HNHacker News
TopNewBestAskShowJobs

nieve

1,134 karma · joined August 22, 2011

submissionscomments
nieve··on Ask HN: What's the largest website hosted on Digital Ocean?
Re SSD disks, as someone who's using both Digital Ocean and Ramnode: For PostgreSQL on smallish data sets (5GB) doing Sphinx reindexes every few minutes Digital Ocean's I/O performance is much lower than Ramnode's KVM SSD instances. Disk benchmarks on the two gave me 5-10x in Ramnode's favor on small instances. Digital Ocean is still pretty decent for a budget provider and Ramnode has all the standard issues of a smaller hoster (and all the advantages), but if I/O performance is critical to you it might be worth taking into account. DO is much slicker, though.
nieve··on Stalked by an online supermarket that wants a 'relationship'
Several of the author's primary examples were actually companies she's continued to do business with, so they're clearly not applying this just to customers who weren't going to come back. I've certainly decided not to deal with a company again because right after the initial transaction I've gotten hit with a flood of this spam. I suspect there's one of the standard measurement issues where companies find it easier to quantify customers who've apparently come back due to this kind of marketing than to measure customers who've gotten sick of it and reduced or ceased their business with the company. It's possible to do a better job, but these companies never seem to and it might be that there's a perverse incentive for the marketers to mismeasure.

Having worked directly under the SVP of Business Intelligence & Analytics at one of the larger US-based social networks I can say from experience that it's really, really hard to get people to consider the things that are hard to measure or hard to interpret. Even in a company that's heavily analytics-driven there's a lot of magical thinking, a lot of anecdata, and a strong bias towards short term effects. Think of it as the marketing equivalent of laying off 20% of your workforce, telling everyone else to work 25% harder, and convincing Wall Street this means profits will go up and stay up.

It's easy to spot the bump in your balance sheet from lower costs and it's hard to quantify the long term costs of decreased morale, lower total productivity, most likely lower individual productivity, attrition, and inability to take advantage of opportunities.

nieve··on Append "_nomap" to your AP's SSID to opt out of Google Location Services (2011)
You may have missed the bits that came out about the capture going on with the knowledge of some Google employees. Th e original developer has plenty of experience with wifi data capture and knew exactly what you could do with the data as well as that you didn't need full packet dumps to snag SSIDs.

"Google long maintained that the engineer was solely responsible for this aspect of the project, which resulted in official investigations, some still unresolved, in more than a dozen countries. But a complete version of the F.C.C.’s report, released by Google on Saturday, has cast doubt on that explanation, saying that the engineer informed at least one superior and that seven engineers who worked on the code were all in a position to know what was going on." - http://www.nytimes.com/2012/05/01/technology/engineer-in-goo...

nieve··on Facebook offers to pay Menlo Park $600k to hire a cop
I should have phrased that more clearly. The cyberpunk thing is that we've got more aggressively open carrying and more dramaticly stupid shooting incidents like the string in Florida, very much the way a lot of cyberpunk has the trope of heavily armed idiots opening fire at the slightest excuse. We have nothing like cyberpunk levels of violence and as you point out it's mostly getting better, but we do have more incidents that'd fit nicely in a book than we used to and we have more people displaying weapons. I'm thinking about the little color bits thrown into cyberpunk to clue the audience in and things like the Zimmerman, Dunn, and Reeves cases fit in perfectly with that (modulo actual human tragedy).
nieve··on Facebook offers to pay Menlo Park $600k to hire a cop
It's been obvious for a while now that we're living in a dystopian cyberpunk future, it just snuck up on us:

