We've Done It
After almost 2 years and a lot of hard work by many people, we have finally achieved what we set out to do. On Wednesday, the Linode servers were decommissioned, and the site is now completely independent and running on its own hardware. All the site data and the domains belong to the community - yes, you own this site.
It could not have been possible without the dedication of quite a few people - some of whom have not made it to today. It wasn't pleasant and it wasn't easy. There were numerous meetings to create a new set of Bylaws. That in itself took many months of often acrimonious debate. Everyone wanted the same end, but they were not happy with how it was being achieved. In the end, as is always the case, it required a lot of compromise, and it meant we needed to say goodbye to the old way of doing things. So to all of those who participated in those meetings: mechanicjay, AudioGuy, Deucalion, Fnord666, MartyB, Separatrix, cmn32480, kolie, hubie, mrpg, requerdanos, fab23, chromas, fliptop, dx3bydt3, tedious, cosurgi, Runaway1956, and even aristarchus, and many that I have probably missed out, a huge thank you from me. And if I have missed you out then that is entirely my fault and I apologise!
The Bylaws dictate how the business will be managed. They do not contain any rules about how the site will be used, or what is permitted in submissions or discussions and what is not, or any of those things that tend to interest the community. But there are some unique differences in our Bylaws from those previously. The Board (which I will get to in a moment) manage this site on your behalf. While they handle routine matters as appropriate to their post, many actions require them to seek the agreement of the community. You will have a say in major decisions. After all, you own the site. The data cannot be used for any purpose other than maintaining this site without a majority decision of the community.
So having created the Bylaws, we could apply to create a company. This was necessary so that the new company (Soylent Phoenix) could receive whatever assets the original company (SoylentNews PBC) was prepared to transfer. For legal reasons, the Board members cannot be anonymous; they have to sign various pieces of legal documentation to create the company. So I asked for volunteers from the community. More than we needed stepped forward and so the business was created. The Chairman of the Board is OregonJohn (6105), the Treasurer is Dale (539) (and he is actually a qualified accountant!), and the Secretary is holeinone (17639). They have also been working hard on your behalf setting up the company, creating Stripe accounts, managing the current funds, and filling in the paperwork that is associated with a business. Technically the Board is a temporary one because they do not yet have the approval of the community. But that is built into the Bylaws, and it allows us to keep moving forward until their posts are confirmed by the community, or challenged by someone else who believes they should be allowed to do the job.
Next came the transfer of assets. Like all things this did not go as smoothly as everyone had hoped it would. Quite understandably, it was necessary for everything to be accountable and legal, and to look after everyone's interests. Again, it was only by having some tough discussions and making compromises on both sides that anything has actually been achieved. So another thank-you to NCommander and Matt Angel. I think we have a solution which, while not meeting all of our initial dreams, is one that we can live with for the future. The site never squandered money and there is nothing to suggest that your donations have been misused.
You will probably now all be aware that two people have donated hardware and a data connection at no cost to the site. Those people are kolie and fliptop, and they have my sincere thanks. Kolie's server is the one that you are using at the moment to read this Meta. In fact, it is handling all of the site including email, IRC, DNS, Varnish, the database, etc. etc. Fliptop's server is currently an off-site backup for our data but will probably require a bit of expansion before it will cope with the full Docker installation. You will also be aware that, over the recent weeks and months, kolie has built the entire software infrastructure into Docker containers, moving each function one by one from the Linode servers and incrementally adding it to our own hardware. He has put many man hours into this task and I for one have enjoyed working with him - although my contribution has been considerably less than his! Again, I offer my grateful thanks.
Finally, none of this would have been possible without the community being tolerant and understanding as we have encountered each unexpected obstacle. The community has, in fact, grown slightly in the last 2 years, but it remains at roughly 10% of what it once was. I thank you all for making this possible. Welcome to your site.
And now for something completely different....
Community Representatives
To provide the community with more visibility regarding the actions of staff and the reason and justification for those actions, we have created 3 posts which will act as Community Representatives. Their prime role is to provide the community with the assurance that management is behaving responsibly and within the rules governing the site, and for the community to have a focus when they have concerns regarding what is going on behind the 'curtains'. They will act as your voice in discussions and decisions. They will have full access to staff discussions but will only have limited access, and then only when absolutely necessary, to the data stored in the database. We have had 3 volunteers for these posts for many months, but I will provide more details once I have confirmed that they are still volunteering for the role, and when we have hammered out the working procedures that must be adopted for this system to function correctly.
