Wednesday, April 14th, 2021 01:06 am

this has been a week

Home Assistant

WHEE! My Home Assistant Blue arrived! I did the migration a couple of days ago, and so far, I'm impressed.

1.) I did not realize the case was metal. Sweet.
2.) It seems to run more smoothly than Home Assistant ran on the Pi, which I'd expect since the Blue was purpose-designed to run Home Assistant.
3.) The resource usage is much better; process temperature is 25% lower and there's a minor but definite increase in speed. And this despite the fact that my Pi ha 8G of RAM and Blue has only 4.

DIY NAS - A New Thing to Do With a Pi

This means that my Pi is now free. Or was, because since it was available, I decided to experiment with diy NAS (network attached storage). Basically, download OpenMediaVault to the Pi, connect my external four-bay enclosure that holds my media to it, and go to town.

I'm still learning OpenMediaVault, so while I did get it working smoothly (with some early annoying hiccups), I want to do some more experimenting before doing a write-up. I'm not married to it, so I am considering trying a few other OSs.

GUI
The GUI is nothing to write home about, and while it's very organized, the design choices are sometimes redundant so it feels more cluttered than it actually is.

I admit I'm spoiled: Home Assistant has a gorgeous user interface and that's just the default; most of what you do in Home Assistant is make it even better and more responsive and more customized. If you have an imagination and a vague grasp of any programming or web design, you can do anything.

OpenMediaVault's UI, like DD-WRT's UI (if you're playing the home game, DD-WRT is the open source software you can use for routers), is--not any of that. It's functional and I will say you can tell no one there made the mistake of trying to break barriers or disrupt the system or rebel against convention and do weird shit with javascript and too much time on your hands, and that's something.

It's utilitarian, and like I said, the design is both clear and redundant. On the left is the sidebar, which is divided into sections, and each section name can expand or collapse all items within. On the right is the main screen, and in a narrow header above it, it has an icon of a house (Home). All items in each section are clickable, taking you to their individual pages. Fine so far.

If you click on Home, the main page shows all the sections in rows, and beneath each section name are all the items in that section and their icons. Everything is clickable: section name on the Home Page takes you to the section page, where all the items are listed; item icon on the Home Page takes you to the item page. A bit redundant, but okay; maybe someone fears sidebars and you can minimize it.

Back to the sidebar; same thing. The section name isn't just a name; it's also a link to the section page with all the items, all clickable to their own pages.

On the header just above this, it shows Home in a button, then the Section Name in a second button. If you click an item--either on the sidebar or on the section page--that appear as a third button. Those buttons are also clickable so you can move forward and back, though you never--so far--get more than three deep and if you have the sidebar open, it doesn't matter, but okay.

It's not easy to explain why I find this weird, except I can't work out why you need three (four?) ways on the same screen at the same time to go to the same place. Basically, this design means you are always at most one click deep anywhere, which would be good but you are only one click deep in three separate ways. It's confusing if you don't expect so much redundancy but it's pretty much impossible to get lost.

I can't tell who it was designed for; DD-WRT was made for intermediate to advanced network and programming people (and it shows in the documentation like whoa), while OpenMediaVault seems to be for everyone and anyone, at least as far as GUI You Will Never Get Lost In Really (so far, there's nothing hidden in nooks or crannies or only appears if you know the right places to check on another random page or tucked somewhere random because fucking with users' heads is fun).

But there's some things I'm not sure an average user is going to know to do or know why using the documentation and quickstart, and if their drives already have media on them--aka not blank or brand new--there are some things are going to be baffling as shit--though super easy to fix--but I'll save that for my write-up.

Verification/Validation

It's an Okay/Apply system. You do an action, click OK, then you have to hit Apply before you do anything else. And almost everything requires it. The first registers the change; the second applies it to the system.

I am not fond of these, but I get why they exist. Most other systems that make me do an OK and Apply is to save time and resources; you can make several changes and click OK for each to store then, and then hit Apply so the system will do all of them at once. On my DD-WRT routers, it was a time saver since each action would take a while individually but massed together much less.