  * Killer drones. Ubiquitous surveillance by both government & business.  
  * Bad data that destroys one's ability to travel, get credit, find work, or avoid being thrown into jail out of nowhere. 
  * More and more armed citizens shooting each other over minor things. 
  * Flash mobs for every purpose under the sun.  
  * Samizdata.  
  * Manipulating the cops into raiding innocent targets while using software to conceal your identity. Absolute and complete lack of airships. 
  * Corporate domination of the political process with unlimited donations.  
  * Massive and growing wealth inequality.  
  * Political campaigns severally wounded by software failures. 
  * "You're not the customer, you're the product." 
  * DDOS for commercial advantage.
  * Arresting people as "terrorists" to seize embarrassing stolen information.
  * International treaties negotiated at the behest of corporations as backdoors to subverting the legislative process.  
  * No-recruit agreements for the rank & file combined with vicious competition for the top talent.  People carrying half their life in portable computer devices.  
  * Ransomware in a myriad of forms.
  * Vast government computer intrusion programs.
  * Criminalization of access and dissemination of public information.  Criminalization of downloading too much from something you've got access rights to.  Criminalization of downloading from open directories with no warnings on them.
  * New diseases, rising seas, freakish weather.
  * A resurgent, militaristic Russia invading various countries.
  * A resurgent, militaristic USA invading various countries.
Admittedly the cyberware is a bit slow in coming, AI is still the same 10 years off as always, self-driving cars are barely at the testing stage, sub-orbital planes are a pipe dream, and nobod thinks the Net looks like brightly colored blocks, but it's still unsettlingly close.
nieve··on Introducing the dotDNS metaTLD
Does anyone who's got some background with DNSSEC (https://en.wikipedia.org/wiki/Dnssec) feel like comparing DNSChain to the basic idea of embedding certificates or signatures in authenticated DNS records? We've had a lot of different proposals for the latter over the years and my impression is their lack of adoption hasn't been because DNSSEC is fundamentally broken (though it is flawed in several ways). AFAICT it breaks down like this:

Against DNSSEC + certificate records:

1) There's a centralized PKI administration issue, but less so than the current CA morass 2) Administrative/rubber hose/legal/monetary compromise of the DNS root administrators (currently US NTIA) would allow substitution of the TLD keys, but this would be blatantly obvious to every organization or individual paying attention 3) DNSSEC has some annoying properties with regard to name enumeration and such 4) Need to either use a primary resolver that supports DNSSEC (Google Public DNS and Comcast, among others) or do a separate query to one who does

For:

5) Minimal extra infrastructure, already supported in some libraries 6) Extremely cacheable with low resource requirements (root & TLD keys, MRU/MFU certs) 7) No need to add a new dependency on a previously untrusted source of code and crypto data 8) As long as the parent TLD has deployed DNSSEC the organization only needs to get their key set up, add a theoretical cert record, and sign records as needed.

I think issues 1 & 2 are at least as big a problem with DNSChain because in practice the hand-waving of how you pick a trusted DNSChain chain is ignoring a huge class of vulnerabilities. Issue 4 is still a win over implementing a large chunk of infrastructure or trusting okTurtles'. The "just use Javascript" bit seems to apply equally to either, though I guess the _query_ portion of DNSChain is relatively simple.

nieve··on Reporting From the Web’s Underbelly
Thanks for the map! I should clarify that I was unaware of deaths specifically attributable to a deliberate swatting incident, the death rate in SWAT raids & no-knock raids is ridiculous. It may be much lower than the death rate from "accidental" shootings of even as specific a group as say young black men who turn out to have no criminal record, are unarmed, and aren't engaged in any criminal activity at the time of the incident, but swatting is more outrageous-seeming to most of the majority who have less worry about being shot by the police. I suspect that the fact that it's a deliberate act of endangerment also increases the perceived significance of the risk - malice is more threatening than incompetence even when it's rare and even if it's less dangerous.
nieve··on Reporting From the Web’s Underbelly
Given the current state of policing in the US swatting someone has a non-zero chance of getting someone severally beaten, tazered, tear gassed, or just shot. I don't know of any deaths, but there have been some really close calls reported in the press and and injury plus property damage. I don't think that really qualifies as a good prank, more of a harassment/endangerment tactic.
nieve··on Opera is really nice.
I just discovered that due to lack of time (if Google Translate is correct) Pano has been removed from the Firefox addons collection by the author and the github repository has been taken down. Since it's MPL 1.1/GPL 2.0/LGPL 2.1 according to the source I've posted a copy of what I believe is the last official release at https://www.dropbox.com/s/3yscqilam8gi8ut/pano%40teramako.gi... (please skim the code rather than trusting me).

