Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
exizt88
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
14 ms
·
121.
▲
by
exizt88
11y ago
Please don't hijack scroll on web pages.
122.
▲
by
exizt88
11y ago
Huh, you're right. I disagree with the author, then. > They’re like the coffee machine on the office: you press the button, delicious coffee magically appears in your cup But even office coffee machines usually have some level of co
123.
▲
by
exizt88
11y ago
I don't think the author ever said that you shouldn't be able to configure an application. It's more about proper default configs. Default configs are, incidentaly, an important part of the Unix philosophy of "program
124.
▲
by
exizt88
11y ago
I'm sorry, but did anyone else find the article extremely hard to read due to the box to the right with the comments and sharing button?
125.
▲
by
exizt88
11y ago
Amazing. Can you change the speed of the simulation?
126.
▲
by
exizt88
11y ago
I've tried overthinking FizzBuzz myself, but from the architectural point of view: https://medium.com/@sbichenko/the-wrong-fizzbuzz-81cb8e67ef4...
127.
▲
by
exizt88
11y ago
Use the Chrome's "Profiles" tab in DevTools.
128.
▲
by
exizt88
11y ago
100 ms delay on every click (measured in Chrome's Profiles tab). I'm on a 2013 MBP. Seems a bit too much for making a number bold.
129.
▲
The Wrong FizzBuzz
(medium.com)
3 points
by
exizt88
11y ago
|
0 comments
130.
▲
by
exizt88
11y ago
https://www.google.ru/webhp?sourceid=chrome-instant&ion=1&es... He really does like that story.
131.
▲
by
exizt88
11y ago
I don't think Tufte ever uses top and bottom borders for tables. Also, the delimiter below the heading should not be so bold; it's usually a very thin line.
132.
▲
by
exizt88
11y ago
A comparably horrifying experience: https://en.wikipedia.org/wiki/Clive_Wearing
133.
▲
by
exizt88
11y ago
I found Node.js to be quite difficult to work with, mostly due to nature of Javascript. Module ecosystem is vast, but each one has an API written in a different style. Combine that with no type-checking and lack of robust IDE auto-completio
134.
▲
by
exizt88
11y ago
I did register for Workshape, though. So, I guess, your little project served its purpose in my case after all :-)
135.
▲
by
exizt88
11y ago
What's the point in the slider? Just show all 5 variants at once. Also, great idea with the "Keep track" button at the end. Registration (and therefore conversion) becomes a part solving the user's problem ("How do
136.
▲
by
exizt88
11y ago
If they have 8 hours of work time and finished their tasks in 6, does that mean they have exhausted their usefulness to the business? Why not spend these 2 hours doing some small miscellaneous tasks?
137.
▲
by
exizt88
11y ago
OOP isn't perfect, but I don't think this post addresses any actual problems. > Since functions and data structures are completely different types of animal it is fundamentally incorrect to lock them up in the same cage. That&#
138.
▲
by
exizt88
11y ago
Your whitespace is all over the place. Try to keep things more bunched up. Like this: http://i.imgur.com/LaR1gaH.png
139.
▲
by
exizt88
12y ago
Genestack - www.genestack.com - UK/Russia - Software Engineers - Fulltime We provide a format-free, web-based platform for bioinformatics research. Work with us on changing how bioinformatics applications are built. BIOINFORMATICS APPL
140.
▲
A Markov-powered mashup of Erowid trip reports and tech recruiter emails
(twitter.com)
8 points
by
exizt88
12y ago
|
0 comments
141.
▲
by
exizt88
12y ago
> Plus, a built-in torchlight provides convenience, whether it’s late at night or after you’ve turned off the lights. So that's what a torchlight is for! Thanks, landing page!
142.
▲
by
exizt88
12y ago
'.' is the proper string concatenation operator in PHP.
143.
▲
by
exizt88
12y ago
> $then = "email" + "@" + "davidpennington.me" Is this supposed to be PHP or Javascript?
144.
▲
by
exizt88
13y ago
> The per-request execution model can be played as an incredible strength, but the moment some framework comes along, starting off by loading all of its code, then instantiating a myriad classes, subclasses, traits, and other crap just t
145.
▲
by
exizt88
13y ago
Not bashing PHP in any way (I'm a PHP developer myself), but what would you consider the PHP paradigm?
146.
▲
by
exizt88
13y ago
From my point of view, you haven't worked with good designers at all. All the good designers I know work hard on their research (both theoretical and concrete, regarding their current projects) and user testing afterwards. In many comp
147.
▲
by
exizt88
13y ago
> How much is that 3 pixels really worth? It's the developer's responisbility to estimate how much work a particular task requires. If it's too much work for too little benefit, no sane manager/designer would insist o
148.
▲
by
exizt88
13y ago
The problem of engineers designing UIs is the topic of one of my favorite books: http://www.amazon.com/The-Inmates-Are-Running-Asylum/dp/0672... The problem is that engineers are not very good at UI design. Both f
149.
▲
by
exizt88
13y ago
Sorry about that. Your post got me all worked up!
150.
▲
by
exizt88
13y ago
If it's such a big deal, just say no. I think the article would, too, argue "why you shouldn't move that button 3px to the left if it makes you rebuild the whole screen".
More ›