madisp

I'd argue that the simple `deleted_at IS NULL` check is not broken - unless your product / domain specifically allows and requires scheduled future deletions adding such logic can easily introduce bugs. For example, you could to get the comparison flipped by accident, and if it's only in one place out of many that bug could go unnoticed for a while.

perhaps instead of nefarious usecases it could just be out of date

Album listener here too. I really liked the "Shuffle albums" feature in iTunes so I built a tiny webapp to give me a random album from my library - https://shuffle.ninja

probably the Lex Fridman podcast interviews:

https://www.youtube.com/watch?v=2BdBfsXbST8 https://www.youtube.com/watch?v=EE1R8FYUJm0

the second one is definitely where the last paragraph in the article is from. Weird that the interview is dated 2021-09-09 and the post is 2020-04-30?

sound samples are often loaded into RAM and can easily cross into 10+GB

what if the tea.xyz domain is deleted? or the site serving it is down? even if it's using something like AWS there is some downtime there..

> tea’s decentralization offers tangible benefits to ecosystem security. Every layer of your apps and dapps is signed and verified on‐chain. Users can rest assured that the software they are using is what all its creators intended. In day to day development our decentralization increases reliability (leftpad is a sore memory) and gives developers the flexibility they need to turn inspiration into empires; when a single proprietry vendor controls how your stack is composed and deployed your stack is centralized and under their control.

and then

  sh <(curl tea.xyz) install deno.land

?

if you don't mind a soldered-on header then berrybase seems to have plenty of Pi Zero WHs

But there is a dislike button in Spotify, specifically in the Discover Weekly playlist where most of the recommendations happen?

https://community.spotify.com/t5/Implemented-Ideas/Discover-...

Ideally you'd use something modern that invokes your function every hour (cron? :P) so that the rescheduling is detached from the function. I think if generation takes X hours of raw CPU computation where `X >= 1` then as long as you've got C cores and `C > X` you should be OK?

The 3600 second + a bit (~1s) is a common thing that happens if your periodic scheduling work is one-shot and repeating is achieved through scheduling the task again after completion. E.g. consider the following code (in Kotlin, but hopefully still readable):

  fun rotateKeys() {
    publishNewKey(key = generateKey(), timestamp = now())
    schedule(::rotateKeys, 1, TimeUnit.HOUR)
  }

if `generateKey` and `publishNewKey` take around ~1s then you'll observe exactly this behaviour - the timestamps will start drifting from some original value.

IIRC the JVM optimizes very aggressively based on runtime profile, and just adds traps to the optimized-out paths to collect new runtime profile.

Imagine code like this, where at runtime foo is always null

  if (foo != null) {
    foo.bar();
    foo.baz();
  }

if foo is always null the compiler will just remove all of the statements inside and places a trap there. If foo happens to be non-null (say, due to external environment changes) then the it hits the trap and the whole method is deoptimized and goes through the profiler again.

The optimizations in the JVM are quite fascinating - https://advancedweb.hu/profile-based-optimization-techniques... I imagine javascript engines employ similar techniques

they failed to mention that malicious code was reading plaintext passwords though

this looks awesome, thanks!

It looks neat but I wish it had physics calculations - I want to be able to write something like

10m * 1000kg * 9.8m/s^2 in kWh

and see a correct result

already true for some models with some features like Nissan Leaf with the remote control app - https://www.reddit.com/r/leaf/comments/4i4f3s/how_do_i_get_r...

But can't you take any MIT or Apache (or similarly permissively licensed) project closed source anyway?

Sadly the data is useless without the exact compiler flags.

I get 18.0658 with default "g++ PrimeChecker.c -o PrimeChecker" and 11.7689 with "g++ PrimeChecker.c -O2 -o PrimeChecker"

edit: the hardware is an i7-4870HQ, quite close to the i7-4850HQ that's in the blogpost (+200MHz clock)

IANAL, but doesn't MIT require you to keep the copyright notice though? I mean you're not waiving away your copyright by licensing something with MIT.

Any reference on it being "generally considered less developer friendly"?

Actually resources.arsc only contains the resource table (think stuff in values/ folders, like strings.xml). The layouts still have their own xml files in res/layout. It is in Android's XML chunk format though, not plain-text xml.

Interesting point - the resource table still has references to the XML files for configuration-resolving purposes, but they are only strings. E.g. getString(R.layout.main) will actually work and return "res/layout/main.xml" or "res/layout-land/main.xml" depending on the device configuration (and assuming one has the layout-land version). Vice-versa, if you have a string resource with the value "res/layout/main.xml" and say, name foo, then calling setContentView(R.string.foo) actually works.

I don't think Slack is a HTML5 app, at least on Android. The apk contains ~2.5MB of bytecode and half a meg of layouts.

Pointers.

I wouldn't ever say that using floats compared to, say, precision decimals in any language is leaking an implementation detail.

As a user of a library/API I need to know if I can rely on it being precise or not.

For fonts, what really helped it for me on Linux was copying the font and removing all hinting information (as outlined here - http://superuser.com/questions/614960/how-to-fix-font-anti-a...)

API is http instead of https and doesn't require signing the request body or parameters :(

Google doesn't update the OS for other manufacturers... Look at it this way - if I fork the rails repo it is up to me to pull/merge the latest changes, nobody will be pushing updates to my fork.

I'm not sure even encryption helps if you only have a locking screensaver. Both TB and FW devices have DMA so if somebody really wants to get in then the password-protected screensaver lock isn't going to stop anyone (given that the encrypted volumes are mounted).

12 was released a year ago, thats their model (and version number, 12 == 2012, 13 == 2013, etc)

Clicking on alpha/deploy leads to 404 :(