one file for each language that can be edited by admins directly.
- SEO: General considerations: There are several factors which I believe to be negatively influencing the SEO performance of most or even all Hive sites today. One of them is that when content is duplicated across numerous sites, it is common for Google and other indexers to penalise the sites in their ranking score. There are ways to mitigate this via directives set within the page metadata, but I do not think this is generally being done. If there are going to be large numbers of new Hive sites being created using this system - professionally optimising this is essential for growth and success. We might not like the big search engines, but they remain a primary way to drive traffic to valuable content.
Suggestion: This requires a full audit and research to be able to ascertain the optimal setup to use for this situation. This needs to be done, ideally before the system is used on a large scale - in order to not negatively impact the domains that host these sites.
Bugs
As always when it comes to me testing out sites, I found some bugs! I was once a professional tester and actually once got sacked for finding too many bugs and making the development team look bad - haha.
There's nothing wrong with finding problems in the code, there is probably no such thing as a perfect computer system. The problems come from ignoring them and not fixing them!
Here are a few bugs that I have found so far:
NOTE: I am testing using Firefox on Windows 10
Language/theme Glitch:
I found that if I clicked the 'Create Post' button and changed the theme from day to night mode, I would get the following screen corruption. This also seems to involve the language selector too:
Refreshing the page then loads the dark theme in, even though I had previously enabled the day theme. I then noticed that this actually appears to be a problem on all pages. Just choosing the day theme causes the CSS styling to break.
Profile layout glitches:
If I view my own profile, I see a few obvious bugs. Including that my Hive reputation figure is displayed as 25, instead of the correct value. The profile URL is cut off and generally the text colours are not contrasting well with the background.
Wallet side menu layout broken:
The name of my community is not particularly long, but it breaks the layout of the side menu on the wallet page. The CSS needs to be improved to accommodate the fact that some variables and text labels can be changed and will not render the same as they do on the original Ecency site.
- 'Sign up with Hive' is Broken:
The new user signup page has a button marked 'sign up with Hive' but it does nothing when I click it:
Conclusion
This technology clearly has massive potential. Once we can enable anyone to easily use it and even to launch their own customisable tokens that can run through their own Hive powered site - Hive can really grow exponentially. Reaching this point will require substantial research, analysis, design, development, testing and optimisation.
In the short while that we have tested the breakaway community system, I have identified a wide variety of necessary improvements - many more than I have listed in this blog. In the coming days we will be focusing on this subject closely and preparing a report that includes detailed suggestions for significantly optimising this software. From there we will discuss the possibilities with the Ecency team and SPK team, plus @valueplan and others.
It has already been suggested that the development for this task should be funded via a DAO proposal, so we may look into collaborating on that.
We have little information on the development of this functionality for the 'Denser' app operated by the core Hive team, but our work should be equally useful to them too when developing their equivalent solution - so we would, of course, be happy to work in conjunction with them too - if requested to.
We are looking forward to helping to enhance this hugely important part of Hive's evolution!
Thanks for reading and if you have any questions or comments, please let us know down below!
To our mutual success,
The @strategizer team!
Reminder: The @strategizer DHF proposal for market research funding is still active and as yet is not funded. Please consider reading/voting on it if you like this post.