Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
joemag
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
31.
▲
by
joemag
8y ago
Sixty Symbols, which is a great channel for these kinds f topics, did a great video on primordial black holes a few years ago: https://youtu.be/gs3mtZPySeM
32.
▲
by
joemag
8y ago
I had to pause for a second to think about what’s so hard about this? And then it hit me - negative numbers.
33.
▲
Seattle Is Dying
(komonews.com)
22 points
by
joemag
8y ago
|
6 comments
34.
▲
by
joemag
8y ago
The thing that bothers me about subscription based models is that there is a non linear relationship between value to the customer and revenue. At some inflection point, increasing the value customer gets out of the product does not result
35.
▲
by
joemag
8y ago
EC2 engineer here. You can connect transit gateway to VPCs owned by different accounts.
36.
▲
by
joemag
8y ago
Space Time is fantastic. Upon watching the entire series (catching up to present time, to be more exact, since new episodes are still being released weekly), I went on a graph traversal spree of related YouTube videos. Another great series
37.
▲
by
joemag
8y ago
Apparently, this is one of the dodgier parts of owning a condo. The condo we used to own had a structural problem - nothing major, but one that needed to be repaired in the next several years, or else it would get worse. Getting the builde
38.
▲
by
joemag
8y ago
This happened in Seattle [1], where 25 floor luxury apartment building had to be demolished. [1] https://en.m.wikipedia.org/wiki/McGuire_Apartments
39.
▲
by
joemag
8y ago
Most cities, you have to get a building permit to do any major renovation. Those permits tend to be pretty specific on what you can or cannot do - some even require the exact construction plans. The article makes it sound that the owner wen
40.
▲
by
joemag
8y ago
Performance hit is barely noticeable for most apps. The biggest surprise is usually heap usage increase. You can increase your heap from 28 to 34 GB, and end up with less free memory.
41.
▲
by
joemag
8y ago
This is from PBS Spacetime series [1]. A fantastic channel, and they did a number of videos on black holes. [1] https://youtu.be/KePNhUJ2reI
42.
▲
by
joemag
8y ago
It comes up as result #2 for me. Although all the “cards” push that towards the bottom of the page.
43.
▲
by
joemag
8y ago
Except propagating a configuration update to multiple clients, without a split brain is a hard distributed systems problem :). Especially in the presence of those pesky network partitions.
44.
▲
by
joemag
8y ago
Agreed that such setups can be simpler than multi-master replication. Yet, there's still a hard distributed systems challenge even in the active/passive setup like you describe. When a failure occurs, the clients need to agree whi
45.
▲
by
joemag
8y ago
You are totally correct that most applications do not require more than a single host to handle all the load, especially given the hosts that are easily available today. However, the far more common reason to go distributed is redundancy.
46.
▲
by
joemag
8y ago
Hi, you used DynamoDB as an example of a weakly consistent system in the opening paragraph, but it actually supports both modes [1]. The point of confusion might come from the fact that the service described in 2007 Dynamo paper was an insp
47.
▲
by
joemag
8y ago
Many Java applications will gladly take 10%, or even higher, performance hit if that means more predictable GC times. If you own a latency sensitive application, then it's common for tail latencies to be dominated by GC times. Bringin
48.
▲
by
joemag
8y ago
I was going to post that the impact of higher FPS is likely too low to justify needing 4x higher processing power. But then the difference in reaction time between 30fps and 120fps is O(30ms). At 60mph that translates to almost 1 meter in
49.
▲
by
joemag
8y ago
I’m perplexed by the use of “honest” and “greedy” as the opposites...
50.
▲
by
joemag
8y ago
Interesting idea, but the article greatly downplays the dos angle. Most applications consume fair amount of resources to start up - there are config files to parse, caches to warm, and so on. So there is an imbalance of cost between an atta
51.
▲
by
joemag
8y ago
Or it does, but the person is smart enough to diversify to avoid suspicion :). Potentially because they tried that several times, each time going back and adjusting the plan.
52.
▲
by
joemag
8y ago
It’s a great model for the users. The fact that it hasn’t seen significant adoption makes me think that it’s not viable one, though. If you think about it, advertisers would put much lesser value on the free tier users, so the free tier wil
53.
▲
by
joemag
8y ago
Law of averages... The distribution might be such that the kind of user willing and able to pay $100 a year for a social network is worth a lot more than $100 a year to advertisers. No matter how you slice it, the ad supported model might
54.
▲
by
joemag
8y ago
Yeah, you usually have one time use keys that can work in cases like that. Or you can get a new key from IT and go through a registration support.
55.
▲
by
joemag
8y ago
In interviewing, like in many other subjective measurements an important trade off is precision (percentage of hires that worked out) vs recall (good hires that got rejected). Experienced interviewers, or an innovative process can optimize
56.
▲
by
joemag
8y ago
That was sarcasm. Many cities have NFL and MLS teams share stadium. The two fields are almost identical dimensions.
57.
▲
by
joemag
8y ago
I’m struggling to understand the complaint. The form clearly states that the app will be granted ability to read/edit or delete email. For as long as we consider third party add ons a viable feature of gmail, apps will require access,
58.
▲
by
joemag
8y ago
Do you even need to play with DNS caching? Aren’t subdomains excluded from cross-origin request blocking? Can badsite.com request a resource from subdomain.badsite.com, and the browser will allow it. The JavaScript on badsite.com can be cl
59.
▲
by
joemag
8y ago
There is upcoming AWS region in hk: https://aws.amazon.com/blogs/aws/in-the-works-aws-region-in-...
60.
▲
by
joemag
8y ago
There have been many well publicized events where a production database was lost, or in some way corrupted. I’ve lived through one such event myself. When that happens, you usually go to a backup, and typically run into two problems: 1) Poi
More ›