Which I thought this was at first because there's a variable pause between OK and Apply. Long enough for me to want to leave the page and it won't let me and a banner appears at top with Apply. Like, the pause is just long enough that you're ready to go and then BANNER APPLY. Argh.

Then--new and frustrating--after Apply, there's a third check "Do you really want to..." and seriously????? I can get a legal gun* in Texas with less hassle than wanting to have SMART notifications sent to me.

* That is hyperbole, but honestly, not very much. And the fact I have to say its hyperbole demonstrates that.

In Closing

I got it up and running, scrubbed Plex and added all my media from new home NAS, and gotta give credit, Plex plays smoother, faster, with a lot less hiccups than playing from my external attached as a share on my router. This is not a bad alternative to buying a NAS; I'll do a price breakdown when I do the write-up, but not including hard drives, I'd say around $250 or less.

School

We're a little over two thirds through the semester; in Intro to Computing, provided I finish at least three more of the four assignments and get full credit on each, I should have an A (if I do all of them and get full credit, I also get an A along with a glow of accomplishment).

Programming Fundamentals is chancier; I have an A right now, but that only includes my first four projects and my first exam; there are two projects ungraded, one I'm doing now, four more projects in the future, and two more exams. My first exam was an 83.75, which was upsetting (I studied for that one), but provided I get a perfect score on all my projects, I can afford a minimum of 67 on each of the other two exams. Which, hopefully, it won't come to that, but I seriously studied for that test (I took notes and reviewed them, even) and as he hasn't yet released the test for us to review what we missed, I still don't know what all I got wrong and that's haunting me. And not making it easy to prepare for the second exam, either.

Finished registering for the summer semester for six hours and fighting myself not to try for nine hours until fall. I mapped out my degree assuming nine hours a semester spring and fall with six in summer, but I want to try and go to 12 per semester within the next eighteen months.

It's not the workload that worries me, actually; I can do it and pass (very probably), but this time, I want to do it with As, and not just an A, but a dramatic A, like a 96-100 in each class across the board.

Educationally Speaking

When I was in high school and college the first (and second) time around, I was never sure that I could do it and was constantly surprised when I did, and not surprised at all when I got a B or even a C; I didn't like it, but the ways of the grades and my brain were mysterious and I had no idea why I couldn't just sit and study and have that actually work and instead have to depend on my ability to learn fast in gulps and short bursts of short term memorization. Back then, I couldn't even take good notes: I either reproduced the book or lecture (until I literally couldn't concentrate a second longer, which was often) or all the wrong things; I could not work out the alchemy of how you decided what mattered and for that matter, how the fuck anyone could stare at that Wall of Text Textbook and learn anything.

Truthfully, until now, I really genuinely did not realize the extent of a.) my ADHD and b.) the effects of medication. Back in 2007 when I went back for a semester, I noticed a difference--this was right after I was medicated--but I got an incredible promotion after one semester and learning that job took all my attention for a while.

When I started this semester, it was pretty much how I started every semester; hopeful but resigned to a best 'better than last time maybe?' I downloaded a program for notetaking (and eventually started using it, but that's skipping ahead), and as these classes are pure online without online or RL class times, read my syllabus carefully, got my online text books (I love love love online textbooks now), and settled in to read productively or die trying. Honestly, I half-expected the latter to become a real possibility, because sometimes, textbooks are really fucking boring.

Now, an ADHD Sidebar )

So yeah, this time, I want A's; I want to turn in projects that exceed not just minimum but maximum requirements and involve many bells and whistles; I want to perpetually have read a chapter or two ahead before class and be overprepared for any assignment; I don't just want to get through class but learn and I mean learn everything.

I mean, I always wanted those things, but now, I think I can actually do them. I live in hope, anyway, and that's new, too.
Home Assistant Blue

My Home Assistant Blue shipped and will arrive by April, and I'm trying to work out when I'll have time to transfer Home Assistant from the Pi to it.

Taking a backup of HA on the Pi and moving it to the new device is the easiest method--with Home Assistant, you can genuinely just do that and not lose data or crash--but years of computer and tablet upgrades (and regular nuke-and-scrub of my Ubuntu Server) have taught me the value of starting with a clean slate when it's feasible. There's always extra customization I no longer need or code I refactored but commented out the original or updates to the base code that mean some code no longer works.

