Social Media Immune System

For years, we have all been exploring the new frontiers of social media, sites like Facebook, Twitter, and the Fediverse and I’ve been thinking about a way to improve life on every social platform.

I like to think of it as an immune system for social media. Curated and shared blocklists. It should be rather easy, just a checkmark box on the other person you follow, or maybe they encapsulate their blocklists, like how it is done in Mastodon on the Fediverse. Click on a user, then click on their shared blocklists, pick which ones you want to follow and then poof, all the users on that blocklist are simply gone.

Blocklists can be for anything. Name them anything, trolls, twats, asshats, or even foreign state actors. Before this, I think folks were very hesitant to even think about this because it will lead to social media balkanization. We’re beyond that point now, giving regular folk the benefit of the doubt simply isn’t going to be possible anymore. We need new ways to banish and shun, as shame has died in our world. This is the natural response to shame going extinct. Nobody is ashamed anymore, and as such, they are wretches that run amok, and so, we need new leashes to yoke them and drive them out of our social worlds.

Just imagine how lovely life could be, without Russian trolls trying to pitch misinformation to sow discord and division amongst us, amongst everyone. This sharable blocklist would effectively push them all out of open windows to the pavement below.

Obviously the walled-garden platforms really don’t want this, because it would actively subvert their algorithm which is designed to be as provocative and enraging as possible, so we really can only look to the Fediverse for innovation like this. In the meantime, one way you could do this and help-yourself is to always write a post about who you blocked and why. It’s not as immediate or convenient, but the time is right for us to stop communicating with each other, because there is absolutely no point in mixing or even acknowledging them. Drive them out, ignore them, and leave them in the dark. It’s the only way to save ourselves. Full throated balkanization.

Testing XMLRPC Hacks

Testing XMLRPC Hacks

Since the Gutenberg editor is such a slob, we can try Ulysses out. As expected, Ulysses doesn’t have any of the issues that the online editor has, no matter how quickly I type into the keyboard.

The real trick will be to see if the xmlrpc hack that I had to perpetrate to get Ulysses WordPress part to function still works. I bet it doesn’t, and so the new workflow to WordPress will likely just be copypasta, as usual.

Lets get ready for disappointment! Huzzah!

Oh Winlogon, Where Are Thou?

I logged into my “before to this point” trustworthy Windows 2012 R2 Server that I have nicknamed Sierra, it told me that I didn’t have rights to the D: drive as the domain administrator. Okay, so I can fix that by getting to the console, brought it up on TeamViewer, and it was a featureless black box. Nothing to connect to, nothing to command, “Send Control-Alt-Delete” did absolutely nothing.

So next stop, plug into the actual VGA console on the server and plug in a USB keyboard and mouse. I verified that the keyboard was alive, it toggled Caps Lock and Num Lock properly, tried Control-Alt-Delete, Control-Alt-Backspace, and Control-Alt-Esc. Nothing. Featureless. Except the local console was a dark blue screen and the monitor was not in sleep mode. It was registering a video signal, nothing but a blue screen. Heh, not a BSOD, that would have been something ROTFL.

I tried to connect to the file shares on the server, that wasn’t a problem, so I knew the server was at least alive. The front panel didn’t show any alerts, so the CPU, RAM, and Array were also just fine. The only problem was, no ability to logon to Windows!

I was able to remotely connect to Event Viewer from the Primary Domain Controller, which helped. There was an error, Winlogon recorded an error event type 6000, with the error: “The winlogon notification subscriber <SessionEnv> was unavailable to handle a notification event.” and then that started a Google search for ways to correct it. Every response was the same, reboot. I really can’t do that, the server has thousands of files open, there has to be another way.

I then connected remotely to services.msc to the troubled server. Nothing there looked promising, no references to Winlogon or any of that. Then it occurred to me that Sysinternals tools might be useful. I ran pslist \\ip-of-server and scanned the output. I spotted winlogon running, noted its PID, and then tried pskill \\ip-of-server winlogon to no positive effect, but I had the PID, so I tried that too. The moment I issued the command, Windows restarted winlogon, I peeked around the corner at the server console and there it was, the time and the entreaty to press Control-Alt-Delete. I don’t know what caused winlogon to crap out on me, but at least the fix was easy. I got logged into the shell on the server, and it is running idle, nice and normal.

