traviscj

Some of my fondest computer memories are working through the LFS guide on a home-built computer in highschool. Wish I still had the time to do stuff like that!

This is why I walked back my claims a bit in my tldr — basically started trying stuff out and had a “whoa not used to the way that looks, but it does _work_...” moment.

I think I still have the “wrong” preference, but I appreciate learning some new stuff!

Doesn't seem like a _bad_ bet considering it lasted 20 years and like 14 versions, though...

I use the gleitz version with

``` alias h3='howdoi -n 3' ```

_really_ frequently, because often the top answer wasn't what I needed.

Project idea: ML-based snafu-avoider that prompts for confirmation when past invocations by other users in the community of that command resulted in manifestations of regret[1].

[1]: screaming picked up by mic; detected maniacal CTRL-C'ing as immediate as it is futile; permanent loss of peer relationship to the project's cloud (indicating subsequent reformatting and/or booting-off-tall-building)

Bit of pointless contrarian opinion from a long time cat|'r:

> Characters saved with "<"? 3. > Hours lost with accidental ">"? unbounded.

Preface: I'm confident on the shell & with my keyboard: I've been using shell redirects for a couple decades, learned touch-typing up to around 80WPM, and used to note-take math lectures live in latex. Yes, I have noclobber enabled. But...

I can't promise myself I'm on a machine configured that way OR that I won't typo this one on a file that really matters.

I kinda use of `cat` as a `immutability-please` command: it _creates_ a barrier/seam between this file and whatever I'm doing with it, such that I'm _guaranteed_ nothing downstream will mutate it (as long as everything only works on stdio).

In a kinda similar way: by starting with `cat x |` (or even just `cat x`, frequently!), the rest of that compound command is _input source independent_ (and even _ignorant_!) -- if I want to hook it up later to a `curl example.com/xyz.json`, I don't need to split it into separate commands/introduce temp files/etc. This is especially because I find myself doing a _really_ quick scan of the entire file ("did it return the actual object this time or just `{}`?"), then piling on transforms from there.

Something about doing `cat x | head` and then kill-wording to fill in `next-thing` feels faster and more consistent than `head x` and then `next-thing x`, maybe because the filenames are often much longer than the commands? Maybe I need a "go to start of line, forward-kill one word" hotkey?

tl;dr: i'm playing psychological games with myself and TBH not sure I'm coming out ahead.

> ... just works as a classic command-line tool.

impossible to understate the value in: not frequently fucking up your ability to run the critical/mundane/everyday/musclememory commands you've come to rely on after maybe 100+ invocations/week as far back as you can remember just because someone you've never met wanted to bump the .ruby-version in your java monorepo

Chuckles don't meet The Bar. /s

What are they testing for? To make sure you’re taking the ADHD meds or to make sure you aren’t taking other meds?

I read it as “there exists a event schema and some database table for which they can be co-constructed”, not “every database table can be turned into a stream of events.”

I don’t have a proof, but For things like state transitions, I’ve often solved this with a <new_state>_at column. Seeing that field populated would indicate a second event need be emitted. (Assumes no cycles in the state machine.)

Often that table is not the one you have, even if (perhaps especially if) you emitted events for every update — seems like some entropy-like force pushes the universe toward emitting details in events that aren’t reflected in the table records, over a long enough timeline.

> fork?

Because it runs contrary to maintainer’s interests?

My guess is that “stateless” means no durable storage. There’s still state, but in-memory (or maybe transiently stored on disk?) and no guarantee about it being there, perhaps?

You’ve got a couple replies saying “not a problem at my office [where I happen to enjoy the use of such a keyboard]” already, so I’ll back you up here:

I have neighbors with mechanicals that bother me so bad I’ve had to pack up and go home for the day. This is over music on noise canceling headphones.

Different people actually have different amounts of trouble with this, with some having lots of problems. Just lucky I guess?

I love this take!

I’ve been similarly untrusting of a lot of “high tech” approaches to various things, and I derive a lot of joy from products/services/etc that take a “back to basics” or at least minimally- or non-digital philosophy. In particular: I have an affinity for automatic watches and carbureted motorcycles.

If nothing else, it’s a bit of a break from what feels like a constant struggle to keep all the gears turning at work.

BUT... I’ll confess I’m also a sucker for innovations / the occasional new hotness. I recently upgraded a Kawasaki KLR650 (a competent but... “well tested”, shall we say? motorcycle) in favor of KTM’s top-of-the-line adventure bike. The technology difference between the two (despite only 3 model years between them) is incredible: the latter adds 5x the power, ride-by-wire, cruise control, lean-sensitive ABS/traction control, an up/down-capable quickshifter, probably a thousand other improvements.

One day, about 1100 miles into owning the new bike, the dash pops up a low tire pressure warning from the tire pressure monitoring system. It showed the rear pressure was fairly low, and sure enough, I’d picked up a small screw between the treads.

Certainly a TPMS is nothing compared to anything self-driving, but honestly it was a bit of a wake-up call — I WANT systems on my bike to increase my safety level.

I’m not really sure what the lesson is here. Maybe “Look for the middle grounds (the ABS/TPMS-maturity systems) between ancient technology (anything on my beloved KLR) and bleeding edge (non-replicable papers on self-driving cars)”? Seems like this holds up ok, especially as a consumer of those techs... But maybe not for the innovators?

A weekend at most.

