<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: MichaelGG</title><link>https://news.ycombinator.com/user?id=MichaelGG</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 09 Apr 2026 08:33:28 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=MichaelGG" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by MichaelGG in "The Engineer vs. the Border Patrol"]]></title><description><![CDATA[
<p>>Well history suggests some answers to that question<p>Such as? American Natives arguably fared rather poorly with immigration. What other examples did you have in mind? Other countries don't seem keen on immigration. Is South Africa encouraging Europeans to come over? Mexico and a lot of Latam have difficult immigration laws. Hell I ended having to pay a few thousand dollars to have my Canadian daughter leave Guatemala because she overstayed her visa despite her mother being a citizen. Imagine the outrage if the US started applying fines to be able to leave!<p>>but they don’t have a solution, period.<p>Isn't it automation plus very very limited purpose-specific visas?</p>
]]></description><pubDate>Mon, 28 May 2018 22:56:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=17175321</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17175321</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17175321</guid></item><item><title><![CDATA[New comment by MichaelGG in "The Engineer vs. the Border Patrol"]]></title><description><![CDATA[
<p>What's wrong with them having preference for people of a similar race? What's wrong for looking at the source countries of many of these immigrants and saying no thanks? No one's in a hurry to make sure lots of whites have positions in Mexico or China.<p>Japan has a low birthrate but doesn't accept losing their ethnic makeup is a good solution.</p>
]]></description><pubDate>Mon, 28 May 2018 22:39:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=17175271</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17175271</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17175271</guid></item><item><title><![CDATA[New comment by MichaelGG in "Kitty Marion: an actress who became a suffragette who planted bombs"]]></title><description><![CDATA[
<p>I agree with your first 2 paragraphs. The problem is that people can be convinced that there is an issue where there is none, then resort to violence on weak grounds. It is easy to get people riled up on spurious reasons. I was about to bring up BLM but see you did so:<p>>considering what they’re protesting, I think BLM has been remarkably NON-violent<p>Last year only 20 unarmed black people were shot and killed by police vs 30 whites. So far this year the numbers are 8 and 11. Counting all police shootings last year, whites are 457 of them, blacks 223. About twice the number of whites have been shot by police. <a href="https://www.washingtonpost.com/graphics/national/police-shootings-2017/?noredirect=on" rel="nofollow">https://www.washingtonpost.com/graphics/national/police-shoo...</a><p>So yes blacks are killed at a higher rate per population. But what about by actual violence? When looking at violence, blacks and whites commit about the same number of homicides despite population differences. Homicide is the most fair stat we have. If we use lesser crimes [like drug possession or other violence], we will see blacks overrepresented due to higher policing and a biased justice system. In fact, as minority areas of cities have a low solve rate, going by homicides may understate the racial difference in crime in favour of blacks.<p>It sort of looks like, if you want to get racial about it, <i>whites</i> are the ones killed too much by police. And if you look at the numbers for Hispanics, it is even worse. (But that might be an artifact of how race/ethnicity/Hispanic is reported.[1])<p>Yet that does not stop perception. There was an ad on TV a while back with a black mother having "The Talk"[2] with her teenage daughter about getting stopped by cops and not wanting her killed. Well, let's see: in 2017 <i>one</i> unarmed female black person was shot by police, and that was when a SWAT team raided her and her bf's place. Such ads that portray this as an issue are simple fear mongering. 2018, so far that number is 0. An ad about how to hide in case of lightning storms would be more realistic. Or simply more training about driving. Or avoiding pools. Or practically anything else in the world.<p>Try it out: Go ask around and see what your friends and others think the true rate is of unarmed people being killed. Ask them what difference in numbers they think exist for black vs white, and ask them about the flip side of civilian violence. My guess is you'll first get some incredibly high statement about how this is just happening non-stop, then when you reveal the numbers you'll get some other excuse about how it is not about the numbers anyways but some other general racial problem.<p>1: I'm using <a href="https://ucr.fbi.gov/crime-in-the-u.s/2014/crime-in-the-u.s.-2014/tables/expanded-homicide-data/expanded_homicide_data_table_6_murder_race_and_sex_of_vicitm_by_race_and_sex_of_offender_2014.xls" rel="nofollow">https://ucr.fbi.gov/crime-in-the-u.s/2014/crime-in-the-u.s.-...</a> for a quick view of homicide stats. The numbers could be off by 2x and it would not change the point much.<p>2: Found it: <a href="https://youtu.be/3s20ePvTaME?t=21" rel="nofollow">https://youtu.be/3s20ePvTaME?t=21</a> "This is not you about getting a ticket, this is about you coming home" to which the girl says "I'm going to be OK....right?" Obviously a dramatization but if you're somehow implying to your teenage girl that cops are going to pull you over and you'll not "come home", you're the problem.</p>
]]></description><pubDate>Sun, 27 May 2018 21:27:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=17169192</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17169192</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17169192</guid></item><item><title><![CDATA[New comment by MichaelGG in "The DisplayPort Pin 20 Problem"]]></title><description><![CDATA[
<p>Applies to software too. Standards need to be strongly defined with no leeway. Parsing should be tight and leave no room for creativity. (Text protocols like HTTP I'm looking at you!) Anything that deviates should be rejected by reference implementations instead of trying to be "robust" by accepting junk.</p>
]]></description><pubDate>Sat, 26 May 2018 19:11:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=17163754</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17163754</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17163754</guid></item><item><title><![CDATA[New comment by MichaelGG in "Coinbase acquires decentralized cryptocurrency trading platform Paradex"]]></title><description><![CDATA[
<p>I'm unsure how it doesn't fix the speed advantage. Anyone can get in front by bidding a millionth cent more. You can keep doing that until the price starts mattering, right?<p>I'm not against HFT at all. But having such granular pricing doesn't do anyone favours. More decimals would reduce spreads as well as silencing HFT critics and maybe make trading a bit more accessible without as much high end systems. But the spread reduction is valuable alone.</p>