There's a maintained and actively developed fork at https://github.com/salamiwst/Panopticon-Forked

nieve··on Opera is really nice.
You haven't needed Bartab for on-demand tab loading for a while now, it's the "Don't load tabs until selected" setting on the Tabs page of preferences. Automatically unloading still requires an addon, though. I've got a serious tab habit as well (459 at the moment) and I'd recommend the Pano addon. It has tab/page title/url search that narrows the tab view to matches, drag & drop re-ordering (en masse), closing tabs directly from the list, tab group management, and visual indicators for whether a tab is loaded (nice when disconnected). It's nice when I'm cleaning out my old tabs to be able to put in ycombi, select all, and close them in one fell swoop. Heavy tab users probably want to turn on some of the optional features.

[Edit] Pano can also hibernate tabs or groups, so if that's the last thing you're using Bartab for it can replace that as well.

http://teramako.github.io/Pano/index.html

nieve··on Linux Distributions that deserve more Love
Do you know of any installer-time options for linux other than btrfs? I don't know of any fs choices, but openSUSE supports btrfs out of the box and even integrates it with the package manager to allow rollbacks. It's pretty convenient and as far as I know you can simply mount an old snapshot (writeable!) and use standard Unix tools to compare them. I'm not sure if that's everything you want, though - a tool that could actually tell you what's changed using the internal snapshot information would probably be much faster and superior to a diff or rsync dry run listing. It's still closer than anything I've ever seen and the ability to simply switch which subvolume (snapshots are just subvolumes) is the default or mount it within an extant fs tree is a win for me.
nieve··on Secrets of the Caltech Cannon Heist Revealed
When I was there in the early 90s the house system had serious flaws, but in part that was because I was in Blacker - then one of the most inclusive, friendly, gender-balanced of the houses (though not without issues). Page was notorious for (self-)destructive behavior and though that was probably unfair it'd become enough of a self-selecting culture that they did live up to it more than most houses. I think most of the serious problems were more products of a small undergrad body (800 total) largely isolated from the surrounding community, lots of pressure plus a remarkable level of sleep deprivation, very bright people with a lot of personal issues, easy access to cheap drugs & alcohol in bulk, and a huge gender imbalance. Stress-induced breakdowns/flaming were common, group stalking & sexual harassment were common enough to have their own term (glomming) and sexual assault may have been common as on most other campuses despite geek culture's self image, harassment & assault by faculty was swept under the rug, blatant homophobia may not have been common but LGB students tended to be quiet about their orientation (trans students were dead silent), and the administration was basically uninvolved, untrusted, and authoritarian (to most students' minds).

I think the house system was a mixed bag there, it kept some people closer to each other than was healthy but it also provided some level of support for a lot of us. I suspect things have gotten much worse since I was there, though. The only honor code I've ever seen work was the largely self-policed "No member of the Caltech community shall take unfair advantage of any other member of the Caltech community" and the administration was clearly communicating that they no longer expected students to be capable of following it. My impression is that the morale effects of that really cut into the effectiveness of a system that had helped keep pranks from escalating to the really dangerous or outright damaging, cheating low, and aggression of any form very rare. It was mostly worked because it was too simple to have real loopholes, the community fetishized it, we got a lot of benefits from keeping it working (take home finals, lab keys, etc.), and fundamentally it kept the peace. The administration still has a page describing it, but my impression is that as a social construct it's a hollow shell of what it was.