So if you have a server like I do, and end up with a mystery blue screen and no way to login, look into downloading the pstools kit from Sysinternals. It saved my day!

Kludgey

This post was written on Mastodon so it has an informal writing style with jargon and a manner for a specific audience. All the spelling and grammatical errors are intended.


I love creating my own problems and then finding a rabbit hole and chasing it until I have a geek solution that is likely foolish. If I load too many tabs on my Macbook, it gets sluggish. So, can I start xQuartz? Sure! Update however…

Okay, that done, can I SSH with the -X flag to my little Raspberry Pi? Yes! Sluggish. Wah.

How about to my “Security” laptop, running Linux? Yes. Sluggish still.

Google Search, find x2go, install it. MUCH BETTER.

So I’m using x2go, running Firefox-esr and connected to my not-work-tabs, including this one. Not seamless, but it works acceptably well enough.

Sitting here, marveling at all this exceptionally complicated computing technology before me, everything has “multiple cores” yet you really couldn’t tell. So instead of running everything from one single computer, we’ve got serious work stuff on one, then a remote desktop window to another running “fluffy stuff”, and then playing Spotify from my !@#$ iPhone. HAHAHAHAHAHA.

Shitty apps, each written by devs that believe that their app is the “King Of The Realm” and you can malloc() forever without having to fret over anything at all. Leaks? Who cares! Look how pretty it is! So, multiple computers, multiple OSes, failures aren’t less, but they are spread out so they don’t block real work quite as badly.

Of course, there is also I/O Blocking to contend with. When the filesystem is doing anything, everything stops. Because I/O is super smexy.

So we contend with shitty development choices by simply throwing entire chunks of technology at the problem. Two laptops, a Raspberry Pi, an iPhone, and an iPad. Each device is good at individual things, but no, we can’t do everything on one single device. Watch that device just chug right to the fucking ground. Ah well. The modern response is “throw a hypervisor on it” and that, wow, what a great way to make an even bigger mess of things.

Bullshit hypervisors make for hilarious blown-out-afternoons. So, Windows 10 on an HP Elitebook laptop, install Hyper-V from the OS, and the Radeon display driver commits hairy suicide. Not only does the driver break, but it cannot be “upgraded” or “fixed”, the only thing you can do, is remove HyperV and… poof, uh, there was a problem? No! No problem! So, you shrug and chuckle and look at the icon for VirtualBox. Yeah, hey buddy…

There are some situations where I start thinking that I should buy a cheap $200 Chromebook just for some things. More technology. SMH. Of course.

Two days ago I remembered the glory-promise of X-Windows and SSH tunnels, with the Display being sent elsewhere. Oh my god, the promise of that… so glittering. So… disappointing.

Oh it works. But it’s like watching slugs have a romantic dinner. Maybe I should just read a book while you request that website, hmmm?

Obviously you turn to Google, the eminent sage and eternal junkie for answers. Ah yes, X-Windows over SSH is a ping/pong nightmare, half the traffic is consumed by just making sure that all the lower layers are functioning properly, constantly. Fine. But then you spot things like x2go, give that a shot, eh… it’s somewhat better.

In the end, the promise bends to tools you already have. Like TeamViewer connected to Windows 10 on a different laptop.

Heh, assuming TeamViewer stays functional that is.

Technology is bittersweet. We have such command of so many wonders. I can’t shake the feeling that it’s all an immense house-of-cards. I suppose I’ve seen too much, I know too much, “I’ve seen things you people wouldn’t believe” kind of running through my head. Like looking at Layer 1 connections secured by… chewed bubble gum.

And all the various cheats and hacks, because you naturally want something, but you can’t get it because the people who provide the thing, won’t provide the tools for the extra things you want. They aren’t going to write the code, their codebase is secret, you can’t submit code yourself, and so you just sit there, google searching and finding kludgy-as-fuck solutions to your headaches.