See, the thing is, until I finalize, I almost never, ever delete code.

If it's inconvenient or makes a script nightmare long or confusing, I'll move the original to backup, just in case. If it's a minor update, I comment out the old code for at least a few runs or until I forget why the hell I'm keeping it. If it's new code or a minor refactoring, I create a backup first. If it's a full refactoring, I usually create a copy of the original, give it a name like codeName_Refactor or something, and keep the original clean and working until I'm done testing, then rename the original to codeName_old or codeName_orig, rename codeName_Refactor to codeName, and move the old code into a folder. And in some cases, I'll just move all the old code or experimental code that won't quite work to the bottom of the script and just comment it out because there's a chance I'll need it back and it's really annoying to paste code between multiple nano terminal windows.

(In VBA in Excel or Word, I move it all to a module named OldModule and add an 'x' to every sub or function name. In Googledocs, I do the same with Javascript. There is code in both older than two thirds of my nieces and nephews. It's like the code version of hoarding or something IDK.)

A clean slate is maybe the one time I can do housekeeping. After doing all the necessary basic configuration and adding in all my integrations, add-ons, etc into the new device, I can move code over either in entire pages or a piece at a time and reload to make sure it does what I think it does (or work out what the fuck I wrote it for).

Fortunately, Home Assistant makes that incredibly easy; the last time when I got a second Pi to run Home Assistant on, after I finished configuration of the second Pi and it was ready to run, I disabled the integrations that couldn't be run on two HA instances at the same time on the old Pi, then just left them both running while I went through all my custom yaml and python and moved it over sometimes a single function or script at a time.

This, by the way, is my idea of the Best Friday Night Ever.

Depending on time, I'm going to try to write up a detailed step by step tutorial on how to set up Home Assistant on the HA Blue. One of the biggest advantages of buying it--other than supporting open source development and the cool blue case--is that Home Assistant ships on the HA Blue already installed, so it's very much plug and play; you literally plug it in, add ethernet cable, then go to your computer, open a browser, and go to homeassistant.local:8123. That's it. The only things you need to do is if you need z-wave or zigbee functionality is buy either usb sticks or a compatible hub, but if everything in your house is wifi, that's pretty much it for hardware. Now its just onboarding, adding your integrations, and trying out all the different theme colors on your dash.

School

Intro to Computing

We're now in Week 5 of School. I'm currently about a week behind in the Intro to Computing self-paced course but while that was mostly due to work + winter storm + other things, honestly it was also because it was the class I could fall behind without penalty. The class is a basic catch up to current technology and the internet and how to use Office; it's shockingly useful for someone who may be coming back to the workforce or needs an non-terrifying intro into current tech and current internet.

Also, it's the one that is ironically both the most work and also the one that's probably easiest to pass without a super amount of effort, which follows the course's purpose. All you really have to do is do all the activities and also create for yourself an Excel spreadsheet to keep a running calculation on the lowest grade you can afford in each activity and still get an A (or B).

But it is a lot of work. For the first six weeks, each week is
1.) one (1) or two (2) book chapter on something about technology (there are six total chapters)
2.) one (1) graded skills test for each chapter
3.) one (1) graded practice exam for each chapter
4.) one (1) or two (2) of three modules on how to use a Microsoft Office Product (Word, Powerpoint, Excel, Access). (There are three to five modules per office product in the course.)
5.) one (1) graded project for each module

On top of that:
1.) One (1) Capstone Project for each Office Product: total of 4
2.) One (1) Exam for every two chapters: total of 3\

Except for the practice exams and two-chapter exams, to get an A you really just need to follow the instructions to do the projects like "Create a Flyer" or "Create a Powerpoint Presentation" or "Create a Business Letter". It's auto-graded so you know immediately your grade, you get a report on EXACTLY WHAT IS WRONG AND HOW TO FIX IT, and you get to submit three times and it takes your highest grade. The graded skill test is a graded review; I'm not sure it's possible not to get a perfect score.

The exams are--not so simple. They aren't exams; they're thirty five to a hundred questions, shown one at a time, you cannot stop until you're done and you cannot change your answer once you go to the next question. No, I'm not kidding; I've never been this stressed by testing in my life.