TOR
We now have TOR up and running again - we had it quite a few years ago and then it disappeared. Eventually, the link will be in the SoylentNews side panel on your screen, but for now it is cut and paste:
http://soylentqarvi3ikkzpp7fn4m5pxeeonbv6kr4akgkczqethjfhmalhid.onion
Elections
As I mentioned earlier, technically the current board is temporary. If you wish to stand for a post then you may do so. We will be asking for nominations in the near future. However, remember that you cannot remain anonymous as your name has to be recorded on the official business records.
We will provide you with a Meta page should you wish to explain to the community why you believe that they should vote for you. The method of voting will be 1 vote per active account, which must be in good standing (i.e. meet a minimum karma level and not be currently banned). The software that will be used is still under consideration as the original voting software was written and operated by audioguy who is not currently taking an active part in the site management.
You cannot nominate someone else for a post. They have to put themselves forward as a candidate. You can of course vote for them should they do so.
realDonaldTrump Account
Accounts that have been dormant for over 5 years quickly attract attention when they suddenly re-appear. In recent years we have seen attempts to resurrect old accounts, sometimes by people other than the original owner. There is an option built in to Rehash which is named "Force User to Verify Account" (FUVA) and it appears on the user's administration page. It simply resets the password and sends the new password to the email address that we hold for that account in the database. The account holder can then reset the password to something of their own choosing and everything works as it should. Currently, it is invoked automatically after the 5-year point if the account becomes active again. About a week ago the "realDonaldTrump" (rDT) account suddenly sprang back into life, having been dormant since 2019, and for the most part it was welcomed by the community. However, by making a comment the FUVA triggered.
I think that the account holder has either forgotten how to access his email or the email address has lapsed for some other reason. So the account is now stuck in limbo. Without a working email contact we cannot send the new password to the account holder. I do NOT want to know who the account owner is. However, they must provide a working email address, which can be a temporary or throw-away address, and they must provide sufficient detail about the account to prove that they are the true account owners. The site can be reactivated as a parody account which will have specific restrictions associated with it.
Community Involvement
Although the site is now independent and 'under new management' there is still much that remains to be done.
- For the present the rules that we have inherited will remain in force, but they will have to be rewritten and approved by the community to take us into the future.
- We will need people to assist with the sys-admin role - which is more Docker management than anything else. Interested people do not have to currently know or use Docker, I am sure that we can provide guidance on acquiring those skills and then explaining the current site structure. It is much simpler than it once was.
- We are investigating how to create a separate environment for both testing code before it is integrated into the main site and as a training facility.
- As ever, we need your support by providing submissions and by participating in the discussions. There are 23 submissions in the queue but 17 of them by one of our 2 submission bots which many people dislike.
Potential Return of Anonymous Cowards to the Full Site
Over the last week or two the level of spamming and abuse has significantly reduced. However, the reasons for this are unclear, and it might not indicate a permanent or even long-lasting change. If the low level continues for a significant period (certainly into the New Year) then it might be possible to re-admit ACs to the whole site again. This is something that is we all want to see, and it would be a welcome enhancement to the site. However, it depends entirely on those few individuals who account for the vast majority of the spam and abuse.
If ACs are allowed back onto the full site then any permanent return is entirely dependent on the spamming remaining at a very low level, and for personal attacks to be rare events.
Withdrawn - It didn't even last until we got this Meta out. JR
(Score: 5, Insightful) by Tork on Friday November 15, @05:22PM (3 children)
🏳️🌈 Proud Ally 🏳️🌈
(Score: 5, Informative) by janrinok on Friday November 15, @05:46PM (2 children)
kolie is already taking a short break! He has certainly earned it.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by mcgrew on Saturday November 16, @01:05AM
You all have. Please don't stop!
Our nation is in deep shit, but it's illegal to say that on TV.
(Score: 3, Insightful) by Reziac on Saturday November 16, @02:28AM
I'm sitting here full of happy astonishment. This is such good news, and many thanks to everyone involved.
And there is no Alkibiades to come back and save us from ourselves.