Yeah, that’s fine, be a prat. I’ve got a python script that scrapes your shit and does what I wanted to do.

Case in point, Signal. I love Signal. I will always love Signal. But I want to search on metadata within the Signal app. No. Like the soup nazi, no metadata for you! Only search on stuff in primary stream. Fuuuuuuuu.

So I have a group chat, it’s all my friends, in Signal, and we use it as a blazing-bright thread, it connects us all, geographically spread throughout the United States. It made the COVID-19 Pandemic less lonely. It was a community of dear friends and we could be together without risk.

So, I journal, have an app for that, but obviously Signal doesn’t work with the journal app, so I can’t just hoover all the Signal content into the journaling software. Sometimes I forget to review everything we said in Signal, so the date-of-chat just slides off primary display. You could scroll, but wouldn’t it be nice to search on metadata? Like take me to the first thing shared on Sep 01, 2021? That would be nifty! NO. NO SOUP FOR YOU.

So, no metadata searching. Fine. So, enter the raw kludgy “fuck it, this is also a solution, damn you all” solution. Can’t search on metadata, but just on raw data, so, lets add the data markers we want to the stream! At 6am, write the date into the stream, every day. Then you can use the tools in the app to search on what was shared, and since the metadata you want is “shared”, now you can search on it! Well, okay! “Sep 01, 2021” look! YAY! That’s what I wanted!

Obviously this creates a “Forking House Of Mirrors”… one bullshit kludgy solution leads to a new problem. I don’t want to wake up at 6am to put the date into Signal stream. OK. Lets automate that. Enter Signal-CLI. shakes head fine. So, lets try to connect to the service, that was a hard climb. Okay, now it’s as group, what groups are there? No groups. What? No. Send something to someone, then ask again. Okay. <<send>> how about now? OH YES, THIS GROUP!? You need a special hex code for this.

If you have this hex code, you’d think you could use that without having to ask going forward. No. New install? You can’t just simply use what you know to peek around the corner, no. You need to run around Robin Hoods Barn all over again, and now you can use it! HUZZAH. FUUUUUUUUUUU.

So, finally, we can send signal data from the CLI. Next, lets figure out the date commands picky-picky formatting rules. How to get Sep 01, 2021?

We’ve got that! YAY! Okay, so lets write a Bash script! Get the date, and at 6am write it out to the Signal group. Write script, change mode on script so it can execute, plumb the foggy memories you have of crontab, and boom. Failure.

FUUUUUUUUUUU

Ah yes, cardinal sin, I didn’t explicitly declare the specific paths to signal-cli, echo, mv, fuck, any command at all. Call the script yourself, works, cron calls? Lost. Fixup. Dive into vim. Find your cheatsheet. Gah.

Finally, good god watch it work. 6am every day, a machine you “rescued from the landfill” with some half-forgotten linux distro you can’t remember is actually working and that’s fine. Now, when it’s Sep 03, 2021, you can search on Sep 01, 2021, to get back and manually journal what you remember telling people, because there it is. Click-drag.

All because metadata isn’t searchable. I got what I wanted. Everyone can benefit from it too. But it is complete mess.

This is why entire afternoons are incinerated on the pyre of “Fuck, I wanted XYZ, but the devs don’t speak English, their angel investors aren’t interested, and nobody but me would ever want this feature… so… fuuuuuuuuuuu”

I suppose I could attempt to ask for whatever it is I think would be good, but devs live on the moon, or as much as would be useful, they do. So no. We don’t tell devs anything. We just muck about, finding fragments on GitHub, trying not to get sick that Microsoft owns them now.

So you find gists, you find forked projects, you find python code fragments. The dependencies aren’t circular-misadventures-into-the-fog, you try to remember basic linux stuff because you haven’t had to screw around with any of it for decades and crontab went off to the same Elysium Fields that Trigonometry went off to…

Google Fu. Another worrisome “house of cards” right there too, but lets not look too closely at it, lest it collapse. Or sell our identity to Belorussians.