Now, because of the number of activities and the weighting: you can, actually, get a zero on every single exam and as long as you get a perfect grade on all the projects, capstones, and skills test, you can get a D. With the Skills test, perfect scores are built in; with the projects, you get three attempts; my lowest grade on a project is a 97/100 with one attempt left (that I can still do until the end of the semester) but just didn't feel like fighting footnotes for those last three points. THe lowest grade I got on an attempt was an 88 before I fixed the problems; this is not undoable at all.

Right now, for me to get a B, as long as I nail every project and capstone, I can fail every remaining exam (I have two two-chapter exams and three one chapter exams left) with a 52 out of 100 or 14 out of 25 (52%) with an extra two points to spare.

For an A, however, I have to average an 80 on each exam and get a perfect grade on every single project and Capstone that remains, though I have an eight point buffer. The lowest I've gotten on an exam is an 84 out of 100; I shouldn't be worried. I read the chapters; I take notes; I study. It autogrades when you're done and you get a full report with every question and your answer and whether its' right or wrong; you can take the exam up to twice.

But. When you start an exam, you cannot stop; each page has only one question and when you answer it, it goes to the next page and question and you cannot change it; there are many questions, which is good for grade weight but very bad for what is already low-grade paranoia. And I say this as someone who tests insanely fucking well; I have answered questions in ways that baffle even me on wtf I was thinking. I have to pause and think about obvious questions like 'what is a CPU'. It's--something.

I had to stop short in horror because I was asked the question "Which of these represents a billion bytes?" and for the life of me could not work out if it was megabytes, gigabytes or terabytes even though a.) I literally do know this, I spend a lot of my free time doing video editing and ratioing sizes, and b.) If I didn't, all had to do was fucking divide by a 1000 to get kilo, divide again to get mega, divide again and if the answer was greater than 999, tera, otherwise, giga. I can literally do that in my head instantly.

I spent three minutes staring at that question without any idea how to math.

I am not in love with those exams.

Programming Fundamentals

Incredibly fun. Well, for me: for my professor, maybe not.

There are ten projects and three exams, one project for each chapter, each project one to five or so scripts to write to do a thing; we just finished chapter four this week, and as it's an online class without class meetings but with hard deadlines, its' one I can't put off.

But also, it's not one that I really have any ability to put off; this is like falling into a few fandom and reading all the fic. The only reason I'm not reading ahead is spoilers; so far, the only way to keep me on pace is taking notes. Most of it is stuff I know already from writing python, but there's a lot of very basic stuff I skipped that I'm learning now and lots of generalized concepts, so it's great.

Then there's the projects.

The project exercises start with me following the instructions to the letter; that part is fine. But that takes me maybe an hour or two. Projects are assigned on Monday; I'm done reading by Tuesday and have my project done by Wednesday at teh latest. Due date is the next Monday; I have free time.

This time is spent methodically going through my scripts and adding bells and whistles. I challenge myself by trying each time to use every single element in every single chapter and preceding chapter in every script and where there's a will there's a way. Then I start adding bells and whistles. Chapter Three was If/Else, and conditionals are my jam, but at least I couldn't trap you in a script forever; it would, eventually end.

Chapter Four was Loops.

Which make my feelings for conditionals look like vague liking; I adore loops. My C++ class I'd trap people in elaborate loops in that horrifying tic-tac-toe game where the only way out was to work out what character I secretly designated for the only escape. I nest loops like Russian dolls and left to my own devices it would never end.

To be fair, I (mostly) restrained myself in the first two exercises; they're optional repeaters but not horrifying. But my resistance crumbled when it came to the third.

I got to ask the user for input three times. That is Christmas.

Short story: it's now four times longer than the first working draft, lets the user correct their answers as many times as desired, run the program indefinitely, and when they're done output their statistics on how many times they ran the program, how many times they corrected their data during all iterations, and how many times they corrected their data during this iteration before saying goodbye. I spreadsheeted test data to validate all conditions and possibilities. The only reason you can escape is we haven't reached error control yet and I can't use it until we do or that's cheating; right now, they can escape by just entering a string instead of a number and killing the program if they get desperate.

