Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
viscanti
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
49 ms
·
511.
▲
by
viscanti
14y ago
Binary Large OBject
512.
▲
by
viscanti
14y ago
SoMa is the only neighborhood with any 20+ story residential buildings going up. It's hard to say why so many of the other neighborhoods are against building up. The geography of SF lends to some beautiful views from atop the hills. Resi
513.
▲
by
viscanti
14y ago
What about the well-intentioned but dubiously helpful PR from someone who just doesn't know what they're doing? What if that were to catch on and you have swarms of overlapping non-programmers submitting PRs? These slippery slope arguments
514.
▲
by
viscanti
14y ago
I guess I don't see the harm in spending a little extra time up front to design things properly rather than just jumping in and writing code, hoping to stumble upon the correct design pattern. I've found it much easier to find and solve pr
515.
▲
by
viscanti
14y ago
I guess the issue here is that it's illegal for the taxis to pass up fares based on where the passenger is going. The fact that it happened 5 times in a row is telling. The taxi industry is the exact group fighting for more regulations in
516.
▲
by
viscanti
14y ago
I've read a number of statements from Uber, and talked to every driver I've had. My understanding is that they all are required to be licensed (and in fact are already limo drivers who are filling free time). I haven't run into a driver y
517.
▲
by
viscanti
14y ago
Is there any evidence Uber is using ANY drivers who don't have the proper licenses to work as commercial drivers in the cities they work? You're suggesting something that I haven't read anywhere. What common good is being protected by requ
518.
▲
by
viscanti
14y ago
The "approved meter that others were required to use" was for taxis, not limo/towncars. Their GPS units are significantly less opaque than the status quo for limos (which really comes down to just naming an arbitrary price, or having to pa
519.
▲
by
viscanti
14y ago
Where have they ever said they shouldn't have to play by the same rules as everyone else? They vet their drivers and ensure that all meet the requirements to drive locally. They're using licensed, commercially insured drivers who meet all
520.
▲
by
viscanti
14y ago
Arbitraging WHAT regulations? They use licensed limo drivers, who are already out driving every day. They're simply helping those drivers fill their excess capacity. Limo drivers have been driving legally, under as much or more regulatio
521.
▲
by
viscanti
14y ago
They DID dig into their own pockets to pay drivers the 2x price. Their "surge" pricing was in effect for a very short time. It's been reported that they paid over 100k out of pocket to help boost supply. They're a young, inexperienced co
522.
▲
by
viscanti
14y ago
Manhattan also has significantly less potential for destructive earthquakes than San Francisco, making it much easier to build tall buildings. It's not just a matter of being bordered by water. A big part of it has to do with no one wanti
523.
▲
by
viscanti
14y ago
This doesn't seem to be an issue of regulatory paperwork though. What does double licensing accomplish and what does it actually mean? I'm all for fines if they're found using unlicensed drivers. I guess I don't know that it's necessary
524.
▲
by
viscanti
14y ago
What does the licensing accomplish? Are drivers double drug tested? Are they double insured? They're using all licensed drivers, which means any arguments of "unfairness" are baseless. They aren't benefiting from lower labor costs becau
525.
▲
by
viscanti
14y ago
Because no good business/technical strategist has ever overseen a complex software project that had a bug in it?
526.
▲
by
viscanti
14y ago
I think part of the "retina" branding was that pixel count doesn't matter beyond that point. It's indistinguishable to the eye, so beyond that point more are not better. It's debatable if that's actually the case, but it's an implicit par
527.
▲
by
viscanti
14y ago
The argument isn't "If it isn't broke, don't fix it". It's that not all change is necessary (or even good). If something is clearly broken, by all means fix it. If it's not, then things get more complicated. Maybe a change is still in o
528.
▲
by
viscanti
14y ago
I had a chance to hear one of the contributors to Golang give a talk about the state of the language at Google. It sounds like the team has actively shifted (for the time being) from developing the language to building things with the lang
529.
▲
by
viscanti
14y ago
I think the difference is between things that work don't need to be changed, just for the sake of change. Things that are broken, or suffer from poor usability should change. Not everything needs to change, but I think it's fair to want b
530.
▲
by
viscanti
14y ago
I would guess that the major alternative for most of the applicants for this program would be studying on their own. You can take as long as you'd like to learn. Doing that, you can spend as much time as you'd like with friends/family, do
531.
▲
by
viscanti
14y ago
It wasn't the copying, it was the culture of AB split testing everything that killed them. AB (or multi-variate) testing can help improve an already successful product, but I'm not convinced it can build a good game from scratch (or from c
532.
▲
by
viscanti
14y ago
Flake8Lint (available for Vim and Sublime Text 2) works really well for me, and my co-workers. The optional complexity checking is a nice bonus. You also get the PEP 8 linting and some logic checks. It hurts at first, to see all the err
533.
▲
by
viscanti
14y ago
This is a good heuristic. For me, single letter variables are almost always iterators in places where space is at a premium (i.e. one line list comprehensions in python). If I'm breaking a loop out over multiple lines (which is almost alw
534.
▲
by
viscanti
14y ago
Haters are going to hate. It doesn't matter what field/discipline you're in. JavaScript is as much a proper programming language as Python or Ruby or Lua. They're all high level scripting languages that do roughly the same things. Only
535.
▲
by
viscanti
14y ago
This would (possibly) be true if all mistakes were weighted equally. As soon as some mistakes have more weight (like maybe the machine killed you on accident because it thought your spleen was in your ear), then looking at aggregate number
536.
▲
by
viscanti
14y ago
I met Ev and Taylor a couple months ago. They're a brilliant team, in a not-so-glamorous market (there's not a lot of buzz around sending mail). These guys are committed to building a reliable/developer friendly email service though. The
537.
▲
by
viscanti
14y ago
Once again: neither am I. You're arguing that EVERYONE's opportunity cost should be factored in. I'm saying that regardless of status (Founder, employee, etc), that opportunity cost is a sunk cost, and examining it with respect to determi
538.
▲
by
viscanti
14y ago
I think any other analysis is too complex to offer any real insight. We can look at the NPV of the possible future earnings, to get a better idea. For the given opportunity cost, what's the impact on future earnings (over the next 10 year
539.
▲
by
viscanti
14y ago
"Sunk Costs" ARE part of the discussion you're having. You're just choosing to include them when they probably shouldn't be. They aren't relevant to whether or not an aqui-hire is a success or failure, mostly because they require us to br
540.
▲
by
viscanti
14y ago
If sunk costs (based on missed opportunity cost) factors in, then you also need to know what future earning opportunities would be for all (or most) founders/employees had they taken a job in Big Corp vs working at the startup that was acqu
More ›