SoylentNews

SoylentNews PBC had a proper business meeting on Friday, to discuss events since the shutdown notice was posted.

This meeting was attended by myself, Matt Angel, and kolie. I was on the phone for about two hours, combined with multiple follow ups in DMs.

Let's get the good news out first.

SN PBC has agreed to continue operations for SoylentNews.

We also had a very long extended discussion on what the future may look like and some points brought up by staff or members of the community were discussed.

Let's cover all the major points below.

We talked about the state of things to come for over an hour, and then I had one on one calls with Matt and kolie. The conversation was fairly high-level, and mostly consisted of a recap of the last few weeks, current progress, as well as what actually rebuilding the site is going to look like. Right now, we have a commitment to finishing the infrastructure overhaul, and upgrade it as is practical.

That would basically bring us to the "ok for now" status quo ...

However, the status quo does not address the dwindling signal to noise ratios or shrinking community. It does not address the lack of moderation or content standards. It does not address our problems relating to SEO or anything resembling modern human interface guidelines. It also does not resolve the long standing problems that lead to here.

We also had a fairly long conversation on the circumstances that lead to here and how we avoid a repeat of ending back on the brink.

We still have the question of what we will do in the immediate future.

I have no illusion that staff are happy to work with me, but I am hoping we can at least define some sort of formal truce in the name of the future of the site. However, members of the staff have been demanding for me to step down and simply get out.

I have been given the impression that if I was to step down, they would simply continue the site as is. I am unaware of any defined plan relating to fixing any of the problems plaguing the site, both as it is now, and those that have persisted long before this point.

I have also been given the impression that if I do not simply GTFO soon, they're simply going to quit.

That's fair and understandable given everything that has happened.

However, I am not the only stakeholder, as currently, Matt as co-owner, and kolie, as an interested outside party who is actively helping to fix the site, have both indicated that they want me to stay involved at this time.

Ultimately, dealing with the deferred maintenance is going to take priority. Given everything, I don't see how I can have a revised plan for SoylentNews any earlier than July, and realistically I expect it to take longer to have a solid agreement hashed out by all stakeholders.

The staff should be a part of this discussion. However, at this point, they have at this point only made demands, and have made no attempts that I am aware of to negotiate anything. Unilateral demands is not a negotiation.

This is part of a larger problem that, at the end of the day, the staff had no true stake if the site succeeded or failed. This is compounded by the fact that no one has been willing to put themselves forward to join as a member of the board of the directors since mrcoolbp disappeared, and since none are apparently willing to work with me, the person who is both president and 50% owner, well, it leaves us at an impasse.

The easiest thing to do is to allow them to present their case as to why I shouldn't be involved and show to everyone that they can lay out a realistic plan to deal with the issues that have long plagued this website. After all, if they're so insistent that I shouldn't be involved, then as a matter of due diligence, I need them to show a plan that involves fixing the site, and actual work being doing towards it.

For my part, I will do the same in the form of a new business plan for the PBC that will specifically explain why we are taking money in, how we will use it, how we are going to deal with raising capital in the future, and everything else that is involved in keeping SN going for another decade. Such a plan is going to require both kolie and I to discuss the specifics of what replacing rehash is going to require.

At the end of the day, regardless of who was responsible, SN decayed to the point that the database was suffering from corrupted tables. That has to be addressed, and ultimately, until some else steps up to the plate, and presents a workable plan, that falls to me

Infrastructure Rebuild

Of course, talk is cheap, so here's what I have actually done in the last two weeks with help from kolie.

We have been making slow but steady progress on this. The plan is to convert the entire site to an ansible playbook, which now exists. We have successfully started deploying site services on a fresh set of Linode accounts. So far, kolie has got the public wiki rendering, bringing it from MediaWiki 1.18 to a currently supported version.