The exercises for Project 3 were all overkill, sure, but they're nothing to Project 4 Exercise 3.

And I can tell you now, it will only get worse from here. I mean, for other people: this is how I have fun on Saturday nights.
Due to so much current stress I've been skipping out on my posting here. I really am not feeling doing an update on life at this time, but I want to spread this far and wide.

I talked about starting to use open source Home Assistant for my home automation needs with a Raspberry Pi. since then I updated to an 8 GB Pi and added booting from an SSD.

Now there's something a thousand times better: Home Assistant Blue - $150

This is Home Assistant's own Automation Hub built by them. You do not need to build anything, like a Pi; it's a single board computer with 4 USB hubs, 1 ethernet, 1 HDMI, with a 128 eMMC hard drive and 4G RAM with Home Assistant pre-installed and a very cool blue case.

It's plug and play. Literally, you plug it in and turn it on; that's it.

Below the cut is a copy of my tumblr post on a reblog about open source that breaks down everything you need to know

home assistant for everyone )

Home Assistant also has addons for running your own DNS server, DHCP server, database, Samba, HDMI CEC scanner, SSH, code editors, ABD server, Tor, encryptors, Plex, and I think the equivalent of pi hole (either official or community) and that's just to start. It can do kind of everything.

Official Stuff
Official Integrations
Official Add-Ons

Community Supported (non-official):
HACS

The number of community integrations you can get in HACS are an order of magnitude more than the official and then there's the UI stuff. And then there's the ones not even in HACS yet which you can find if you just google "Home assistant" and "Your Thing" which are like--well, a lot lot lot.

If I can get enough people interested in trying, I want to start a DW comm where people can ask questions, post tutorials, code, or just get support and help. It can be intimidating to start something like this, but help and support make everything easier and fun, and I'd love love love to work with/help/talk/support and/or be main venting person to anyone who jumps in.
I'd been back and forth about trying Home Assistant for my home automation needs, mostly due to a.) inertia and b.) SmartThings pretty much covered what I needed.

Then: SmartThings has been transitioning to a new platform and app for two years, and they managed to break a lot of my custom code and other people's--basically, half of why I loved it. Worse, you could not longer choose what devices to expose to Alexa through it.

This is probably where I should go into a little more detail before I start.

the home automation wars: zigbee, z-wave, and wifi )
example: my home and my home automation, a breakdown )
the home automation wars: this is where it got complicated )

Back to Now

With that perspective, the changes to SmartThings--the loss of a lot of functionality--have been disappointing and frustrtaing. Until now, I used it as a central Hub of Everything with Alexa for voice; now, I couldn't, especially when it became impossible to selectively decide what devices in SmartThings were exposed to Alexa.

Example: my Philips Hue lightbulbs were already compatible with Alexa, so they were directly connected to Alexa. To use them in SmartThings, though, I had to direct connect them there as well. So when I connected SmartThings to Alexa, my Hue lightbulbs would connect again to Alexa through SmartThings and I'd have two of everything.

(In some horrific cases, I had three copies of everything in Alexa, all with the same name, the only difference if I opened properties and could see where the device came from. Worse, some had slightly different names, and Alexa would get confused.

Can't lie, it also bothered me on an anal retentive level; everything was messy.)

My two choices were to
a.) remove everything from Alexa, add it to SmartThings, then let everything go through SmartThings to Alexa. That wasn't feasible; some of the wifi stuff wasn't compatible with SmartThings, and I'd never bothered to test the wifi workarounds available because I'd never needed to.
b.) remove everything from SmartThings that was already in Alexa. As it turned out, that was everything but my z-wave and zigbee devices.

I depressingly chose number two, which was only marginally better than one. This, as a result, mae Alexa my central smart home hub, not just my central Voice Stuff.

Alexa is a terrible central hub; there's limited access when not on a mobile device (the web interface is--horrific). Worse, all connections to Alexa were in, not out. With SmartThings, I could connect them into Alexa, see them and use them in Alexa, but anything connected to Alexa did not connect back to SmartThings. Worse, the automations (routines) you can create in alexa are functional but integration with devices isn't perfect. Alexa's routines simply didn't connect well with zigbee and z-wave devices even when they said they worked. There was delay, a pause, or most likely, not work at all.