(Score: 5, Insightful) by turgid on Friday November 15, @05:30PM
Well done to all involved.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 5, Insightful) by Snospar on Friday November 15, @05:50PM
Well done to everyone involved, I know a huge amount of work has gone into this and I'm thrilled that this site and this community can forge ahead into the future. To infinity and beyond (or something)
Huge thanks to all the Soylent volunteers without whom this community (and this post) would not be possible.
(Score: 5, Insightful) by Barenflimski on Friday November 15, @06:01PM
Great job y'all! This is all good news. I'm impressed. Thanks for everything you do!
(Score: 4, Insightful) by jelizondo on Friday November 15, @06:24PM
Drinks for everyone!
Great way to start the weekend.
Cheers
(Score: 3, Interesting) by VLM on Friday November 15, @06:32PM (10 children)
Grats to all.
You seem very optimistic. All the "cool stuff" about docker needs a fair amount of knowledge at least to avoid the worst foot-guns.
For example say you're doing a K8S deployment of a docker image, you can include stuff like this in the deployment's YAML:
readinessProbe:
httpGet:
path: "/"
port: 80
periodSeconds: 10
startupProbe:
httpGet:
path: "/"
port: 80
periodSeconds: 60
failureThreshold: 10
What that will do is try to pull "/" from the container on port 80 every ten seconds and if it fails it gets kicked out of the load balancer by K8S. Also you get ten minutes on initial startup in case its a slow starter.
You can also use a liveness probe to forcibly restart a container. Like you can assume its locked up somehow if it hasn't generated a log entry in an hour.
You can do a whole lot more other than httpGet probes like run commands inside the container that better return "0" as a result code "or else", or gRPC calls or test if certain TCP ports are open. Its pretty cool.
Adding to the fun, "K8S old timers" like me know readinessProbe and livelinessProbe but startupProbe is new like just the last year or two (at most) and "in the old days" we did weird things in livelinessProbe to avoid accidentally marking a slowly starting up container as a dead container.
Everyone makes the mistake of imagePullPolicy: "Always" about one time. Its a moth-to-a-flame situation. Think about what happens at a large installation if there's something like a power failure or mass reboot for whatever reason "which should never happen" but does anyway. You won't like the results. Especially when anonymous dockerhub access goes into rate limiting. Very funny ha ha. It'll all come back up eventually. Eventually.
And then there's the people who don't get K8S secrets or even K8S configMaps so they put plain text passwords in deployment's "env" section very funny ha ha.
Setting resource limits in a K8S deployment is an entire art form also.
Oh here is an artform. Think of Deployment "revisionHistoryLimit" if its too high you'll run out of storage and if its too low you'll stick yourself in a corner in the inevitable emergency roll-back scenario. 2 is probably pretty dumb unless there's a good reason, 200 is probably WAY too high. More than you're willing to scroll thru in a CLI during a disaster is too many.
Another funny fun one is you can trivially deploy stuff using Ansible into K8S but it can get ... exciting if you are not very careful. Uh I was staging something I didn't know it would restart the prod database server whoopsie.
Also people who are "not familiar with Docker" like to do things like avoid editing your own DockerFile instead doing configs in shell scripts provided by configMaps or even crazier stuff. Generally a bad idea.
What else is hilarious about Docker / K8S / etc... how about people who don't understand their load balancer, so if they use metallb but don't know what a allow-shared-ip is they end up in a hell of their own making, generally very confused about what happened, that's a typical noob networking mistake on K8S. Or the people who don't understand "port" vs "targetPort" in a K8S service.
Its not impossible, every noob started at the start, but it'll be exciting. Another piece of "noob K8S advice" is you have to learn very quickly and early the difference between real facts and opinions. Some people think putting /something/.ssh/authorized_keys into a ConfigMap is sheer genius, others think it sheer insanity. Sometimes both are correct, sometimes it depends on your local AAA design. K8S labels are a typical example, this is like variable naming standards except its worse in K8S, much more confrontational. Oh another one, people who like/hate StatefulSets and Deployments. You'll run into people who will run Deployments with "replicas: 1" and insist everyone else is wrong. Its interesting to reminisce about all the dumb things I did while learning K8S, mostly documented or alluded to above. I have never "made the news" by doing something dumb, AFAIK, so at least I've been lucky so far.
