Archive for October, 2019

I Can Knot Not Code

Tuesday, October 29th, 2019

I think knots are so cool. Knots, as in, tying rope in special ways to accomplish different tasks with the rope. This website is way better than it even needs to be and I love it.

I even have this very fancy hardcover book that is like the classic knots book: The Ashley Book of Knots.

Ya know what else?

I suck at tying knots. I don’t know a single knot beyond tying my own shoelaces and honestly, I’m not even particularly good at that. I even have some nice rope at home and plenty of space to practice.

I think it’s because my thinking knots are cool and desire to learn how to tie them isn’t enough to actually get better. All the want in the world and fiddling with rope once in a while isn’t getting me there.

Concentrated practice over a long period would probably do it. Putting in the hours.

But… I probably won’t.

I don’t have any stake in the game. I’m not in a position to need special knots very often, and when I am, my garbage knot skills get me through anyway.

We’re all different, so I’m just speaking for myself, but I can only learn what I need to because I have a stake in the game. I have to learn because the situation is right in front of my face day after day.

Something something learning to code.

Caramelized Onions

Monday, October 21st, 2019

I made French Onion Soup the other week. In the end, it just made three big dinner-sized bowls of soup, but the recipe started with carmelizing 5 big onions!

Once that step is done, they must be a tenth of their original volume.

Probably some kind of metaphor there. Like things can feel so damn big in your head right when they are happening. But let them simmer for a while, they get a lot smaller.

I Still Kinda Don’t Get Event-Based Hashtags

Thursday, October 17th, 2019

I think I understand hashtags in general. On services, like Twitter, that support them, I can put a word like #party and, when published, it becomes a link to some listings page of posts that also use that same hashtag. They are user-generated classifications that help users get to related posts. Discoverability, I suppose. Make a tag unique enough, and you might have it all to yourself.

But events use them all the time. They ask attendees to use the hash tag.

I’m at #CoolConf. Woohoo!!!!!!

Now if some other user clicks the #CoolConf link, they’ll see posts from other people who used the same hashtag. Except, with no context at all.

What’s the event about? Where is it? Who runs it?

I hesitate to use the hashtag because I worry people will click it and be totally confused about the lack of context.

I’d rather use the handle of the conference itself. That way you’ve included something that if someone else clicks on, it has a profile attached to it that explains the context of the conference.

Plus, you’ve still included a unique bit of text that makes the related posts searchable. So if I’m searching for posts from a conference, I can search for “CoolConf”, and get posts that both mention @CoolConf and use the hashtag #CoolConf.

Another benefit to using a mention is unique properties of it on platforms. On Twitter, if I start a Tweet with an @mention of a specific account, it won’t show that tweet to my followers who don’t follow that account. Some people use that to scope tweets to a certain event, so you are participating in a conversation with other conference people, but not bothering your other followers.

In that sense, a @mention is just like a more-useful hashtag. Right?! Or am I missing something here?

Coincidences and Succession

Wednesday, October 9th, 2019

I enjoyed Succession for the most part. I just finished season one last night. Here’s one brief thought though of something that bugs me a little (aside from, you know, the fact that I spend so many hours watching a TV show entirely comprised of deeply awful people).

There is a list of rules that Pixar has for crafting stories. #19 goes like this:

Coincidences to get characters into trouble are great; coincidences to get them out of it are cheating.

So like, a meteor crashes to earth and sends a bunch of dinosaurs flying off in different directions, and they are very sad because they have lost each other. Fine, a random coincidence causes trouble. To solve that trouble, the characters have to try and go through a journey. If a magic portal opened randomly up and they all stepped through it and were together again, we’d be disappointed in that story. The magic portal being a coincidence to get the characters out of trouble.

In the middle of the first season of Succession, there is a huge power-play with Kendall trying to forcefully remove Logan from the company. It requires a board meeting and a vote. Kendall is supposed to be in the room and Logan was supposed to be out of the room. It is a time-sensitive thing, but Kendall is essentially stuck in traffic and doesn’t make it there until the very end of the meeting. It made for an intense scene. But, we have every reason to believe the outcome of the meeting would have been entirely opposite if Kendall was there. Having a major plot point be determined by traffic just wasn’t very satisfying. It feels like a coincidence getting a character out of trouble, or at least, having undue influence on the plot.

It would be one thing if this was a one-off, but it plays out again at the end of the season. Again Kendall has a plan to oust Logan. Again the plan is thwarted by coincidence. In this case, a deer in the road causes a catastrophic event that torpedoes the entire plan.

I haven’t seen season two yet, but I’ll probably watch it. My hope is that we get to see some plot play out without a random coincidence having so much plot impact.

My Guitar

Monday, October 7th, 2019

UPDATE: Lol this blog post was me starting to realize I had the wrong guitar. It was a big mystery that is now solved.

I have a pretty nice guitar that I bought from my friend Mark maybe 4-5 years ago. He was wanting to put some money down on a piece of land in Ashville, and gosh dangit if they didn’t get that land and live in the house on it part-time.

I have no paperwork for the guitar. There isn’t a lot of information on the thing. I bought a case for it when I got it, so nothing there to say what it is really.

I remember two things from what Mark told me:

I’m essentially blogging this so I have something to remind me any details at all about the thing.

It looks a lot like this. Kinda modeled after the classic Martin D-28. The headstock on it says Yamaha though, so I don’t know if just the body was crafted by Dennis Overton and not the head/neck or what. I paid $5,000 for it, so it’s definitely the most expensive instrument I own. It’s nice and loud and clean and clear and warm, so I’m damn happy with it.

I should take some pictures of it to add to this blog post for posterity.

Reading Outside the Sphere

Monday, October 7th, 2019

Sometimes the most pleasurable articles to read are the ones that are shared with you by someone else. That “you should read this” moment is kind of special. One reason I appreciate having co-workers is this. They share great stuff that I probably wouldn’t have come across otherwise. Two just last week.

One was about selling your software to enterprise companies, even if you really aren’t set up for that.

Enterprise pricing is discontinuous with normal pricing. If the $250 a month plan entitles you to 500 foozles and an Enterprise needs 5,000 foozles, that costs thousands or tens thousands of dollars per month. If an Enterprise only needs 500 foozles, that costs thousands or tens of thousands of dollars per month. If an Enterprise only needs 50 foozles, that costs thousands or tens of thousands of dollars per month. This is partially justified by the amount of pain you’re signing up for by doing an Enterprise sales process, but is mostly just pure, naked price discrimination. Enterprises are not price conscious. Don’t attempt to sell them based on your price.

Sounds right to me. Dealing with enterprise can be such a pain in the ass, that charging $5,000/month for a $50/month plan is actually fair. All the process and hoop jumping isn’t just annoying, it’s taking away from the other work you should be doing. So if this is the work you are doing now, the numbers need to work out right.

The other was Stevey’s Google Platforms Rant. 8+ years old. The whole point is that Google does everything right, and Amazon does everything wrong, except for one thing, and that one thing is so important that it puts them far ahead. That one thing is dogfooding all their own services.

The Golden Rule of Platforms, “Eat Your Own Dogfood”, can be rephrased as “Start with a Platform, and Then Use it for Everything.” You can’t just bolt it on later. Certainly not easily at any rate — ask anyone who worked on platformizing MS Office. Or anyone who worked on platformizing Amazon. If you delay it, it’ll be ten times as much work as just doing it correctly up front. You can’t cheat. You can’t have secret back doors for internal apps to get special priority access, not for ANY reason. You need to solve the hard problems up front.