TL;DR? The house system was less damaging than most frats due to self-selection for house membership, but that also concentrated some really unhealthy behavior and the rest of us were already pretty broken without the houses helping.

nieve··on Secrets of the Caltech Cannon Heist Revealed
When I was at Caltech in the early 90s the general belief[1] seemed to be that we'd stolen the cannon from Harvey Mudd originally and the concept of MIT stealing it from the Flems[2] was hilarious. The Flems would be flustered, the Mudders would have to get it back from Massachusetts, and the rest of us would applaud and go back to setting things on fire, painting on the walls, and lock-picking. Inter-house rivalries were more of a running gag than an actual conflict - especially since there was heavy overlap between house memberships (you could and often did join more than one) - but it played into the continual pranking/stacking.

[1] Short institutional memories confusing the issue further. [2] Fleming house members, basically stereotyped as Caltech's jocks at the time. I can't throw stones, I was a Mole and there were plenty of jokes about us, plus I almost dated a Flem.

nieve··on When valuable employees quit or resign
Oddly enough the original question is what a CEO can do, but after discussion how hard it is for the CEO emotionally the all of the specific advice is aimed to the valuable employee who's leaving. I'm not clear on how exactly the post author missed the point of the question so impressively and then submitted it here - perhaps he couldn't think of anything else to write in response? The tips for the employee are good, though a bit patronizing and basic, but it's not exactly very useful to a CEO. I've been on both sides of that table (though as manager for the employee who was leaving, not CEO) and post that actually answered the question would have been more than welcome. It's usually too late to keep the employee when they're resigning, but there's often some room for a positive relationship when they leave. In my experience keeping up morale, keeping a potentially valuable contact (you wanted to them for some reason, right?) available, avoiding a reputation hit with other potential hires & partners, and maybe being able to tap them for tribal knowledge or the occasional bit of consulting is well worth a lot of effort for the CEO. How you do that tends to be a bit less clear...
nieve··on A list of free programming books
Higher Order Perl is the book that actually got me excited about Perl again after using it for ten years and then not needing to touch it for five. As usual MJD is clear, absorbing, and has great taste in what to write about. So many techniques are a great fit for Perl/Ruby/Python and yet are completely absent from books on those languages.
nieve··on Angry entrepreneur replies to patent troll with racketeering lawsuit
If we're going for diagnosis-via-internet I think that it sounds a lot more like ILD (Intermittent Lawsuit Disorder). It's a pretty classic case - sudden displays of aggressive legal threats, weirdly self-righteous justifications, conflating criminal & civil law, completely disproportionate responses, and assertion of bizarre offenses. I think it meets most of the criteria in the C&DSM IV.

Less flippantly I think that there may be a pretty strong correlation between these behaviors and personality disorders, but bullying and animal threat display behaviors are still relevant models. It's potentially a big mistake to confuse one of these people who is panicking and trying to use intimidation as a response with someone who has slipped into a state of irrational anger & obsession. The former can be dissuaded or possibly negotiated with, the latter is more likely to just keep escalating. In particular I'd presume that the latter are more likely to actually file the frivolous lawsuits and it's a higher priority to get everything locked down on the legal front. Generally a good idea anyway.

nieve··on Epic, The Privacy Browser
I just checked myself and the identifying info for me was by a huge margin fonts & plugins. User Agent will become far less relevant if Epic gets more adoption and keeps a lock-step updating pattern. Perhaps as use of web fonts becomes more common perhaps it will become more feasible to send just a static list that's the same for all Epic users[1]. I suspect there's a potential cache-enumeration issue that would help tag browsers there, but I'm not sure.

As for plugins, that's a bit more difficult - I wouldn't mind an option to ask if I wanted to let a site enumerate plugins, but you start getting into header games & pointless reloads there if you're not careful.