I wouldn't put your K8S Secret yaml files in public on github LOL but its probably safe to upload deps and svcs and stuff; you might get some free advice about those. Not just from me shitposting, I'm sure there are other people here who do this occasionally at the day job (or contract for me). Or maybe you're using terraform to manage K8S I donno.
(Score: 5, Touché) by janrinok on Friday November 15, @06:52PM (4 children)
Fascinating - but we are not using K8S.
The cool stuff is written in English, not a different arcane language for each function that is so easy to cock up.
I'll cross your name off the list of potential volunteers then? :D
Enjoy your weekend!
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 3, Interesting) by VLM on Friday November 15, @07:15PM
Oh. Well then. It's a lot of fun and automates a lot of stuff you'd otherwise be doing by hand. You're doing K8S "stuff" anyway LOL, you're just doing it all by hand LOL.
Back in "the old days" I ran some docker containers by obtaining a "secure" NFS mount on all my hosts running docker and then in /docker/something I had shell scripts like "start.sh" "stop.sh" "upgrade.sh" that did about what you'd think. You do a volume mount on the docker command line like -v /docker/something/config:/config would map /config on the container onto my NFS share. That way I could in theory run any container on any host which I certainly did sometimes. I stored all my configs and config files in git making rollbacks easy. Not as easy as K8S LOL. IIRC the main pain point was mapping DNS to the IP address of the host that's currently running that container and forget about trying to get load balancing working. Oh it was also hilarious if I tried to start a container twice simultaneously on two different hosts. A nifty feature of this is I only need to backup my /docker NFS share because everything else was generic bare Ubuntu. The above is a good small timer way to run Docker containers.
Or you could be using something like docker compose; I kind of skipped that tech going right from the stuff above to K8S. It has its own "language" to learn, looks interesting.
(Score: 2) by vux984 on Saturday November 16, @12:26AM (2 children)
docker but not k8s? swarm?? Something else? I'm curious why not k8s?
I've been using k8s for a few years now for our core services, and feel I've barely scratched the surface. VLMs post brought some pretty cool features to light that i didn't know about. (Not secrets and configmaps, we've been using those for ages, but some of the health monitoring stuff I didn't know about - that's been on my todo list for a while... right after we improve logging. :p .) At least health monitoring hasn't really been an issue - the containerized services are pretty robust - crashes/restarts are pretty rare, and "hangs" non-existent.
(Score: 3, Informative) by kolie on Saturday November 16, @03:35AM (1 child)
I like k8s. K8s is a lot of complexity here, and additional texting and overhead. We are running on one node right now. Realistically on k8s your going to need 4 nodes to even get off the ground. Maybe in time. I can show others the ropes on docker, finding people up to managing k8s is going to be a challenge.
(Score: 2, Interesting) by RobC207 on Saturday November 16, @09:44PM
Nice work, kolie, on completing the migration into Docker containers! This is the most difficult part of the journey. Having to resurrect a reproducible build from the bones of an existing infrastructure is a fine art form.
Buuut, now that you have done it; you have done most of the hard work that you'd need to attain scalability and fault tolerance. Those are the features that you really need in order to survive unexpected hardware failures or (gasp) software maintenance events. I mean, look at the recent stream of kernel releases. You're going to need a better long term plan than reboot and pray. What's the point of having high availability at the host level if your services themselves are not configured for fault tolerance?
I have to agree with some of the other commenters on this thread. Docker is cool; docker-compose is the next step, maybe, but Kubernetes is so much better. The best benefit for a small cluster is the easy host maintenance. Even in the event of a full power failure, it's generally hands free recovery to a working state. Of course, I am already standing on one of the plateaus, having slogged the vertical learning curve. I have been hosting in Production for many years. More specifically, Rancher on k3s. Rancher provides an excellent user interface, as well as an easy way to provide user authentication control. You don't need to know k8s or yaml in order to get your services to run. You can do it all from the web interface. Once you are happy with the state, you can download the yaml from the cluster, clean it up, and store it in your deployment repository. It's also very easy to host multiple copies of the site infrastructure on the same hardware. The nodes don't need to be huge. You save money with more, smaller nodes. My home cluster runs on four Pi 4's and one x64 node.
