> Along similar lines, I've wondered about how often restaurant and bar owners have bugged their own premises.
In a few select Houston strip clubs.. all the time, or at least ten years ago some did, the ones owned by certain organizations. I haven’t been around that scene for a decade, so I can’t speak to now. I know of a few other places that seemed to magically never have trouble with city officials or permitting. Another such place, an after hours club, was frequented by the mayor’s “party-oriented” daughter, never had trouble with police raids, or fire marshals. The venue survived unscathed until the next mayor took over and such leverage became unavailable. The Texas Alcoholic Beverage Commission was the only real challenge, but often the local cops would be knowledgeable about pending raids and would graciously provide some advance notice. It certainly helped that many of those cops were paid as off duty private security by the venue. DEA was another frequent adversary, but those folks aren’t as undercover as they thought they were. I might suggest that the DEA was (or maybe still is,) one of the more sketchy law enforcement agencies in the government.
Just under the surface of “normal,” there is some really fascinating stuff that goes on.
Not sure about the latest PSUs, but in the past high efficiency PSU's had to make trade offs with increased ripple/noise so they weren't ideal where maximum stability was desired.
My favorite question to ask in software engineering interviews is one that I believe to be un-burnable.
> It's 2140 AD, New York is under water up to X feet high. Buildings have been retrofitted with <magical-ish material> to withstand the water. You are in charge of keeping your building dry. If water gets in and damages the foundation, a few thousand people die or become homeless.
> Design a system that ensures that doesn't happen.
How candidates approach this, how they think about redundancy, how they deal with additional constraints or extra scenarios thrown at them tells a lot about how they approach things. The question is not about software (explicitly) on purpose so that it gets people out of the coding mindset.
With more software-focused systems you always get candidates who starts writing code and designing objects and classes and stuff. No, I want you to design a system. Stop writing code.
The reason this question is unburnable is that there's no right answer. You're being asked to show how you work through an abstract problem and design a solution. You aren't being asked for an answer.
So far everyone who did well on that question has turned out to be a great engineer. Whether they were fresh out of college (and learned a lot fast) or they were already experienced.
On the standardization issue: the advantage of such a method that we presented is that as long as there exists a standard for model specification, we can encode every image with an arbitrary computational graph that can be linked from the container.
Imagine being able to have domain specific models - say we could have a high accuracy/precision model for medical images (super-close to lossless), and one for low bandwidth applications where detail generation is paramount. Also imagine having a program written today (assuming the standard is out), and it being able to decode images created with a model invented 10 years from today doing things that were not even thought possible when the program was originally written. This should be possible because most of the low level building blocks (like convolution and other mathematical operations) is how we define new models!
On noise: I'll let my coauthors find some links to noisy images to see what happens when you process those.
I did some reading of the PNG specification and it may be possible to take advantage of Adam7 interlacing (8 passes total) and scanline filtering (several methods) to write out an image where the scanlines of 'known to be blurry' areas contain only 1 pass of image data and 7 passes of highly-compressible scanlines with filters that generate blur at decode-time.
Doing this formally at scale would require the compositor and the encoder to cooperate, as the encoder would benefit greatly from having access to both the 'blurred' and 'unblurred' areas without the blur filter having been applied to the former, as it could then construct low-fidelity, low-bandwidth, visually pleasing blur for the 'blurred' segments.
This exceeds my ability to write PNGs by hand and it certainly exceeds the bounds of what most people think 'an encoder' should be capable of doing, but at least it presents a path forward. I'll post to HN someday if I ever somehow manage to do this.
I only pay my accountant around $2000 to handle a bunch of housekeeping at the end of the year that I'm to lazy to do during, and to file my taxes. For years I did my own taxes using Turbo Tax Self Employed and Turbo Tax Business Editions, but in 2016 I decided to outsource all of that headache, the first year, my taxes were half of what they were the year before with similar income. Then after the TCJA in 2017, my taxes were only 1/5th what they had been before the accountant, with close to double the raw income.
The biggest changes from doing it myself: I got an accountant; I incorporated (s-corp) instead of doing self-employment income; I draw W2 income now (using Gusto); I set up a SEP-IRA and put 20% of my W2 earnings as an automatic debit from my business account (in Vanguard); TCJA chops off 20% of my distributions on my K1; and I'll state again, I hired an accountant.
An inside look into the operation of the Supreme Court is Bob Woodward and Scott Armstrong's "The Brethren" [1], which covers the years 1969-1975.
There's a surprising amount of persuasion, negotiation, and coalition-building in order to find the votes to grant cert or hold a majority. It's interesting to see how the interpersonal dynamics among the Justices influence decisions.
An example of this is looking for unanimous decisions to send a strong message, whether it is for school desegregation:
> Since 1954, the Court had always been unanimous in school cases, its strong commands to desegregate joined by every member. For fifteen years, the Justices had agreed that it was essential to let the South know that not a single Justice believed in anything less than full desegregation.
or to reject Nixon's executive privilege claims in Watergate [2]:
> The Nixon challenge had to be met in the strongest way possible. An eight-signature opinion would do it [...] The country would benefit from such a show of strength now.
These decisions required significant reconciliation between the original ideas of the Justices. They don't operate in a vacuum of law study. The Muhammad Ali draft dodging case [3] also demonstrates this, where a 4-4 deadlock is turned into a unanimous 8-0 decision centered around a technical error which would not set precedent - the book describes Burger as considering that "it might be interpreted as a racist vote" if he dissented.