In other words, my bathroom lights stopped coming on when I came in the room and that meant war.

(I seriously don't remember how to turn on and off bathroom lights when I enter a room; this is not going to change. I live Star Trek; I'm not going back.)

When I got the NVIDIA Shield TV to take over Plex, I had a Pi with nothing to do with it. So last weekend, I sat down and started the process of learning about Home Assistant.

Home Assistant is a program that makes your Raspberry Pi into a home automation hub. It can connect to any almost existing hub you have with their integrations, bringing them all together. Much like SmartThings but even more so, Home Assistant depends on community integrations, so there's even more made by the community, mostly for devices/ecosystems that don't have an open API.

Basically, it's what SmartThings was doing for me before: I connected all my devices to it and it controlled them, I can create automations for lights or whatever. Better, it allows me to connect to Alexa only things that I want to, so I can connect my Hue Lights to Home Assistant and they won't travel over to Alexa and show up twice or five times or whatever.

There are a billion awesome differences between Home Automation Pi and SmartThings (and other home automation hubs) but there's one big one: the Raspberry Pi does not come with z-wave and zigbee functionality. You either have to buy the parts--a z-wave stick and zigbee board--and make the connections yourself (not hard), or use an existing z-wave/zigbee hub that can be connected to Home Assistant.

In my case, I had my SmartThings Hub, so that is my z-wave/zigbee device. And while SmartThings has made itself more annoying, it does still possess the ability to connect to almost any z-wave or zigbee device in existence one way or another, and much to some ecosystem's despair, even when the break the zigbee or z-wave standard so you can't

(Apparently it took about a day? for users to work out how to connect Aqara devices to SmartThings, even though they deliberately tried to use non-standard zigbee so you'd have to buy their overpriced hub. Good try, Aqara.)

home assistant, an introduction, finally, with pics! )

I'll do a second post about configuring Home Assistant as time permits, but hopefully, this made you curious. As I want friends in my journey and will get them any way I can.
I am taking a month off social media--by that I mean tumblr and twitter--to see if that helps calm me down. I'm officially bored with anxiety; I didn't know it was possible to experience both together, but here we are.

In the last few weeks;
1.) learned new exercises for PT.
2.) installed Home Assistant on my Pi, which is something like SmartThings hub but like, so fucking much cooler and better and more awesome.

After problems with SmartThings meant I de-integrated everything that wasn't z-wave or zigbee. SmartThings went to a new standard and also new app and lost functionality, so it was no longer fun; all my scripts were being killed and all the ones from the community were dying. However, it's still an exemplary z-wave and zigbee hub, so it could be used for that (a z-wave stick is about the same price as a SmartThings hub and that's without zigbee), so I finally sat down with my Pi and installed Home Assistant.

It's about a thousand times better than SmartThings. For one, it's web based, not app based, so the web components are both not an afterthought and also exist; for another it uses Javascript but mostly Python and YAML, and I've really been wanting to learn those two.

I will do an entry on this eventually because holy shit, this is fun.

Related....

3.) Did my first fork and pull on github to bug fix a python script. Yes, I'm finally using github as intended.
4.) Am finally learning python (and yaml). I learn by doing, which generally means I need a project to work on, specifically one that does something I really want to do and can't make work in any language I already know. Making my Sengled Color Bulbs work, as it turns out, was the trick.

When doing the integration for my Sengled bulbs into Home Assistant, the Sengled script had some problems and since python is Javacript With Indents, I went to read the script then read the developer's notes. He didn't have some of the bulbs I did, so I sent him my logs and then made some corrections to the scripts since it would be harder for him to do that part without the right bulbs to test (Sengled doesn't have a public API so he was doing it all with the app, a Sengled Hub, and watching traffic and jsons). Anyway, he invited me to submit a pull instead of sending him script bits, so I learned github while python to do the corrections. It was fun.

adventures in scripting )

Adventure indeed.