I would like to donate time in the sys-admin role, and to help you make the leap to k8s. I help software support teams migrate from bare VMs to Docker and then onto Kube in my day job for fun and excitement. I couldn't stand the stress of admin on a single Docker host.
Anyway, let me know.
(Score: 2) by sbgen on Friday November 15, @09:18PM (4 children)
Sigh... Are you trying to discourage me from volunteering? I am not from tech background and was hoping to start learning by contributing. Here you are - smashing my dreams.
Warning: Not a computer expert, but got to use it. Yes, my kind does exist.
(Score: 4, Informative) by janrinok on Friday November 15, @09:51PM (2 children)
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by sbgen on Saturday November 16, @04:28AM (1 child)
This ought to be fun :-) You said to drop a mail and I cant figure it out - that should tell you my vast tech expertise. I do have expertise but in a different discipline and I am sure I have some skills there.
Thanks again to all of you for what you have been doing for this community.
Warning: Not a computer expert, but got to use it. Yes, my kind does exist.
(Score: 2) by janrinok on Saturday November 16, @06:30AM
janrinok@soylentnews.org
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 3, Informative) by VLM on Friday November 15, @10:16PM
If anything I'd encourage you to play around with K8S at home, although apparently they're not using K8S here so it doesn't matter.
It's a barrel of fun and you can do all kinds of crazy stuff.
There are two older books that still check out, one is "Docker in a Month of Lunches" and the other is "Kubernetes in a Month of Lunches".
I'd suggest starting with docker and eventually you'll find yourself writing shell scripts and procedures to automate all of that and then you'll notice you can replace an enormous amount of effort with K8S... I use RKE2 at home. Work is a complicated story revolving around cloudy things.
"GitOps and K8s" that book is too high level for someone starting out.
Oddly enough CKAD study guide type of books are a pretty good broad based intro to K8S.
"Kubernetes for developers" not a bad book choice if you're a developer.
I would start small. The stereotype is use Docker for your web frontend and backend with multiple API and web servers and database clusters and stuff. Naaaah start with getting a dokuwiki appliance running. Or a Linuxservers.io "webtop" running.
I would say that perhaps K8S is the most vertical learning curve I've gone up in a long time.
WRT containerization and stuff, doing easy things is easy but proper architecture and security is harder than non-containerized stuff.
(Score: 3, Interesting) by VLM on Friday November 15, @07:04PM (4 children)
Wanna see something cool?
In your ansible:
roles:
- { role: deploysomething, k8snamespace: 'prod' }
Or namespace "dev" or "test" or whatever. You can run this multiple times with different k8snamespace variable names that do about what it sounds like.
Then in your deploysomething.yml role file you will have a variable named k8snamespace containing 'prod' in this example. You only have to write ONE role ONE time, not one for prod and one for dev and one for test and one for betatest and ...
IIRC you can template your role so in Ansible when you run kubernetes.core.k8s it looks like this
- name: deploy on the K8S cluster
kubernetes.core.k8s:
namespace: '{{ k8namespace }}'
src: deploysomething.yaml
IIRC you don't "need" to set a metadata namespace in your deploysomething.yaml (or maybe I used a template to insert it, I don't recall this is all from memory). People THINK you have to include a namespace line in your K8S yaml file but it'll inherit from kubernetes.core.k8s. I think. I would need to log in and check. If that's not how I did it I used templating, which you'll have to use anyway.
If the above insanity is unclear, your Ansible main.yml is calling an Ansible role deploysomething.yml with a variable injected to tell it if you're using dev or prod or something else as your namespace name. Then in deploysomething.yml you tell kubernetes.core.k8s to use your injected variable as its K8S namespace.
This means prod and dev will be exactly the same so you will be doing at least some templating in the config files to make sure the web frontends talk to db-prod or db-dev or db-{{ k8namespace }} in general. I would have to log into Ansible and review how I did this but its was probably similar to using ansible.builtin.template with the template_vars passing it a dictionary based upon the simple string variable and then having kubernetes.core.k8s call the result of ansible.builtin.template or something like that. It probably made more sense at the time I wrote it. Had to do something similar with DNS templating, such that something-dev or something-prod would be provisioned. Long story.
Ansible can do very cool stuff but it can degenerate into looking like line noise if you're not careful.
(Score: 3, Informative) by janrinok on Friday November 15, @08:21PM
Thanks, and noted.