WORST—-and I mean WORST—case, could crank out a clone during a hack week.

My read was that the _companies_ get the tax break. The employees get taxed to shit, but it's still better than not getting the RSUs.

> put everything we could into the database (work queues and pup/sub included). It was magical.

I've been building systems on some pretty similar ideas for the last 5+ years, and have a similarly optimistic outlook on it. Just to add another datapoint here, we've been able to scale them up to 20-40x higher than your figure (a couple thousand jobs/sec) on a single (albeit beefy) MySQL instance. It definitely takes a fair bit of care & we certainly have some time invested in it, but it can be done.

I guess my services haven't had (gotten to? :-) ) to share databases with other services for quite some time -- they did when I started my job! -- but even with that concession, the benefits of DB guarantees have been amazing. There's nothing like a transaction for preserving your sanity.

The best tactic I've learned from these systems is to think really hard about the batching & partitioning behavior to keep msgs/sec high enough & transactions/sec low enough to stay afloat.

This guy taxes.

In seriousness, though, how do you build up this level of comfort/familiarity with the tax code? Did you do it professionally?

This doesn’t seem to be the case from the article [1] linked in a grandchild of your original post.

[1]: https://www.consumerismcommentary.com/verify-funds-on-that-c...

Why wouldn’t this require just the info on the check (routing + account #), which everyone involved probably already had/needed to answer the query?

I feel like I've noticed a shift in apparent motivation between 2014 and now. I remember a lot of people talking about doing it as a side gig or whatever in the past, but now I hear a lot about drivers working 8+hr shifts, sleeping in their cars between shifts, commuting from 1-3hr away to drive in SF, etc.

Great point, thanks!

Fun story: A coworker and I are in the process of tracking down exactly violations of this in a misbehaving app, due to an accumulation of RPC within transactions across the entire app. (I used to be responsible for it as a greenhorn ~5years ago, but walked away for the last several years to work in a different part of the org.)

The approach is actually kinda interesting: we add an interceptor in the RPC layer for every outgoing call that asks the DB machinery “am I currently in a transaction?”, and emits a warning (containing at least the RPC being made, perhaps a (partial) stack trace) if so. We had so many occurrences of those warnings that our logging framework proactively started dropping log lines! So the next step is to only emit the warnings once per $duration per message.

Another day in the life...

(No significant experience with either SQLAlchemy or ActiveRecord, but work in Java at a shop that really strongly emphasizes the unit-of-work pattern.)

"give up connection after every query" is, IMO, not _less_ of a footgun, but a _more subtle_ footgun. That is to say, sure, for the most part, it will work ok.

_But when it doesn't_... like when your database goes down for a couple hours, or your rack loses network, or your proxy to whatever cloud provider is running your database, or maybe some other dependency had an outage between the queries... you end up not really knowing what executed and what didn't, with some fraction of your 10k req/sec leaving behind some dangling state -- a partially processed $whatsit -- and no clear/easy way to fix it or even necessarily identify how many $whatsits were affected. At least a few times, I've needed to either wait until records make their way into some OLAP system for offline analysis, or deploy some fixup code, or emulate either/both to fix stuff up after something weird happens.

At least that's what I am constantly afraid of. Do you also worry about this kind of stuff? If not, what prevents it from being a problem in your environment?

Signed,

The guy constantly badgering seemingly everyone about minding their transaction boundaries.

They probably use their sharklazers emails to sign up.

I think there could be an argument that the personal freedom you mention, when risks are realized, can degrade the experience of the world at large. Lazy example: a botnet running on many machines compromised as described above sending spam email to innocents.

I’m still on the fence about whether that justifies their protocols. I think I actually lean toward “no”, but I’ve also lately become keenly aware of the difficulty of even simple things like keeping everything up to date, and my lack of real insight into what those updates include. If I’m effectively trusting them anyway, might as well trust them to get it to me ASAP, right?

I’m also enough of a realist to assume there’s a Fight Club style “A times B times C > X” reputational/financial risk logic going on here. If there’s few enough of the devices out there, it’s probably cheaper to apologize (legally, as in settle).

It would seem my insanity wolf idea was only low-tier insanity! Thanks for sharing.

I’m also kinda reminded of aws lambda URLs, or I guess URLs in general.

Edict: all future library names must be UUIDs?

I’ve been really distracted lately thinking about what things need named handles in my data models/representations, and what they should be handles for, and how to do so, and how to change them in sane ways. So far all my thinking has just convinced me that I am a mediocre architect.

Piling on, I’d also recommend against a 15 hour workday.

I’m much more cynical:

1. It was probably a tax write off.

2. It was probably also both a community service project AND a reflection of lower expected insurance payout.

3. Maybe the straight-A students are correlated with insurance payers that would rather pay the repair known 1-off payment (or skip doing anything at all) vs the unknown recurring increase in insurance payments?

(Source: drove a dented-up pickup for over a year after a short romp along/into an orchard)

Maybe a digital voice recorder thing (like [1]) so neither of you has to take explicit notes but can still capture all the discussion?

[1]: https://www.amazon.com/Sony-ICDPX370-Digital-Recorder-Built/...