Profile

seperis: (Default)
seperis

Syndicate

RSS Atom

Tags

Quotes

  • If you don't send me feedback, I will sob uncontrollably for hours on end, until finally, in a fit of depression, I slash my wrists and bleed out on the bathroom floor. My death will be on your heads. Murderers
    . -- Unknown, on feedback
    BTS List
  • That's why he goes bad, you know -- all the good people hit him on the head or try to shoot him and constantly mistrust him, while there's this vast cohort of minions saying, We wouldn't hurt you, Lex, and we'll give you power and greatness and oh so much sex...
    Wow. That was scary. Lex is like Jesus in the desert.
    -- pricklyelf, on why Lex goes bad
    LJ
  • Obi-Wan has a sort of desperate, pathetic patience in this movie. You can just see it in his eyes: "My padawan is a psychopath, and no one will believe me; I'm barely keeping him under control and expect to wake up any night now to find him standing over my bed with a knife!"
    -- Teague, reviewing "Star Wars: Attack of the Clones"
    LJ
  • Beth: god, why do i have so many beads?
    Jenn: Because you are an addict.
    Jenn: There are twelve step programs for this.
    Beth: i dunno they'd work, might have to go straight for the electroshock.
    Jenn: I'm not sure that helps with bead addiction.
    Beth: i was thinking more to demagnitize my credit card.
    -- hwmitzy and seperis, on bead addiction
    AIM, 12/24/2003
  • I could rape a goat and it will DIE PRETTIER than they write.
    -- anonymous, on terrible writing
    AIM, 2/17/2004
  • In medical billing there is a diagnosis code for someone who commits suicide by sea anenemoe.
    -- silverkyst, on wtf
    AIM, 3/25/2004
  • Anonymous: sorry. i just wanted to tell you how much i liked you. i'd like to take this to a higher level if you're willing
    Eleveninches: By higher level I hope you mean email.
    -- eleveninches and anonymous, on things that are disturbing
    LJ, 4/2/2004
  • silverkyst: I need to not be taking molecular genetics.
    silverkyst: though, as a sidenote, I did learn how to eviscerate a fruit fly larvae by pulling it's mouth out by it's mouthparts today.
    silverkyst: I'm just nowhere near competent in the subject material to be taking it.
    Jenn: I'd like to thank you for that image.
    -- silverkyst and seperis, on more wtf
    AIM, 1/25/2005
  • You know, if obi-wan had just disciplined the boy *properly* we wouldn't be having these problems. Can't you just see yoda? "Take him in hand, you must. The true Force, you must show him."
    -- Issaro, on spanking Anakin in his formative years
    LJ, 3/15/2005
  • Aside from the fact that one person should never go near another with a penis, a bottle of body wash, and a hopeful expression...
    -- Summerfling, on shower sex
    LJ, 7/22/2005
  • It's weird, after you get used to the affection you get from a rabbit, it's like any other BDSM relationship. Only without the sex and hot chicks in leather corsets wielding floggers. You'll grow to like it.
    -- revelininsanity, on my relationship with my rabbit
    LJ, 2/7/2006
  • Smudged upon the near horizon, lapine shadows in the mist. Like a doomsday vision from Watership Down, the bunny intervention approaches.
    -- cpt_untouchable, on my addition of The Fourth Bunny
    LJ, 4/13/2006
  • Rule 3. Chemistry is kind of like bondage. Some people like it, some people like reading about or watching other people doing it, and a large number of people's reaction to actually doing the serious stuff is to recoil in horror.
    -- deadlychameleon, on class
    LJ, 9/1/2007
  • If imitation is the sincerest form of flattery, then Fan Fiction is John Cusack standing outside your house with a boombox.
    -- JRDSkinner, on fanfiction
    Twitter
  • I will unashamedly and unapologetically celebrate the joy and the warmth and the creativity of a community of people sharing something positive and beautiful and connective and if you don’t like it you are most welcome to very fuck off.
    -- Michael Sheen, on Good Omens fanfic
    Twitter
    , 6/19/2019
  • Adding for Mastodon.
    -- Jenn, traceback
    Fosstodon
    , 11/6/2022

Credit

November 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 2022
Page generated Apr. 23rd, 2025 09:08 am
Powered by Dreamwidth Studios