|
Durchlass Kilometer 5,698
@
emareaf
Suebia
|
|
Bio can't include '' ''.
|
|
|
13.685
Tweetovi
|
34
Pratim
|
592
Osobe koje vas prate
|
| Tweetovi |
|
Durchlass Kilometer 5,698
@emareaf
|
5 h |
|
|
||
| Durchlass Kilometer 5,698 proslijedio/la je tweet | ||
|
Durchlass Kilometer 5,698
@emareaf
|
30. kol 2015. |
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
21 h |
|
Betelgeuse might have planned ahead
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
21 h |
|
Have a look at meaningness.com
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
3. velj |
|
Also some arguments are like the car turned because the steering wheel turned vs. the car turned because the driver wants to exit the freeway.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
31. sij |
|
Depending on my day and whom I talking to what I mean is in the yellow column or in the pink one.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
31. sij |
|
Rumor has it that men who are in a relationship become more attractive to other women. People be talking about social proof but what if it's just that being in a relationship makes you straight up a better person
|
||
|
|
||
| Durchlass Kilometer 5,698 proslijedio/la je tweet | ||
|
Adam Strandberg
@The_Lagrangian
|
10. sij |
|
the sorites problem doesn't hold if instead of a grain we add a heap's worth. but then, if we remove just one grain from the amount we're adding, it still doesn't hold. by induction, the sorites problem doesn't hold for the 1 grain case
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
30. sij |
|
Evolution can be outsmarted, it can't be outselected.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
30. sij |
|
Are they misleading without the later parts? Otherwise it seems best to publish them.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
To create a good thing you have to be able to imagine how it can be surpassed and be okay with that.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
A cordless drill is an intensible tool because it only got a hole that spins in which you can stick tools that do useful things when spun.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
The map and filter functions in functional programming languages are intensible software.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
Extensible software does one job and lets you stick another piece of software onto it so that it can do a job and a half. Intensible software does half a job and has a hole in which you can stick another piece of software so that it can do a whole job.
|
||
|
|
||
| Durchlass Kilometer 5,698 proslijedio/la je tweet | ||
|
Person with a Computer
@trytoprogram
|
23. sij |
|
The design advice that there should be zero, one, or an unbounded number of anything is not very good advice - it turns out everything has an arbitrary size limit, and if you don't declare it yourself, nature will declare it for you.
|
||
|
|
||
| Durchlass Kilometer 5,698 proslijedio/la je tweet | ||
|
Tom MacWright
@tmcw
|
29. sij |
|
one of the themes in 'on the metal' is that programmers today don't know how their software really runs, or don't care. i think the corollary is that it's just way harder to know, now that all the dumb parts of the stack are smart and make their own sneaky decisions.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
Quality is the result of deciding from the beginning what the limits of what the thing will ever have to do are.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
Only have a zero, one, or infinite amount of things yada yada but when you know that you never have more than a thousand sprites on the screen you can make reliable calculations about which performance budget you have for each sprite to consistently get 60fps.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
Premature optimization yada yada but when you know this is as fast as it gets you can make calculations and do actual engineering.
|
||
|
|
||
|
Durchlass Kilometer 5,698
@emareaf
|
29. sij |
|
"write it in a slow language, rewrite the performance critical parts in a fast language" is a bit of a trap because it makes it more difficult to predict and therfore plan around performance.
|
||
|
|
||