fyp

Hmm why not sell everything and hold cash if you just want to keep everything at 0?

You're just paying fees for no reason. Is it just some tax trick to keep long term capital gains for the index funds? (does it even work like that?)

They do expose some limited interfaces that are pretty easy to browse. For example:

https://www.google.com/settings/ads

https://myactivity.google.com/myactivity

https://www.google.com/maps/timeline

https://www.facebook.com/off_facebook_activity/

I highly don't recommend buying from them. I bought a steelcase leap v2 in platinum color and it's clear that the base of the chair is a cheap knockoff.

Crappy silver spraypaint that flakes, the wheels barely work, the piston was broken on delivery (which they did replace for free).

I've had problems with the chair being too low at max height and I am a short guy. I still don't know if it's their cheap base that's the culprit or that's just the design.

I thought half life alyx wasn't released yet?

Yes they do mutate and there are really pretty graphs for it: https://nextstrain.org/ncov

Why the fuck are government websites like coveredca.com sharing information with facebook? I don't use facebook logins with it.

  {**d1, **d2}

is very natural if you also write javascript where their spread operator looks like:

  {...d1, ...d2}

Did anyone else find their "eight failure modes of the apocalypse" incredibly arbitrary given the important sounding name?

I am currently evaluating whether I want to use the generators for a typescript node.js/react project.

Which typescript generator are you using? Do you use typescript on the backend too and if so is it easy to keep the generated type definitions in sync? Also for backend what do you use for json schema validation?

I read until the end and it turns out to be a nonconstructive proof via the probabilistic method!

https://en.wikipedia.org/wiki/Probabilistic_method

Another real world example I saw recently was in programming competitions.

There's a tradition of displaying a color for your username based on your ELO rating (for example people would brag about being Red or Yellow on topcoder).

So people tend to stop competing forever after hitting the next level just to preserve their color!

Graph distribution of ratings: https://codeforces.com/blog/entry/52470

Should also search HN for "piercing the corporate veil" to see the numerous past discussions on the caveats.

https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

My google ads profile correctly identifies me as male: https://adssettings.google.com/

I imagine the high level business use cases for these machine learning APIs are similar. For example, analyze shopping mall camera footage to figure out demographics shopping there and coordinate the (socially accepted and segregated) men/women department stores accordingly.

This is such an profitable use case I cynically can't see why they would cripple their API like this, regardless of their ethical stance.

I can kind of make sense of why humans(animals?) evolved to see brightness in relative terms. The same object will have different amount of light reflecting off it based on the time of day so you don't want your brain to register those as different scenes.

Hue also makes sense. It's the distribution of wavelengths being reflected.

But what does saturation correspond to? When do things become saturated/desaturated in the natural world?

The reference links are also pretty cool: https://michaelbach.de/ot/mot-pigeonNeck/index.html

> If you always had to be 95th percentile in your current peer group (general couch potatoes, high school championship participants, world championship participants, Olympians, Olympic medalists, Olympic champions, Olympic champions twice in a row), then the 95 just doesn't mean anything. Then you just have to be the single very best according to your logic.

This is the typical experience for the ivy league types common on HN though.

They had to beat their way into the 95% of their zone school, then 95% of their magnet high school, then 95% of their elite college's graduating class, then 95% of their big N corporate ladder, etc.

Despite the filtering at every step of the way, the curve is always bellshaped! And it feels like shit if you find yourself on the left side of that curve among your peers.

But I guess it's healthier to think along the lines of: "What do you call a medical student who finished last in their class? Doctor."

I wonder why OrderedDict isn't outright deprecated (or reimplemented as a trivial wrapper over dict).

You can get the first/last key of a 3.8 dict with dict.keys() and reversed(dict.keys()) which you can then delete to reimplement OrderedDict.popitem.

Deleting a key and reinserting it will let you reimplement move_to_end.

The only cool thing that OrderedDict's implementation might be useful for is to move to front (or anywhere else not the back) but it doesn't expose that api.

I wonder if dropbox might have more photos data than facebook. People only share a small subset of all the photos they have taken and have backed up. I can see that being valuable. Otherwise I am not seeing why they would want to acquire them.

I am having the same problem and I think I am going to go with json schema.

The problem is that a lot of the things you want to validate aren't easily expressible as typescript types (e.g., valid email address, make sure two fields are always the same length, etc). If json schema are more expressive you want to use that as your source to generate the typescript interface instead of the other way around.

It's a tradeoff. In libc++'s version you still have string size stored in the top 7 bits so you just need a bitshift to get size. It sounds like fb's implementation would require looping until null terminator to get the size.

I thought the example in the article was a little artificial at first. Like why would you only have 12 requests if you have 12 backends?

Reframing it in terms of capacity cleared that up. If the rate of incoming requests is higher than the total rate your backends can process, your queues will grow infinitely!

So something like an average of 12 incoming requests per second with each backend capable of processing 1 request per second is actually fairly realistic. And I think the math still works out the same there.

The precise statement is that you need to have O(2^(N/2)) items before the probability of finding a hash collision is greater than 50% (or whatever nontrivial percentage). English is hard and I think whoever cares about the details can look it up themselves.

Birthday paradox is way too important to omit from this list!

For example if you're uploading stuff to a bucket, you can compute its hash first to figure out if a duplicate already exists and if so, skip the upload.

Why can you do this? What if it was just a hash collision? Shouldn't you still compare the contents to really make sure they are the same?

Turns out if your hash function is N bits you will need to have 2^(N/2) items before you see two hashed to the same thing by chance. If you choose a 256 bit cryptographic hash function like SHA256, that's 2^128. This probability is so low you have a higher chance of encountering a cosmic ray bit flip!

https://en.wikipedia.org/wiki/Content-addressable_storage

https://en.wikipedia.org/wiki/Birthday_problem

The dropbox cli already lets you ignore folders. I've have had the following script for many years to ignore node_modules:

  exclude_folders=$(find . -type d -name "node_modules" | grep -v "node_modules/")
  echo "Excluding $exclude_folders"
  dropbox exclude add $exclude_folders
  dropbox exclude list

The feature I want is pattern-based ignore in a .dropboxignore file.

It's an interesting idea. You can make two funds sum up to a constant by taking opposite and equal positions.

But I am not seeing how you can control the direction the money moves. For example if you knew fund A will gain $x and fund B will lose $x, shouldn't you simply not make the trade in fund B?

> 12 years, 5 months at Google

> many Google-internal CLs ("change lists" == commits == PRs)

> 3,064 Android CLs

> 10,787 Go CLs

This is at least an average of 3 commits a day (4 if he didn't work weekends). How is that possible?

I wonder if you can make the same demo with real images using seam carving.

See the example images in wiki: https://en.wikipedia.org/wiki/Seam_carving

Or online demo: https://alexander.soto.io/seam-carving

In case people read the comments first, this is about postgres data types.

It's pretty comprehensive and I especially liked the section about how datatypes are implemented internally (pg 25). I wish it went into more details though.

Is that how languages that always give you zeroed arrays (e.g., java) do it?

I remember in an intro bio class we were shown simulated animations of how reactions occur. For example it depicted molecules floating perfectly and almost intently into an enzyme demonstrating the lock and key model.

The teacher made it point that the animations were not realistic in the sense that there should be a lot more stuff flying around. And a lot more denser and faster otherwise reactions simply wouldn't occur. But that chaos isn't educational to visualize.

So I guess my question is why does it look so neat in this real life video?