latch

27 ms·

211.
by latch 5y ago
As it's described in the article, neither side is presenting a compelling case. The government isn't saying what problem it's trying to solve, the teacher's union isn't saying why it's bad (1). I went through t
212.
by latch 5y ago
All of these vulnerabilities impact desktop and servers. I guess I wasn't clear, but when talking about "multi-tenant hosts", I meant servers.
213.
by latch 5y ago
The best mitigation is to not use multi-tenant hosts.
214.
by latch 5y ago
Having been on the receiving end of a "How cute", it's extremely condescending and obviously violates HN guidelines. (the entire comment is just a "shallow dismissal")
215.
by latch 5y ago
Plotting is both CPU and IO bound. Plotting can't fully take advantage of multiple cores. The workaround is to start N plots, each taking ~400GB of space and ~2 cores. On an 8 core server with 2x800 GB SSD, you can do 4 plots in parall
216.
by latch 5y ago
nginx has proxy_cache_purge (1), but I agree that nginx doesn't expose a particularly good API around its cache. Plus, the awkwardness of doing disk IO from OpenResty (unless this has been improved), is a drag. Maybe this is something
217.
by latch 5y ago
Why not an http proxy with disk caching, e.g. location / { proxy_http_version 1.1; proxy_set_header Connection ''; proxy_set_header ..... proxy_cache s3_zone; proxy_pass https://
218.
by latch 5y ago
slower?
219.
by latch 5y ago
This (1) thread from François Chollet describes 'Artificial Intelligence' very well in my opinion. From this point of view, it's obvious why you need to fallback to other data/metadata. Further, to the linked tweets and
220.
by latch 5y ago
Their margin on bandwidth is literally over 1000%. Quick google says that S3 costs 320% more than Backblaze (which, presumably, isn't running at a loss). > At a guess The comments in this discussion that try to provide actual number
221.
by latch 5y ago
Sure, but you actually have to go through the steps. The very first step, back of the napkin, indicates savings. More than enough to warrant a more detailed evaluation. Then you can start considering the more complicated factors (redundancy
222.
by latch 5y ago
> Let's do the math. Offers no math. At retail, 625 16TB drives is $400000. This is about 2x the MONTHLY retail s3 pricing. Further, as we all know, AWS bandwidth pricing is absolutely bonkers (1). I think your conclusion that S3 is
223.
by latch 5y ago
It's 100% our mistake(s). It's only AWS fault indirectly, in that AWS is complicated and requires a lot of non transferable knowledge. As a small example, we currently pay $750 for Route53. We don't know why (it isn't tr
224.
by latch 5y ago
I'm one of those "you'll get my baremetal and systemd out of my dead cold hands", kind of guy. But I have reasonable exposure to both AWS and GC and I can say that, by far, Google Cloud is easier to reason about. As a co
225.
by latch 5y ago
It's democracy when it works in our favor, and war when it works against us. People in Bolivia and Peru struggle to afford local food (e.g., Quinoa) because they're outbid (by money, but certainly not necessity).
226.
by latch 6y ago
Did you look at whether Go was part of the problem or were you using something else and you found that Go was part of the solution?
227.
by latch 6y ago
So if I'm compiling PostgreSQL from source, should I be doing: export CFLAGS='-O3 -march=native' Before ./configure? Because if I don't, it's using -O2 without specifying an architecture.
228.
by latch 6y ago
We're a small transport company, trying to get employees to and from work. No matter the time of day, there are always people commuting (or planning to commute). When's a good time for outage? Imagine it's 3:30am, you just go
229.
by latch 6y ago
Both SaaS and multi-tenant hardware have massive surface area. For multi-tenancy, it isn't about trusting AWS/Azure/GCP, it's about trusting everyone you're sharing hardware with. Cloud products are difficult to set
230.
by latch 6y ago
Yes, but no. I'm a staunch believer that multi-tenant hardware and managed services are _obvious_ no-gos for privacy reasons. But, having done B2B where i had to deal with security procedures/questionnaires/documentation&#x2
231.
by latch 6y ago
PostgreSQL is great, but I don't think your statement is particularly true. Process per connection is pretty easy to accidentally run into, even at small scale. So now you need to manage another piece of infrastructure to deal with it.
232.
Things I hate about PostgreSQL (2020)(rbranson.medium.com)
439 points by latch 6y ago|272 comments
233.
by latch 6y ago
I think these two rightfully get highlighted, but I've found it difficult to organize complex Go code (though clearly some people manage), which I think is now my biggest issue with the language. It's inability to deal with cyclic
234.
by latch 6y ago
Wait, am I misunderstanding, or are you saying the code would usually be written as: func (c *GZipComprosser) compress() []byte { if (c == nil) { return []byte{} } return []byte{c.internal} } ? I don&#
235.
by latch 6y ago
If you go to the article and read the quote under the title, you'll see where the 50x comes from.
236.
by latch 6y ago
Did you read the article? There are 2 cleanrooms. The 2nd story one is undamaged but needs the 1st story one to be repaired (I assume it's a structural thing?). So I'm curious how, from 1 picture, your comment can be informed? Als
237.
by latch 6y ago
I just spent this past weekend playing with FreeBSD. Pretty sure the answer comes down to whether your hardware is supported. I gave up trying to install it on my XPS due to the lack of support for 802.11ac and issues with video. But it ins
238.
by latch 6y ago
Yes, but I prefer Ubuntu and I'm under the impression that running Linux on Macbook's isn't great ??
239.
by latch 6y ago
I'm on my 2nd XPS 13. I switched to it and linux after being severely impacted by the butterfly keyboard (i.e. having to fly to a different country to get it repaired, only to have it break again a few months later). I don't _hate
240.
by latch 6y ago
> You could do it from scratch, but it would take so long that you would likely never be able to manage the actual complexity you care about Except plenty of people have and continue to manage? Do we think that _most_ people are using Do

More ›