The first priority was to get it transferred and working. Kolie has achieved that. We accept that there is still outstanding work to do to produce the scripts that automate the entire build but hopefully there is a little less pressure now and the pace can ease somewhat. We do not underestimate the remaining work.
I think many in our community think that we still have the number of staff that we had 2 years ago. I'm afraid it isn't so. We can almost count the active staff on the fingers of one foot!
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by Geotti on Sunday November 17, @02:25AM (2 children)
Oof, if you think kube is vertical - try nix. Loads of fun there with reproducible environments.
Other than that @janrinok, definitely use some orchestrator for the containers! k3s is a good start, with not too many moving variables, but it's not too hard to wrap your head around kube, once you realize what you need to manage something running off of containers.
(Score: 2) by VLM on Sunday November 17, @05:20PM (1 child)
I think I agree completely, although I'd phrase it differently as my path was once you start automating your container operations using shell scripts and written procedures... it all starts looking an awful lot like "write your own K8S" except with far fewer features and more bugs.
There are exciting corner cases and situations to look for as I previously listed and the learning curve is a bit steep. Imagine learning someone ELSES docker automation bash scripts, and they're 100x more complicated and capable than your own, it's a rough weekend or so of effort.
I would recommend that after someone writes a couple homemade shell scripts to automate Docker stuff, they'll understand K8S a lot better than someone who just started playing with K8S. On the other hand that path will take longer.
(Score: 2) by Geotti on Thursday November 21, @02:35AM
Yep, also Kelsey Hightower's Kubernetes the Hard Way comes to mind for the ones, who want to truly understand setting up a k8s cluster.
(Score: 5, Interesting) by Gaaark on Friday November 15, @07:18PM (5 children)
Thank you kolie and NCommander:
my fear was this would turn into some kind of Dresden fire bombing. You put positivity back into my life.
Thank you again for bringing this site back into reality and into something worthy of being a part of.
Thank you to all those involved, the most visible being Kolie and NCommander and Janrinok and all those others i'm too old to place. Thank you to all. All.
All.
I will be donating now.
Spank you to all.
All!
Heil SoylentNews.org!
--- Please remind me if I haven't been civil to you: I'm channeling MDC. ---Gaaark 2.0 ---
(Score: 3, Informative) by Gaaark on Friday November 15, @07:23PM (2 children)
Or...not?
I encountered an error when subscribing.
will try again.
Okay... have either subscribed twice or not at all.
--- Please remind me if I haven't been civil to you: I'm channeling MDC. ---Gaaark 2.0 ---
(Score: 2) by Gaaark on Friday November 15, @07:27PM (1 child)
emailed the admin at admin@soylentnews.org
probably some config problem.... grrr... :)
--- Please remind me if I haven't been civil to you: I'm channeling MDC. ---Gaaark 2.0 ---
(Score: 4, Funny) by janrinok on Friday November 15, @07:54PM
I've replied - hic.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 5, Informative) by janrinok on Friday November 15, @08:01PM (1 child)
NCommander donated everything that the original site had at no cost. I am not sure, but I think that also means that they wrote off their initial investment. Thanks indeed to NCommander and Matt Angel.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by Gaaark on Friday November 15, @09:34PM
So we have TWO Angels! :)
--- Please remind me if I haven't been civil to you: I'm channeling MDC. ---Gaaark 2.0 ---
(Score: 2, Insightful) by pTamok on Friday November 15, @07:59PM
Thank you to all who contributed.
Looking forward to making a monetary contribution.
I wonder if 'The Register' will report this - we've appeared in their reporting before.
(Score: 0) by Anonymous Coward on Friday November 15, @08:06PM (2 children)
is to not be dependent on yesterdays slashdot
(Score: 4, Informative) by janrinok on Friday November 15, @08:16PM
We do not use anything from Slashdot. If the stories aren't topical then that is not our fault. We are not responsible for making the submissions. The editors' job is to process them, which we are doing every day. The bots are only used when we are running low.
Weekends are a little bit different. We like to get the queues set up early so that we too can enjoy the weekend. I am half way through preparing Sunday's output, but it will have to wait now until tomorrow. My day has ended having spent over 5 hours on here today, as well as living my own life.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 3, Informative) by hubie on Saturday November 16, @03:02PM
I made a clean break from Slashdot 10+ years ago, coming over here with the initial migration. I don't go there at all, not so much out of principle, but I regularly frequent very few places these days. I lost the ability to log in over there many years ago when I forgot my password and my Slashdot account is tied to an email that no longer exists.
