HNHacker News
TopNewBestAskShowJobs

runningdogx

735 karma · joined April 7, 2009

submissionscomments
runningdogx··on June 8th: the day your phone won’t stop ringing
"Expecting everyone to have IPv6 connectivity sorted out throughout the enterprise network before June 8th is ludicrous..."

It's 2011. What's ludicrous is enterprise desktops with broken dual-stack support.

If IPv6 day weren't happening, remaining problems would not get fixed. Then in a year or two, major companies would start dual-stacking their public webservers in an unsynchronized fashion, and end users would be complaining about miscellaneous brokenness off and on for months, possibly years.

runningdogx··on 56% of Americans have Internet data caps; FCC asked to investigate
I'll pick on at&t uverse since that's what I've got, and since they're the latest to join the transfer cap party.

Those grandmas are not paying $65/mo for the top service tier. They are paying $35/mo for 3Mbit/s lowest-tier uverse service, and they still get a transfer cap of 250GB.

Streaming netflix users, who are pretty much all going to be in the top tier or two of service (24Mbit/s or 18Mbit/s) get pinched by the same cap. 250GB is 3.2% utilization of an 24Mbit/s download link, over a month, btw.

Even in some alternate universe where caps make sense, the least AT&T could do is grade the caps by line speed. It seems to me like I'm subsidizing grannie's transfer allowance, even if she doesn't fully use it.

Since the primary issue is traffic congestion at peak usage times, the obvious solution is to rate limit heavy users, rather than punishing them with hard caps which indiscriminately targets users who transfer lots of data in off hours.

Given all that, why does AT&T cap all uverse tiers the same? The first explanation that comes to mind is that they know if they drop the caps too much for lower tiers, because despite what they claim 250GB really isn't that much, they'll start cutting into too many grannies' modest internet usage. They want to keep their low-end customers from being hurt by lower caps, but they want to gouge their higher-end customers for going over the same caps.

And why do they cap instead of rate limiting heavy users? Rate limiting doesn't make them money. Caps might (though they will lose some customers over this outrage, and others will cut back usage or service tiers, so AT&T might even lose money overall).

runningdogx··on 56% of Americans have Internet data caps; FCC asked to investigate
As do I. Throttling per user (as in per customer port) rather than per traffic stream (the latter can cause net neutrality issues) is obviously a superior technological solution. Traffic caps and overage charges do nothing to prevent people from saturating links at peak traffic times (evenings, for home ISPs).

How do you know Charter doesn't throttle bandwidth at peak times? Do you track packet loss, and would you notice if it went up by a percent or two?

Even if an ISP doesn't voluntarily throttle traffic, border routers automatically will with WRED or some other more modern queue management strategy if there's too much traffic buffered for the link it's trying to send on. It's more a question of whether packet loss is managed on a per-flow or per-ip basis, on egress, or whether it's managed on a per-port basis on ingress from the customers to their ISP.

runningdogx··on 56% of Americans have Internet data caps; FCC asked to investigate
Verizon did something interesting with their wireless network: they track heavy bandwidth users and throttle them during peak traffic periods. [1]

That seems like a more reasonable option than hard traffic caps. It's no surprise more wired ISPs aren't using that strategy, though... hard caps generate revenue, while rate limiting heavy users doesn't. That wouldn't be such a problem if there were healthy competition in most metro markets, but there isn't.

Verizon FIOS is one of the few good [fast] U.S. wireline ISP options left that doesn't cap traffic. It's too bad many markets are still de-facto duopolies (from the days when there was a meaningful distinction between "telephone" and "cable" companies aside from layers 1-2). FIOS isn't available here, and if it were I'd switch in a heartbeat because of AT&T's incompetence [2].

[1] http://www.tekgoblin.com/2011/02/03/verizon-to-throttle-high...

[2] http://leaf.dragonflybsd.org/mailarchive/users/2011-02/msg00...

runningdogx··on How fast is the Internet at Google? Mind blowing.
Are you in a major city or a more outlying area? Is that a wired, or wireless, connection? Even wired access could have wireless backhaul... assuming yours is wired, is that the reason for the slowness, or have they really been mad at work all over India laying conduit with copper instead of fiber?
runningdogx··on TrueCrypt 7.0a's Performance, Analyzed
No.

http://www.truecrypt.org/docs/random-number-generator

The RNG pool is used for creating: the master key (for encrypting the data in the volume), temporary secondary key (for generating a pseudorandom stream to initially overwrite a volume's empty space), salt (for passphrase hashing, using pbkdf2), and (optionally) a keyfile.

That is all limited to the process of setting up a volume, and so the RNG pool is not used for typical operations on a volume after that volume is created.

runningdogx··on Linode launches native IPv6 support
Just follow the HE tunnel setup instructions like you would for any non-linode host or border router.
runningdogx··on Heroku's AWS outage post-mortem
From 9:07 to 20:43, the status updates were generic and not very helpful in answering two questions customers want to know:

1. What exactly is going on? 2. When will it be fixed?

In the middle of a crisis, saying "we're aware of the problem, and we're working hard to fix it," for hours does not really count as communication. It increases customer aggravation rather than decreasing it. Customers want to know answers to the above two questions. They don't care that you know about the problem and that you're working on it, unless you're not doing those two things in which case they will be (and should be) furious; those two things are expected.