[1] I force my own fonts for everything, so no UX issues for me. No idea how much of an issue it is for other folks.

nieve··on Message threading (1997-2002)
As far as console clients go Mew does very well with imap, in fact more reliably than most of the gui clients I've tried. It happily pulled down multi-GB Gmail mailboxes and kept them in sync without any issues, has sane disconnected-mode behavior, handled network failures well, etc. It's implemented inside emacs, but you don't have to be a heavy emacs user to use it and one of the vi-emulation modes might help if not. Built-in ssh tunneling support; good support for mime, pgp, s/mime; and pluggable search options were what convinced me. It depends on stunnel for ssl/tls, but I'm happy they're not reimplementing support. The manual could use some work, but it's relatively extensive.

https://github.com/kazu-yamamoto/Mew http://www.mew.org/en/info/release/

nieve··on New Features of Firefox Developer Tools: Episode 25
The combination of being able to freely modify network requests before resending them and being able to execute javascript exactly where you want it seems perfect for debugging other people's websites with less friction. Or figuring out exploits, depending on your point of view. The former is going to save me a lot of pain - I've been running Aurora for a year or two now and it's been good to me, but I've slacked on following new features.
nieve··on How does Trello access the user's clipboard?
I believe the security risk is primarily sniffing the clipboard continuously when you expose a tab (or in horrible implementations, have a tab open). Use a password manager that puts the password into the clipboard, flip past the hostile tab, the site now has one of your passwords. Copy a big chunk of text from a sensitive email or document, now they've got that as well. Secret urls, auth tokens in urls, private email addresses? Same.

Putting something into your clipboard is kind of annoying and could be a small security risk, but pulling the contents out is much worse. If it's a site you don't visit often it would be limited exposure, but if someone manages to hack Soundcloud/Pandora or inject JS into them you're talking about a lot of illicitly acquired information. Most of it may be useless, but it's a sucker-bet to assume no-one will get hurt by it.

nieve··on RSS mistakes: let's not make them again
Yep. I offered the info in case it was useful, but I don't think it lets Mozilla off the hook at all. It's slightly better than completely getting rid of the feed detection code, I guess. That's probably the next step...
nieve··on RSS mistakes: let's not make them again
You can put the RSS icon back in the toolbar (but not the url bar) by going into View->Toolbars->Customize and dragging it wherever you want it. It's a bit less obvious than the original, but it's still useful.
nieve··on Systemd is not portable and what this means for our ports
Thanks, that's a definite win for pulseaudio and I hadn't really considered the developer side of it at all (I'm primarily backend/network/db). I certainly don't think pulseaudio is an inherently bad architecture, but I do feel it's matured too slowly in terms of stability. On the flip side as you it's more user-friendly than Network Audio Server (which has fewer features) and and as you point out supports more of the "I've got an hdmi output playing music through my tv, but a bluetooth headset for voip" use cases that a decent-sized class of users run into.

It also occurs to me for the first time that the rise of pulseaudio was not that many years after laptops largely switched to integrated codec chips instead of discrete sound and I had all sorts of problems with those above and beyond the "Damn it, give me hardware volume controls" frustrations. It could have inherited some of the problems around quirky/flaky hardware that I remember ALSA facing in that period. At this point I just want something that can reliably play music or take audio input and for whatever reason pulseaudio isn't it for me. Hopefully that'll change before most distros switch to whatever the new hotness in sound setups is...

nieve··on Systemd is not portable and what this means for our ports
Oh, I definitely wasn't singing their praises even if ALSA was a step up from OSS, PulseAudio was just a new low for me in sound systems. My biggest frustration with it was the tendency it's always had on my systems to simply stop producing sound with no obvious cause and when it was working block anything that didn't use it from producing sound. Sometimes a random process had either accidentally muted everything or triggered a bug that did so, other times audio would get sent off to some completely useless sink, occasionally there were no noticeable causes and you just had to kill and restart the daemon (if you were lucky). On top of that a lot of choices were made that tended to bind it to one console, so switching to another would cut out all sound.

