Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

You've heard of “yes men”. Now learn about “no men” (2011)

rachelbythebay.com

Post reply on HN
Print view View original

111–120 of 151 posts

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#111

Post by JamesBarney »

hrktb wrote: post #19

I don’t get it. 1 - there is a Mr.No guy but, the author only knows him by reputation 2 - there is a technical problem that seems to revolve around memory 3 - the author has an idea of a solution, she actually tries as people don’t believe her 4 - her solution works, but she is told the “No” guy said “no swap” once a long time ago Her reaction: “Did I care what he supposedly said? No.” Why doesn’t she go talk to him…

This is my experience too. Dealing with "no men" is usually far easier in person.

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#112

Post by lr4444lr »

rbavocadotree wrote: post #88
shados wrote: post #79

Earlier quoted context omitted.

JavaScript is indeed a very good example of this. Not just in the ecosystem, but even just the standard itself. People are pushing for more and more stuff. People on the committee can say "no", but there's an infinite amount of bullshit to push, and only a limited amount of time to push back. So sooner or later, bullshit makes it in.

What are examples of bs in the javascript standard?

String (capital "S") class.

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#114

Post by scarejunba »

egdod wrote: post #103
scarejunba wrote: post #40

Earlier quoted context omitted.

It makes it sort of like the sci-fi book Ancillary Justice by Ann Leckie. You have no idea what the gender of the protagonist is. Perhaps future gender-neutral speech will sound like this instead of choosing a third pronoun.

Yes! I listened to a good bit of that audiobook, but it was such a pain having to puzzle out who was doing/saying things. I eventually gave up. I mean... it was an interesting idea, writing a book that way because the characters sexes shouldn’t really matter. But in practice, it was a mess.

It was easier to follow in the written form.

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#115

Post by icxa »

shados wrote: post #80
novok wrote: post #50

Earlier quoted context omitted.

That is why you write that in a standards document, along with the reasons why and have a consensus about it. That way it becomes impersonal. Same thing with auto formatters and linters sidestepping a lot of pointless bickering and bike shedding.

That works for things you want to DO, but is hard for things you DON'T want to do. As I mentioned in another post, as your org grows, there's going to be an infinite amount of things people push, and pushing back is hard. If you have to stop and write a consensus document for every single one, you'll be doing nothing else.

How hard is it to say "these are the types of things we want to do. anything must look like these types of things in terms of (insert x y and z criteria: size, scale, opportunity, profitability, growth, etc)

spoiler alert: it isn't hard and this is what we do at my startup that is currently experiencing hyper growth and is on multiple "unicorn" lists

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#116

Post by nwallin »

jcrites wrote: post #91

I agree with the author's overall premise, which is that other people who don't own and who are not responsible for your system should not have authority over how you build and operate it. One of the things that I like about the current company where I work is that owners always have the final say about their systems, not any external technical authority. There is no such authority who could say no where I work (with…

> generally no swap in production.

This is a bad policy. The problem is that it breaks Linux's overcommit logic. Many applications allocate huge chunks of virtual memory far in excess of what they actually use. Due to the magic of virtual memory, this doesn't get mapped to physical memory until it's actually used.

The trouble is that there's no mechanism in any programming language I'm aware of that allows the OS to retcon an allocation. The only tool the kernel has at its disposal is the oom killer. So if you allocate too much, and you have swap disabled, the oom killer has to step in and kill something even if you're only using half your ram.

This is bad enough, but in addition to this, you have vfs cache to worry about. If you have 64GB ram, your applications are using 40GB, and you have 22GB cache, you and I might assume you have 24GB free, but the OS only sees 2GB free, and a few allocations might trigger the oom killer if they happen faster than the vfs can dump cache. (which is surprisingly slow) With swap, this isn't a big deal, because the pages don't actually get assigned until they are used, which happens later, so the vfs has plenty of time to evict.

The point of swap isn't to give you more memory. It's so the OS can keep its promises about memory allocation and overcommitting.

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#117

Post by nailer »

lr4444lr wrote: post #112
rbavocadotree wrote: post #88

Earlier quoted context omitted.

What are examples of bs in the javascript standard?

String (capital "S") class.

Why don't you like Strings? All prototypes (Number, Array etc) have a capital letter.

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#118

Post by nailer »

slewis wrote: post #109

I know exactly who she's talking about! When I was at the unnamed company I worked on a project that he retained control over and said no to a lot. We found that if we sent him a very well reasoned, documented and measured change he'd say yes. I was there when the no swap policy was under lots of active discussion. It was there to ensure that application authors could have strong expectations on how long a memory acc…

What the app something that required deterministic latency? There are situations where your rather hit OOM and kill your app and go buy some RAM rather than have it operate slowly - executing trades too late, or pointing at targets that have moved, etc.

Re: You've heard of “yes men”. Now learn about “no men” (2011)

#119

Post by nailer »

8ytecoder wrote: post #31
pjc50 wrote: post #16

Earlier quoted context omitted.

Swap causes not disk decay but performance decay; the system gets unamanageably slow and would have been better off with a clean crash and failover.

Only if there's thrashing. It’s prohibitively more expensive to build a machine that has enough memory for way more than peak usage than to have some swap that can handle some level of overflow pages. Like everything swap has to be monitored and there’s no need To disable it entirely

It depends. Sudden peaks of latency can get comparatively more expensive on a trading floor than buying an absurd amount of RAM.
Post reply on HN