Barring the ability to tell your customers "we will be back up at X:00", I think the best approach is to share as much information as you can without getting into proprietary information. That's why I think GP considered their communication a failure. That's why I consider their communication a failure, although I've seen this pattern enough from different companies that I don't hold it against Heroku as long as they learn from it.

runningdogx··on The Linux Security Circus: On GUI isolation
...a major difference being there are not robots yet that can go around breaking into people's less-secure houses and torturing residents for information, all in an automated fashion.
runningdogx··on Court rules Linux kernel infringes on patents. Lawsuit avalanche to follow.
Scavenger hunts are set up as linked lists.
runningdogx··on LibreOffice And OpenOffice Will Not Merge
MySQL already has three maintained patchsets or complete forks, and there's no reason to believe any of them would die if Oracle stopped open sourcing MySQL tomorrow. Percona Server and MariaDB are drop-in compatible, while Drizzle went in a different direction. All three projects have some pre-Oracle MySQL developers working on them.

Far be it from me to discourage someone from adopting Postgresql instead of MySQL, but there are other options that retain compatibility with mysql-only apps.

runningdogx··on An Introduction to Compassionate Screen Scraping
Screen scraping is taking visual data and transforming it into structure data. A screen scraper would graphically capture a window and try to identify or pick out data. Bots for MMOs tend to do that, alnong with providing input to the MMO depending on what they "see".

Web or data scraping is what the article talks about. Still a hard problem, easily broken by minor changes to the scraped webpage, but not subject to the vagaries of OCR and computer vision or graphical interpretation problems, which is what I was expecting from the title.

runningdogx··on Git Cheatsheet (Visualization)
Nice, but can you normalize the height of the command-name strips? They can range from 17-19px (before +4 padding) in height, depending on the local font setup I suspect. Cumulatively that's enough of a difference so that in an unusual case (when they're all 19 tall, as they can be in a gentoo+chromium setup), the Local Repository column of strips extends down and obscures the description.
runningdogx··on Oracle gives up on OpenOffice after community forks the project
{MySQL: [:"Percona Server", :MariaDB, :Drizzle], Java: [:Clojure, :JRuby, :Scala]}
runningdogx··on Out of Fuel: Why Hasn't Innovation Provided a Reliable Alternative to Oil?
How are biofuels viable as a replacement for oil in the quantities our current lifestyle demands? They can't be produced in the same quantities that oil can be, and producing them requires farmland, so food prices go up.

stats and graphs: http://earlywarn.blogspot.com/2011/02/us-ethanol-production....

runningdogx··on Nginx 1.0.0 is out
Not to mention that Cherokee's configuration file format is very human-unfriendly, and its configuration utility is a GUI that requires a web browser (and port forwarding if you want to configure a server remotely, which would typically be the case outside of dev environments).
runningdogx··on Nginx 1.0.0 is out
It (nginx) is bsd licensed. If you don't like the development model, you can always fork it on github, or start a company that develops and supports a closed-source fork.
runningdogx··on Google Modifies HTTP, Makes Chrome 50% Faster
http://www.google.com/support/forum/p/Chrome/thread?tid=0d78...

I've seen two things implicated in several google support threads about the issue. First, having both the system and google's sandboxed flash versions enabled as plugins, and second, an older version of Trusteer Rapport (typically offered by banks) was implicated.

Both may be fixed by now, but if it's still crashing, try about:plugins, enable Details at top right, and disable the system version of Flash.

runningdogx··on Show HN: CSS3 Animated Glowing Forms
Chrome 10 (stable-latest): exhibits the problem

Chrome 11 (beta-latest): renders properly

runningdogx··on MIT's artificial leaf is ten times more efficient than the real thing (Wired UK)
Since most reactions do not end up with all the constituent atoms as free ions, the real question is how the sum of bond energies of the reaction product(s) compares to the sum of bond energies of the starting molecule(s).

In the case of water, 2 H20 -> 2 H2 + O2 is endothermic, but reactions do not have to be endothermic, or else chemistry would be very boring and cold.

runningdogx··on RSA hit by targeted attacks, SecurID 2-factor auth possibly compromised
It's my understanding that such login banners are used to provide support for prosecution efforts, should someone break in and claim they did not know that access by random people was prohibited.

I would hope nobody thinks such warning messages are an adequate substitute for human-based and technology-based security measures, any more than wearing garlic cloves or putting up "no guns allowed" signs are replacements for being wary of vampires and armed criminals.

runningdogx··on Japan suspends work at stricken nuclear plant
(According to the IAEA) There was one measurement of 400mSv between reactors 3 and 4 [1], I think which actually occurred sometime on monday GMT (possibly Tuesday morning in Japan), which prompted the evacuation. It fell again, which probably prompted the authorities to allow workers back in.