It doesn’t take much at all. Fragile houses of cards built on other fragile houses of cards. People mobbing on top, like hapless Eloi sitting down at the picnic tables and never having a single bright shiny thought in their pretty little heads because food is always right there, on the table, same time every day. Meanwhile, underneath, the Morlocks are banging on pipes, and every once in a while grabbing an Eloi for a snack.

That’s the Internet. Humanity on top of the Internet. The rot in Layer 8.

And all you really do is shrug. You hope for a better world. Every once in a really long while you stumble blindly over something truly elegant. It’s like tripping over Rivendell and spotting an Elf walking along a curated beautiful path of perfectly carved scrollwork.

And it’s only momentary. The pile of constantly shifting wreckage we call the modern world continues to shudder and throb. It all works, and you marvel that these people manage to continue to live in all of this… wreckage.

VAR Blues

I had to step away from the VAR I was using at work because of a recent change they had instituted with my business account. For years, I had enjoyed a classic relationship of having a single VAR Account Executive assigned to my account, where the AE would learn from me and get to know me, and I would get to know them as well. It was a very successful working relationship, and had been the way of things for six years. Around two months ago, the company made a change. They moved my business account from the structure that I was familiar with over to a team-based structure, and billed the benefits to include “There will always be an AE to work on your account” as a value-added proposition. I was worried that the change would instead eliminate the engagement, the learning, and the developing relationship between customer and reseller.

This new structure included a single shared email address that many people had access to, the AE’s assigned to the “Pod” and the “Pod Manager” who also kept a view on the shared mailbox. I was supposed to send every correspondence to this shared address. At first, I enjoyed the value proposition that there would always be someone to get my messages and to execute my requests. Although, to be brutally honest, access to my Account Executive was never really a problem, so this value proposition was actually a “solution in search of a problem” that I didn’t have. It wasn’t until much later, in the retrospective analysis, that I came to realize this as more significant than I considered it at first.

It was after this, when the rest of the feature set for this new structure started to appear. I’m certain that the VAR thought that all of these things were only enhancing value for customers, but really every step just led me further away from where I was most comfortable. I wasn’t able to “get to know” my team of Account Executives, they remain faceless, voiceless text in email. This lack of humanity was at first not considered to be an issue, but later on became significantly problematic. The disconnection accelerated as we progressed. I was no longer handing work to an Account Executive, asking them for advice and tips, and there was a significant amount of value that I was suddenly unable to access. I had come into the arrangement with a habit of asking my Account Executive to send me quotes on various items, and they would seek the best fitting item that suited my preferences and hand me a quote for the recommended items that best fit my needs and, during pandemic, had a better chance of being in stock. This habit was broken by the new way of doing things. I was no longer able to reach out to an Account Executive to get advice, to have their vantage point much closer to the manufacturers and distributors that we all were using to acquire technology, now I was supposed to simply go on the VAR website, find what I wanted, do my own shopping, and then assemble my own quotes. This feeling of being cut loose became pervasive because it was just another touted feature, considered by the VAR to be part of the “Value Added”, and quickly included not only writing my own quotes, but submitting my own orders as well.

The loss of engagement, the anonymity of the Account Executives involved, and how I was supposed to move all of my previous activities to self-directed work, ostensibly leveraged on the VAR website, all touted as “value added” components were actually just the opposite for me. It wasn’t until I started actually living in this new environment, doing my tasks this new way, that I realized just how much I had missed the old way that I used to do things. The value proposition was always above board, nobody was intentionally being manipulative or malicious, but the result was cold, impersonal, and made me feel like there was an erosion of all the value that at one point was part of my “value added” experience with my VAR.

Whenever there is a change, items can be lost in translation, they can get missed, I do not fault anyone for missing say one or two small things as the customer and the Account Executive in the VAR start to grow together and establish a working relationship together. I didn’t want to, at the time, hold people’s feet to the fire, but that’s exactly what I ended up having to do. I maintain a strict three-strikes policy when it comes to faults, if it’s awful, and you did it three times, that means that it isn’t a mistake, it isn’t overlooking something, it’s part of the design.