Like the link, I am also curious about why these delays have occurred, especially the most recent one. Wildly speculating, maybe the Court does not want to appear too politically activist in this moment; or maybe the members are trying to find consensus for their preferred outcome before granting cert at all. One other option, though I hope it's not this one:
> [Stevens] said, it was "pointless" to grant cert only to have the majority reaffirm its well-known view [regarding Liles v Oregon, an obscenity case].
> breaking out of that hyper-focus was like killing my dog
That's such good way to describe it for those who just don't experience it.
It's like if you had just reached the decision logically to turn off life support for someone you love. It's not like you just straight away walk over and switch it off. There's this awful reluctance that you have to go through first.
Medication sounds like quite a mixed bag, but I'm feeling somewhat more willing to give it a try.
It took over a decade, but I finally realized that this is the original purpose of the dialectical journals I was required to write for my high-school English classes, before they were coopted as an assessment tool. As it’s a record of your own thinking about the various things you’ve read, it provides a way to figure out where ideas came from in a way that full-text search can’t handle.
These days, I use a traditional index card system combined with a spaced-repetition flashcard system. Each concept that I feel like collecting information about has a small stack of cards that starts with a 1- or 2-sentence definition, in case I’m using a term slightly differently from other people. Below that are cards that represent quotes, notes, summaries, and recommended reading on that topic.
The flash card deck is to make sure I know what things exist, and what words are used in the literature to refer to abstract things. If I need more detail than I can remember, those keywords will get me to a relevant wikipedia page, at least.
As someone going through this right now, here's some things I do to make the process easier.
1. Put together a ~medium detailed listing of what work needs to be done, broken down into sub-tasks.
2. Include estimates for how long each sub-task will take.
3. Roll that up into a gantt chart with dependencies reflected.
4. Identify opportunities for parallelization so you can answer the question of "will adding more folks make things go faster?"
5. Identify and be prepared to speak to which portions of the schedule represent padding, and explain why it was padded in that way. Share your thinking in the trade-off you made and be open to adjusting -- and de-risking the schedule in other ways.
6. Realize this is a bit of a negotiation, not necessarily you vs. them but more you and them vs. what you can deliver. In a negotiation it's important to not appear intractable. Have things you can give away.
7. At the end of the day the engineers are writing the software. If you're not actively dragging your feet, what you say does go, so it's a question of where you're spending your time, and on what. With that in mind, your boss has a job to do too, so help give them the tools they need to do so.
8. Don't burn bridges, y'all have to work together.
There are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies and the other way is to make it so complicated that there are no obvious deficiencies. — C.A.R. Hoare, The 1980 ACM Turing Award Lecture
I was recently talking to a registered psych nurse, and we got talking about Cognitive Behavioural Therapy.
I believe I've been self administering a form of it for a couple years, and I summarized my understand of CBT as "moving more thinking from the amygdala to the prefrontal cortex", and she confirmed that with "in laymen's terms; yes".
It's not like the fields are completely isolated, I guess is what I'm getting at with that anecdote. It's hard to go from neuroscience to psychology, but that's always being looked at. I reckon most big advancements will start coming when we start understanding the connectome more, but it's not like all advancements will come from there, and it's not like people aren't working right now to bridge neuroscience and psychology.
Sounds like a bank, a government agency, or any business over a certain size.
I tell people you only call it politics when you are losing. More accurately, it's a layer of literal stupidity above the competent to shield the money side of the company from the leverage that operations people would have if they had any information about how the money side worked.
Instead of a hierarchy, rethink a company as a hub and spoke model with concentric rings. The main differences are the implication in a hierarchy that there is "gravity," keeping people down and that they need energy and leverage to climb "up," which further implies there is a place to "fall," and that there is only one way "up," instead of many possible paths to the centre from all directions. There is no gravity, only gates and barriers, and even these are just information. Politics is how a middle manager runs interference and creates distractions to make sure you can't see over, around, or through them, and that the people behind them closer to the money can't see you. Tech is usually outside the main perimeter, mediated by contracting companies or middle managers whose job is to compartmentalize the value people create, and be sure it is replaceable.
Viewed this way, of course this demented political farce is how Apple works, because it's how everything seems to work when you have internalized the precise and specific mental model someone uses to take advantage of you.
Sorry if you can't unsee it now, but hopefully it will be funny and we can get good, competent people who value tangible skills into positions of power.
I think a summary of it is that some things are broken by design in a large org.
If you've worked in a huge org like Apple or the government, you've seen incredible inefficiencies. How can all these smart and often well-paid people be doing something so wrong (i.e. wasting time with political battles, etc.)?
But that's only true if you look at the "medium view" of the organization. If you look at the large view, how the money actually flows, then it might be beneficial for one part like IS&T to be kind of broken, as long as the rest of the company works.
I have seen this dynamic at other companies. Internal tools can sometimes hold too much leverage over the organization. They can almost "blackmail" people into getting their way because they have a literal monopoly over what they do. It might be better to let multiple teams fight battles amongst themselves, which seems inefficient if you're working as a regular joe, but could be efficient from the CEO's perspective.
----
Edit: A related idea is the principal-agent problem:
e.g. if you've worked at any large org, you know that not everyone acts in the company's interest (nor should they strictly speaking). A trivial example of this is putting extravagant meals and vacations on the company's card. (Although the nontrivial instances of this go all the way up to fraud/embezzlement.)
A less trivial example is making up fake engineering work that sounds good when review time comes, but doesn't actually advance the company's business.
So some things that seems mystifying on the ground level are pretty much a defense against this inherent conflict of interest.
In an adjacent domain, a lot of HR is a defense against employees acting in their own interest. (I see a lot of veterans on here saying: "remember HR doesn't exist for you", which is presumably a hard won experience.)