The 1000mSv figure is most likely a misreporting of the transient 1008 or 1080uSv (I recall it had an 8 in it) measurement reported over the weekend, or some subsequent measurement... there have been lots in and around the plant that have been on the order of 1000uSv.

I really wish everyone in the media speaking about the nuclear crisis would timestamp their comments accurately, ideally in GMT. People all around the world are trying to follow events, and it's nearly impossible to know when some reporter says something about an event whether they're using Japanese time, their local time, or GMT.

[1] http://www.facebook.com/notes/international-atomic-energy-ag...

runningdogx··on Video of reactor 1 at Fukushima plant
Some reports[1] indicate radioactive iodine and cesium are being detected at monitoring stations. How could that happen if confinement hadn't failed? (fuel rods melting, containment failure of the reactor vessel and/or primary coolant loop, and the obvious lack of a containment building at this point)

Based on the description of two-stage containment for BWRs at wikipedia[2], and the video, if that's the containment strategy at Fukushima, it seems the drywell containment structure blew up, since steam being vented into a pool of water shouldn't cause a visible breach.

[1] http://e.nikkei.com/e/fr/tnks/Nni20110312D12JF520.htm

[2] http://en.wikipedia.org/wiki/Containment_building#Boiling_wa...

runningdogx··on Wolfram Alpha doesn't "get it"
You read every ToS for every service/product you use?

Community and social standards often trump legalese. For instance, most absurd restrictions in apartment rental agreements are not enforceable.

In this case, linking to another site, no matter how "deep" the link is, is common practice.

If WA really doesn't like it, they can prevent it by checking the referrer, or going even further by adding a pseudorandom token to their query page, switched every 5 minutes and valid for an hour or a day, and redirecting deep links which don't have a valid token.

Of course, doing that would probably kill some traffic they like, so they have a silly, selectively-enforced and probably unenforceable ToS clause instead.

I don't see any reason why such provisions need to be enforced legally, since they can be enforced technologically in most cases. Expanding the law because of laziness is a horrible precedent.

runningdogx··on Steve Huffman on Lessons Learned at Reddit
I don't understand the fetish with caching every variation for every user.

Here's a thought: dynamic apps should render a page exactly one way in html, and rely on javascript and cookies to post-process the site so it APPEARS to be customized for that particular user. That includes admin widgets.

That particularly applies for displaying how many minutes ago something was generated. Serve it with a date-time format, use javascript to post-process into "x seconds ago" or "x minutes ago".

If someone isn't logged in, or isn't an admin, and they hack their javascript to display user or admin stuff, who cares? The user/admin requests sent to the webserver won't succeed anyway, because they rely on having an admin session cookie.

If there's meaningful rather than just UI stuff that users or admins get to see, then you have to cache that separately, but you can even do stuff like loading it dynamically with js so the publicly visible (cached) content can still be used, and you cut down the amount of stuff your server has to auto-generate. It can cache the separate pages (xml, json, whatever) that serve the logged-in user content, as well.

runningdogx··on Dispelling the New SSL Myth
Source-IP based persistence on the layer 3/4 load balancer solves that very easily. The src-ip cache merely needs the same timeout as the webservers' ssl cache.
runningdogx··on VideoLAN launches new website
They're finally displaying the version number on the download button! That was my main gripe with the old site.
runningdogx··on Tell HN: There is a Scammer Amongst Us
In that case, Mallory was a fool. Mallory should have installed the MITM cert in the browser's certificate store, to prevent warnings. How many people routinely audit their browser's SSL cert list?
runningdogx··on How to Disagree
Complementing that essay somewhat, a list of fallacies: https://secure.wikimedia.org/wikipedia/en/wiki/List_of_falla...

And cognitive biases: https://secure.wikimedia.org/wikipedia/en/wiki/List_of_cogni...

Although not all are equally bad.

runningdogx··on Google algorithm change launched
Would it be feasible to introduce a search customization setting: "lower the estimated probability that I mistyped"? I assume google uses something like two cut-offs: A < B, where if P(mistype)>A, it suggests but doesn't search for the correction, while if P(mistype)> B, it displays auto-corrected results. Simply an user option to nudge P(mistype) lower would probably suffice, or an option to increase B.

A major usability problem for me with auto-corrected results is the two lines with links: "Showing results for <link>" and "Search instead for <link>". Interpreting those two lines takes me out of the flow of analyzing search results. I haven't yet been conditioned to click on the second link. I never want to click on the first link, because google is already showing me those results.

I frequently search for unusual acronyms, terms, variables, made-up words, and gibberish that shows up in logs, among other things that google likes to "correct" for my benefit. I realize those use cases are not typical, but current auto-correction behavior can be extremely aggravating in those cases.

Auto-suggest a correction all you want, but auto-correcting and suggesting the original is going too far I think, unless there are zero hits, or some equivalently strong algorithmic determination is made that I couldn't possibly want the query the way I typed it.

← PreviousPage 3 of 5Next →