All of that is to say, I don't know what is posted over there and when, nor do I really care because it has never been the mission of this site to post things that aren't on Slashdot. I assume they release stories more frequently than we do here, so there's that as well. Plus, one could just as well complain that Slashdot is yesterday's Hacker News, I suppose.
My primary concerns when submitting or editing stories is to check whether they are dupes already posted and whether I think they are generally interesting to readers of this site or would generate interesting discussion. Given that this site split off of Slashdot, it isn't too surprising that the same kind of stories would show up in both places.
This is a good opportunity to point out that the frequency, and thus timeliness, of stories here depends upon the number of story submissions and the number of active editors. They need to grow together, and if they do, we can post new stories more frequently.
(Score: 3, Insightful) by Deep Blue on Friday November 15, @08:19PM
A long path travelled.
(Score: 2) by Rosco P. Coltrane on Friday November 15, @09:21PM (4 children)
Great job guys. I love it when one of my favorite sites does well.
Now then...
Where's my title?
(Score: 2) by Gaaark on Friday November 15, @09:37PM (1 child)
How many shares do I get? ;)
--- Please remind me if I haven't been civil to you: I'm channeling MDC. ---Gaaark 2.0 ---
(Score: 4, Informative) by janrinok on Friday November 15, @10:25PM
I know you ask the question in jest - but that is a significant difference with the new site. There are no shareholders. Every account has a theoretical 'share' in the site (in reality it is a vote) and thus any significant decisions that would affect the site should be agreed by the community before any action is taken. For instance. the routine running costs can be handled by the treasurer but any significant expenditure after we have finished setting the site up has to be agreed with the community. So no hookers and blow, for example, and no hidden payments to keep Martyb in his luxury New York penthouse suite,
There have been several donations of cash, hardware and services - some of which are anonymous gifts, and others I have described in the Meta above. We are not starting with a debt, nor do benefactors expect to be repaid. When everything is fully organised the accounts will be made available for everyone to see as a display on the site. They will be 'sanitized' so that no personal information will be revealed. If all goes as expected the site running costs will be significantly less than the previous site but there will still be some essential expenditure to meet legal requirements, to maintain the domains, etc.
As well as building the site, we have to start rebuilding the staff so that nobody is required to work more than they would want to, and rebuilding the community. That last item will not be an easy task but it is essential if the site is to have a lasting future.
It is important to note also who is NOT on the Board. Neither kolie nor I have a place on the Board. This has not be a power-grab to take control of the site. We asked for volunteers to form a Board not knowing who might step forward, or indeed if anyone would. Fortunately, people did.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 5, Informative) by janrinok on Friday November 15, @09:46PM (1 child)
You are Senior Vice Rosco P Coltrane. When you visit the secret SoylentNews HQ, you can use the indoor toilet to pee rather than going out the back behind the dumpster....
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by Tork on Friday November 15, @11:40PM
🏳️🌈 Proud Ally 🏳️🌈
(Score: 4, Funny) by sbgen on Friday November 15, @09:26PM (1 child)
So good to read this submission, and thanks to all of you who worked actively for so long. Hurray to all!
It sure took a while, two years? As I age I have come to realize everything takes longer than what is in my mind, and I worry a little less every time that happens :-) Also, I read NCommander donated everything the original site had at no cost. Shows that in the end everyone wanted the same thing - a sane community to continue.
Now let me take my glasses off and wipe a tear ...
.
.
cant find one but I wiped anyway.
Warning: Not a computer expert, but got to use it. Yes, my kind does exist.
(Score: 2) by fliptop on Saturday November 16, @07:54PM
It's a loose collection of people scattered all over the globe using email and some IRC to communicate. There was a point in time last winter when it seemed like we were in 4th gear while chain-deep in mud but after I applied for and received the 501(c)(3) it moved forward w/ the by-laws, getting the initial board together, setting up the servers (thanks kolie!) and transferring the site.
Our Constitution was made only for a moral and religious people. It is wholly inadequate to the government of any other.