The first fault was completely missing the deadline on renewal of security software that my company depends upon to protect us all online. Thankfully, the manufacturer has a very gracious fifteen day grace period, where deadlines are much softer than how they actually sound. The fault resolved, and we moved forward. The second fault came shortly after the first one, and again, the same manufacturer. Missing the renewal of contractual agreements that enable me as a customer to approach the manufacturer’s technical support center if I have any questions or problems. It was addressed and we landed on our feet, but again, we had to sag backwards into the fifteen-day grace period. The third strike was one of tragic poor communication, and one of the most egregious failures I’ve ever witnessed. This failure also coincided with a new Account Executive team member whom I had never communicated with before.

The lack of experience and knowledge on both sides of the divide, again, became a problem that really got in the way. This new Account Executive asked me over several email exchanges questions that were too vague to answer because there wasn’t any included detail. IT is a detail-centric category. We thrive on details, we need exact details, like numbers, or topics, some way to clearly identify what it is that we are talking about. It doesn’t really work when people try to use vague communication styles packed with pronouns and references to unknown objects. Exchange after exchange in this manner became tedious and incredibly tiresome. After several iterations, where I had also started carbon-copying the Pod Manager, did the truth of the situation reveal itself. Once I learned what the object of the conversation was, I tracked it and realized that the subject work should have been completed months before when they had already invoiced my company for the work completed, invoiced and paid.

That was the last straw, the VAR relationship had a tragic and lethal attack right on that spot, right at that time. I began to pursue a kind of “re-entry to the VAR marketplace”, essentially shopping for a new VAR. I found one, chatted them up, had several fantastic meetings and the new VAR has more energy than I’ve seen from the previous one, more professionalism, and more effectiveness. Furthermore, I was also clear with the old VAR, telling them that it was unconscionable how things had unraveled between us, including the “Pod Manager” who never even once attempted to intervene. It was like complaining at a brick wall, for all that I got out of the subsequent correspondences.

The way I was treated was more educational than bothersome. It was a lesson for how important my companies account was to the old VAR, that during the COVID-19 Pandemic, our purchasing slowed because the supply channels also slowed. We wanted technology that was on extensive backorder, and so as our purchasing slowed and stopped, our value to the old VAR ebbed away. The group arrangement was a lesson in and of itself, we were too small, too insignificant to assign to a singular Account Executive, and so, we were effectively downsized as a customer.

We were expected to do all our own work, be our own VAR, as it were because we simply weren’t buying enough to be relevant to our previous VAR. This in itself carries a rather embarrassing knock-on side effect because we had ordered a particular kind of technology from a particular manufacturer and we had eleven items on extreme backorder with the VAR. The old VAR never valued our account, and this was proven out to us by the later revelation that the eleven items on extreme backorder actually slipped into “End Of Life” from the manufacturer. The VAR couldn’t be bothered to re-evaluate the old Open Orders unprompted, discover the EOL surprise themselves, and try saving face by explaining to us what had happened and offering alternatives. What had happened instead, was that the customer had started conversing with a new VAR, discovered the EOL condition, that highlighted just how little the prior VAR cared.

It didn’t matter what the old VAR even wanted to attempt in recovery efforts for the now fully dead business relationship because the single thing that they bring, their “Value Added Reseller” nature, was proven to be totally absentee. We didn’t buy technology for lack of funds, we didn’t buy technology because the people meant to handle the reselling never noticed that what they had already sold ceased to be for sale by the manufacturer!

So I walked away. I moved many orders from the old VAR, spec’ed them out with the new VAR, and actually ended up solving nearly all the seriously backordered gaps in our purchasing stream in one singular afternoon. I sometimes wonder, idly so, if the old VAR thinks about the suddenly cancelled orders, where we were waiting since April with extreme backorders, and then interest fades. Do they even care, do they even notice? They didn’t care enough to look at any old open orders, to even see that the item that we were waiting for shipping on was never ever going to ship because the manufacturer simply stopped selling it. Not having the attention to detail on historical items makes it not really any surprise that they kept on fumbling until the customer simply walked away.