Meanwhile, I've been digging deeper into rehash. In any scenario that involves the site continuing, we are going to need to be able to deploy code changes. Fortunately, when I did the work to get the site running on Apache 2, I left myself a lot of good notes on how it all works, as well as the "make build-production-environment" target which handles a lot of the worst parts of how to deal with the mountain of legacy Perl.

This is slowly coming together in building a Dockerfile that's on the public rehash repo.

As of writing, I have gotten it to the point it can successfully run the install target. I wrote some rather hacky code that handles shoving connection database information into the system DBIx::Password module, because that's what rehash requires, and I was reminded again why this is the codebase from hell.

We have also had a longer discussion towards implementing new services like status monitoring. A member of the community submitted a long and detailed plan to implement Prometheus for service monitoring. We're not quite ready for that, but we also need to talk about specifically access to the backend.

Infrastructure Access

When I formally announced the shutdown for SoylentNews PBC, I locked out all shell accounts to the backend as well as limited access to the Linode panel. This was done both to protect the site and as a matter of liability. When the circumstances changed, access to the backend was not restored. Access to rehash's administration panel however remained available, which is how janirirok and the other editors have been posting articles. The #chillax channel remains up, although I'm no longer in it.

I didn't restore access after the situation changed. There are quite a few reasons I could give, but given the sheer amount of hostility I have received from certain individual members of the staff, I could not and cannot rule out the possibility that someone would simply "rm -rf /" the production environment out of sheer spite. Also, these machines are going away. There isn't going to be shell access beyond this point aside from a control node.

Right now, the current plan is to simply get the site back to the level of functionality it had back in November, with as many parts of infrastructure being fully up to date as-is possible. This includes going through the configs, removing obsolete bits, and basically reviewing every aspect of the underlying nuts and bolts.

I am going to be taking a solid look at getting rehash ported to the current versions of Apache 2.4 and mod_perl. It mostly depends how much of the stack can be built on mod_perl 2.4 easily. At least with the site on deployable infrastructure, it drastically simplifies what it will take to move forward.

Timeline Moving Forward

Fixing SN is going to require people to be involved and dedicated to rebuilding and essentially relaunching the site as well as fixing many of the problems that have led to here.

To ask that without providing some sort of compensation is folly but without a defined plan, well, we end up in a catch-22 situation.

So, here's what I'm going to do and what I am going to ask of anyone who stays involved.

We will have the site migrated to less broken infrastructure by the end of June at the latest, and likely well before that point.

After that migration is complete, kolie and I are going to negotiate a contract that will handle either overhauling or outright replacement of rehash. This agreement is going to define any new functionality that will be built into what will essentially be version 2 of this website.

We also need to define what specifically is the role of various volunteers in the upkeep of this site, as well as defining our options in case we ever end in another situation like this five or ten years down the road.

If there's one thing I have surmised, no one is happy with this situation, but a big part of moving forward is having an agreement on how it will be maintained and have some flexibility going forward.

A major point of this is growing SoylentNews's cash flow to the point that it can reasonably afford to have at least one paid staff member and provide some actual incentives to keep people involved in its upkeep and overall maintenance.

I have quite a few ideas on how to approach this, but I rather have a chance to write them out in-depth and explore them.

As part of this, I also want to deal with raising the overall signal to noise ratio, increasing the number of comments per article, and hopefully bringing in new blood to the site. It's very hard to have any idea as of the state of the health of the community as we don't have any analytics, and have historically only ever run PiWik for a short period of time.

I think we need to consider doing that again, if only to have an idea of where we are in terms of actual readership and engagement.

Finally, there needs to be a reason why someone might post inbound links to SoylentNews, and once someone clicks on them, sticks around longer than a few seconds. That also makes finding such content easier for both humans and search engines.

These are just some ideas off the top of my head. Realistically, I'm going to need to sit with a pen and paper, sound out which ones are practical with kolie and Matt, and then start taking steps to implement them, with whomever is willing to join us going forward.

I will keep you all posted,

~ NCommander