(Score: 5, Funny) by carguy on Friday November 15, @10:06PM
An invitation for all to join me in a hearty round of:
BUCK FETA!
Congrats and thanks to everyone that contributed, and continues to contribute.
(Score: 3, Interesting) by ShovelOperator1 on Saturday November 16, @02:31AM
Thanks a lot for keeping this site running.
About Tor version, instead of links, there are some HTML options for it - then Tor browser will automatically ask should it switch to Tor version. However, it looks like it's different again - when I checked it previously, it was just a Meta tag. Now it seems to be some kind of header too.
(Score: 2) by The Vocal Minority on Saturday November 16, @05:32AM (3 children)
My infinite thanks to all involved.
When was the peak number of users that we are now only 10% of?
(Score: 3, Informative) by janrinok on Saturday November 16, @07:29AM (2 children)
TLDR: about 350.
Accurate figures are hard to pin down as nobody can tell how many lurkers we have.
So we look at active accounts that can be identified individually. We had around 5 years ago approximately 3,500 regularly active accounts. They weren't all active at once but over a period of several months they were visible in some way or other. They logged in, commented, moderated, or made a submission. Something that is recorded in the database.
Over the last few months I haven't managed to do a similar analysis, but we have stats that are provided daily and they suggest that we have around 350 active accounts. That varies of course - there are noticeable daily changes in activity with weekends attracting some who do not appear during the week, and vice versa. Additionally, all ACs are recorded as the same user ID - "1". So whereas humans can often identify different ACs from each other, the database and software in Rehash does not. Someone with the appropriate access can run their own software if they wish to refine the data that Rehash records.
A second way of guestimating the level of activity is to look purely at page hits. These have also improved slightly over the last 2 years. However, it is also necessary to remember that we do not know how many bots or crawlers are accessing the site.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 3, Informative) by janrinok on Saturday November 16, @07:32AM
That should read
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by dilbert on Saturday November 23, @05:40AM
> Accurate figures are hard to pin down as nobody can tell how many lurkers we have.
Thanks for all your hard work from this lurker who reads regularly, but doesn't typically post.
I am willing to help out on the systems side if my linux admin skills are needed (you probably have enough volunteers from more active posters)
(Score: 3, Interesting) by acid andy on Saturday November 16, @04:31PM (2 children)
This is great news and a seriously impressive achievement. Well done and thank you to all who had the determination to keep working towards this.
In the text you strongly imply but do not actually directly state: Is the handover from NCommander and Matt Angel now fully complete and legally signed off? I am not trying to stir things up, just observing from a distance I have seen these things take a lot of time and obviously you have the servers and the domain now, but I was just curious if there were any other loose ends at this point?
Welcome to Edgeways. Words should apply in advance as spaces are highly limite—
(Score: 4, Informative) by janrinok on Saturday November 16, @05:24PM (1 child)
As far as I am aware yes. But I will have to refer to the Board for confirmation. I'm trying not to interfere with other people's roles. There is nothing worse than volunteering for a job and then somebody telling you how to do it. If anyone needs help or advice then we all chip in, but that is not the same as butting in when not invited. :)
We certainly have ownership of all of the things that we asked for and, as I have stated elsewhere, NCommander was happy to donate them to the new site at no cost.
NCommander has not been having an easy time recently away from SN matters, so there may be things outstanding that just require a signature. He has been invited to release a Meta if he wishes to do so.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by acid andy on Saturday November 16, @05:56PM
Thanks for the swift and detailed response JR.
Welcome to Edgeways. Words should apply in advance as spaces are highly limite—
(Score: 3, Funny) by Username on Saturday November 16, @05:29PM (1 child)
Dearest soylentists,
I will nominate myself for the post of Director of the Department of Governing Efficiency (DOGE). My sole requirement is the on demand use of the company jet (when the company acquires one).
Sincerely yours,
Username
(Score: 3, Funny) by fliptop on Saturday November 16, @08:00PM
You'll have to confirm w/ the board, but I do believe it's nearing completion [etsystatic.com].
Our Constitution was made only for a moral and religious people. It is wholly inadequate to the government of any other.
(Score: 2) by NotSanguine on Monday November 18, @08:00AM
Your efforts are greatly appreciated!
No, no, you're not thinking; you're just being logical. --Niels Bohr