I think that the critical lesson for me in all this drama with the two VARs has been the hidden value that engagement had throughout the entire experience. I never really had a visceral feeling for just how important the engagement was between myself and my VAR Account Executive until it was eroded almost completely. Once engagement disappeared, it was a game-changer for me because it illustrated just how important VAR Account Executives are in the process, how much I had come to rely on them. The VAR Account Executive sits in a very high place, able to see things that customers cannot. Their fingers on the pulse of international transport, distribution, and delivery. I only wish that my prior VAR had not allowed six years of solid, dependable positive experiences go down the drain like it did. I am far happier with my new VAR. With the right engagement between customer and Account Executive, I have already spent $25,000 with the new VAR! These dollar values are still small potatoes in comparison to other customers, but $25,000 is certainly more profit for the new VAR than for the old one who is just sitting on a pile of cancelled orders.

Crochet Yarn Estimating

I have been thinking about starting a new crochet project and it struck me that there were no measurements or calculators online that could give you an estimate of the amount of yarn you would need based on the measurements of the projects result, with fixed terms such as the weight of the yarn, the hook used, and which fundamental stitch was used to create the final result. The goal is to answer this question,

“If I wish to make a square of fabric that is 34 inches by 34 inches, made of half-double crochet using a 5mm hook and worsted weight yarn, how much yarn would I need to complete the project?”

So I decided to find out the answer manually. I worked several styles, including:

  • Foundation Chain
  • FSC
  • FHDC
  • FDC
  • SC
  • HDC
  • DC
  • SC Strip
  • HDC Strip
  • DC Strip

Where the strips were just a single row of that kind of stitch, and the non-strip measures were measuring the yarn after creating a complete work in the square based on the index, which for all of this was five inches. So, a SC Strip is 5 inches long, one row of SC, and SC is 5 inches wide by 5 inches high.

Here are the numbers that I measured, these are inches of yarn consumed for each category (YCR – Yarn Consumption Rate):

Stitch Five Inches LengthAverage YCR per Inch
Foundation Chain 25.54.8
FSC8016
FHDC8717.22
FDC8215.5
SC 785.531.42 (per sq inch)
HDC 80232.08 (per sq inch)
DC 79031.6 (per sq inch)
SC Strip5310.6
HDC Strip78.515.7
DC Strip10220.4

What surprised me was that Double Crochet uses less yarn, a little bit, than Half-Double Crochet does, and then just how close the three fundamental stitches are to each other when you hold width and height at a fixed value.

So my answer is, for a 34” by 34” square of fabric, Half Double Crochet with a Foundation Half Double Crochet will require:

FHDC = 34” x 17.22 = 585.48 inches of yarn

HDC = 34” x 34” = 1156 square inches x 32.08 = 37,084.48 inches of yarn

Full Project = 585.48 + 37,084.48 = 37,669.96 inches of yarn, with 36 inches per yard results in 1,046.38 yards of yarn.

So with this information, I know what my project would potentially cost me, outside of the project time itself, so if I select a superwash merino yarn, for example on yarn.com, their Valley Yarns Valley Superwash is $6.49 a ball, and a ball has 97 yards of yarn in it. That requires 10.78 balls, well, really 11 balls. That’s $71.39, with their 20% off deal, their out the door is $64.10.

The really nice part of knowing the rough yarn consumption rate for these stitches is that you don’t have to overbuy your yarn and then end up with extra of anything. There will almost always be partials but there is a big difference between having a third of a ball left versus having 3 1/3 balls left.

If you find these values to be useful, I would only ask that you leave a comment here letting me know that they helped.

Ulysses 18.7 and WordPress 5.4

Aside

I just had a devil of a time with my Ulysses to WordPress integration. Something underhanded happened on the way to the Forum. Either it was something that WordPress tweaked in 5.4, or my host did something clever to get in the way and didn’t tell me. Someone left a very important bit out, which broke Ulysses, my editor of choice for blogging.

The solution was to be found in these two sites:

David Bosman’s Blog – Ulysses and WordPress and

Hans Bruins’s Medium Post – Ulysses and WordPress

So if you were using Ulysses all along, and it suddenly crapped out on you with WordPress, these instructions seem to do the trick, it did so for me!