I guess my core complaint is that ALSA could be a pain to get working properly on a new system, but once it was working it didn't usually spontaneously break on me. PulseAudio was only slightly more likely to work out of the box, but was much more likely to break without my making any changes. I guess it's partially a sysadmin mindset, but I usually come down on the side of preferring up-front pain to long term unreliability. Perhaps I didn't use as wide a set of audio programs as you did, I didn't tend to run as many of them at once, or my hardware just had better drivers. For streaming sound (I didn't care about video) I usually stuck to network servers that could shove audio down a Shoutcast or http stream and listening with something like Amarok or MPD, I'm sure PA was nicer if you were trying to relay what your HTPC was playing back.

[ETA: I use linux consoles very heavily when I'm coding on a linux machine, I know this isn't exactly the most common behavior among more recent unix users.]

nieve··on Systemd is not portable and what this means for our ports
Poettering oversold its state of stability and then seems to have largely lost interest during the most critical period of adoption. The distributions are responsible for believing the hype and shipping too early, but that doesn't change the fact that the developers were hyping it well beyond what was justified. PulseAudio has a lot of nice features, but it's still less stable and has more bizarre failure modes for me _9 years_ after its initial release than plain old ALSA. I literally know no other linux users who have ever had anything good to say about PulseAudio. Entirely anecdotal, but when it comes to trusting developers to do a good job in the first place, much less help with a major transition I'm not inclined to discount that.
nieve··on VP of Engineering - "best of the best" = no more than 12 years experience?
The minimum isn't the important part - they're putting a cap on experience at 12 years which will exclude the vast majority of protected workers, basically anyone who didn't switch careers very late.
nieve··on Saltstack: Changing RSA public exponent from 1 to 65537
Doing your own crypto is generally hard and risky, but with a little googling there are often reasonable defaults and suggestions available from trustworthy sources. The biggest danger seems to be that people pick some random set of values & constructions and assume that's good enough without further checking.

On the flip side this should have set off warning bells for them - it can be a little bit of a code smell to pick 1 for a configurable value in your encryption. Not always a bad choice, but 1 has a lot of properties that can simplify various mathematical operations and Basically setting a crypto value to the smallest positive integer (and a non-prime at that!) is like deciding that a set of numbers in your code should always be represented by two-byte signed ints. Sometimes that's _exactly_ what you want (say, various fast math libraries), but if you're doing general purpose code you should really stop and think about whether you might end up with a value outside that.

nieve··on Yelp Hipster Finder
And don't forget the red blotch of shame where Ballard has been getting gentrified into the ground. Hipsters here have more overlap with yuppies than most places I've lived.
nieve··on Der Spiegel says US bugged EU offices in Washington
Snowden has been releasing documents steadily & feeding stories to other outlets(1) and has stated he has quite a few more stashed with people who will decrypt and distribute the full set immediately if anything happens to him. Given that what we've seen has been cherry-picked from what appears to be a much larger collection and he's stated he has more, the odds are pretty high. He's probably passing some to Wikileaks, but I'd be surprised if they have the full dump.

(1) The South China Morning Post out of Hong Kong broke the PacNet story: http://www.scmp.com/news/hong-kong/article/1266875/exclusive...

nieve··on Evi Nemeth, retired Boulder computer professor, missing at sea near New Zealand
Weirdly enough that article mentions Dyche as the skipper and she's only listed as "a woman named Evi". AFAIK she's the shipowner and has plenty of experience as captain as well as being the most well known person on board(1), but I guess either the reporter didn't do their research or the authorities only mentioned Dyche.

(1) Beyond her general profile in CS and administration a quick google check shows 136,000 hits for her name and 13,000 for "David Dyche", almost all of which appear to be someone else.

← PreviousPage 14 of 16Next →