No public commits to highlight, at this moment mainly because I haven't been able to prioritize the devportal, since my proposal hasn't been funded yet. It's just a hobby, right now. So I thought I'd go over the previous commits.
The way Hive does Testnets, they must be redeployed at least once every three months. Assuming there's development on hived, it should be much more frequent. I ran a public testnet prior to the Eclipse Hardfork, but have since shut it down since it's a bit of an expense for me.
Anyone who absolutely needs a testnet right now can always use tintoy, which just requires a basic docker setup and a few minutes to deploy. A full, public testnet is a different story.
I have kept the testnet tutorials updated and even found some small omissions that have started to affect deploying a testnet on Hive lately. Whereas previously, resource credits did not impact testnet setups very much in the past, they've started to become more of an issue. Special thanks to 
As mentioned, in addition to the published full-sized testnet documentation, I've been maintaining docker and gist repos for deploying a toy-sized testnet, called tintoy. If you're looking for a testnet, tintoy is the quickest and most reliable way to get started.
Also, special thanks to 
Early on, it was tempting to just do a mass search-replace, looking for steem and switching it to hive. But doing this blindly was problematic. For example, it would change "steemconnect" to "hiveconnect" (which is nonsense) instead of hivesigner. Or references to "Steemit, Inc." would become "Hiveit, Inc."
Context matters.
Also early on, it didn't really matter that much, except as a point of pride. The initial fork of hive retained the steem identifiers, like sbd_interest_rate. Those identifiers changed in the eclipse hardfork, so it became imparative to the devportal.
Also, there are situations where it still makes sense to have "steem" in the devportal, for historical context (and to avoid the mandela-effect):
There's still some cleanup to do in this regard, but most of it is consentrated in libraries used by tutorials, which I'm currently (slowly) auditing.
SBDS was something of a predecessor to Hivemind, sort-of. Whereas Hivemind was focused on communities, SBDS was much broader.
The idea was to improve performance of API endpoints with database-agnostic look-ups. Most of those bottlenecks have been addressed by Hivemind instead.
Not only that, but search/replace would have presumably renamed it to "HBDS," and there is no such project.
Mostly just re-branding and verifying that it still works as documented. I also updated the "helpful" UML diagram with the proper terminology.
By far, these tutorials are the most time-consuming aspect to the entire devportal, and we have three languages to maintain. The latest commits involved the node.js changes. Simple search/replace would have just clobbered these.
I'm most proud of these paragraphs:
Developers love quick-starts. This is as quick as it gets. During the audit, I'm removing the previous "git" commands overview with a link to gitlab as a .zip file. It's a neat feature provided by gitlab that lets you drill into a repository and grab the exact part of the tree you're interested in.
I will be adding these paragraphs to the python and ruby tutorials, when I get around to them.
You can see all of the development commits to devportal here:
https://gitlab.syncad.com/hive/devportal/-/commits/develop
I very much enjoy working on this, so regardless of the level of funding I get, I will continue. But I'd also love to be able to prioritize this work beyond a hobby.
If you'd like to vote for my current proposal:
https://hivesigner.com/sign/update-proposal-votes?proposal_ids=[151]&approve=true
Also see: https://peakd.com/proposals/inertia