]]></description><pubDate>Thu, 24 May 2018 02:02:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=17140416</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17140416</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17140416</guid></item><item><title><![CDATA[New comment by MichaelGG in "Coinbase acquires decentralized cryptocurrency trading platform Paradex"]]></title><description><![CDATA[
<p>SEC can "fix" HFT instantly by allowing more decimals. There's no good reason to only have 0.01/0.001 increments. Making it so granular forces speed over accurate pricing. Make it 8 decimals and the speed advantage goes away.</p>
]]></description><pubDate>Thu, 24 May 2018 00:27:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=17139805</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17139805</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17139805</guid></item><item><title><![CDATA[New comment by MichaelGG in "Coinbase acquires decentralized cryptocurrency trading platform Paradex"]]></title><description><![CDATA[
<p>Who has Tether defrauded? I agree Tether may have just issued without backing, but they don't promise being able to withdraw. So who is actually being harmed? And why is it unfixable? What if Coinbase promised to completely cover any usdt shortfall?</p>
]]></description><pubDate>Thu, 24 May 2018 00:21:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=17139761</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17139761</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17139761</guid></item><item><title><![CDATA[New comment by MichaelGG in "Stopping the RSI pain that almost destroyed my programming career (2016)"]]></title><description><![CDATA[
<p>It's bizarre. I tossed the book away in disgust because it's crazy pseudoscience....and it still worked.</p>
]]></description><pubDate>Wed, 23 May 2018 15:52:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=17135123</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17135123</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17135123</guid></item><item><title><![CDATA[New comment by MichaelGG in "Congress wants to extend the copyright on some sound recordings to 144 years"]]></title><description><![CDATA[
<p>At this point, it's unlikely he or any other artist will release more music in the 1900s or even the 2000-2017 period.</p>
]]></description><pubDate>Fri, 18 May 2018 21:18:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=17105028</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17105028</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17105028</guid></item><item><title><![CDATA[New comment by MichaelGG in "YouTube Music, a new music streaming service, is coming soon"]]></title><description><![CDATA[
<p>If it'll let me play YouTube audio on my Sonos it'd be a great feature. Otherwise I'm not quite getting the value.<p>Workaround for this is to buy an HDMI splitter and send the audio to a sound system then use Chromecast, but Sonos is just so convenient.</p>
]]></description><pubDate>Thu, 17 May 2018 06:10:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=17089136</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17089136</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17089136</guid></item><item><title><![CDATA[New comment by MichaelGG in "Validating UTF-8 strings using as little as 0.7 cycles per byte"]]></title><description><![CDATA[
<p>There is probably a trade-off depending on the content of the string, right? So the API probably needs a general-purpose and a "this should be all ASCII" version?</p>
]]></description><pubDate>Wed, 16 May 2018 15:43:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=17083292</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17083292</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17083292</guid></item><item><title><![CDATA[New comment by MichaelGG in "Validating UTF-8 strings using as little as 0.7 cycles per byte"]]></title><description><![CDATA[
<p>I meant Rust should have a SIMD optimised version that assumes mostly ASCII. I'm guessing there is a trade-off involved depending on the content of the string.</p>
]]></description><pubDate>Wed, 16 May 2018 15:41:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=17083275</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17083275</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17083275</guid></item><item><title><![CDATA[New comment by MichaelGG in "HTTP headers we don't want"]]></title><description><![CDATA[
<p>Some things are just wrong. I've implemented SIP, a horrible standard. Lots of compatibility issues just from their insistence on a "human friendly" text format alone.<p>At any rate there's lots of things you just have to ignore, drop, reject, and otherwise muck about with in order to run a sane network. These standards are not written with software experience. They're written much in a vacuum and out of touch. This varies widely across RFCs so it might not apply to RFCs you like.<p>Example of a MUST for SIP and HTTP: line folding and comments in headers.  Apart from being crap for performance (so much for being able to zero-copy a header value as just a pointer+len) there's zero legitimate use for these "features" of the syntax. Simply rejecting such messages is in your best interest as a network operator.</p>
]]></description><pubDate>Wed, 16 May 2018 15:07:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=17082996</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17082996</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17082996</guid></item><item><title><![CDATA[New comment by MichaelGG in "Validating UTF-8 strings using as little as 0.7 cycles per byte"]]></title><description><![CDATA[
<p>Fast checking is really useful in things like HTTP/SIP parsing. Rust should expose such a function as well seeing as their strings must be UTF-8 validated. Though it's even faster if you can just avoid utf8 strings and work only on a few known ASCII bytes, it means you might push garbage further down the line.</p>
]]></description><pubDate>Wed, 16 May 2018 14:49:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=17082829</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17082829</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17082829</guid></item><item><title><![CDATA[New comment by MichaelGG in "iOS 11.4 to Disable USB Port After 7 Days: What It Means for Mobile Forensics"]]></title><description><![CDATA[
<p>>I could see Cook & co. make the case that these devices are so dangerous in that regard that they shouldn't be able to be used by anyone.<p>OK but that is not a legal strategy. You're not providing any basis other than that Apple should have some magical power to prevent people from touching devices they legally have access to.<p>OK and if you got Planned Parenthood or the Humane Society listed as a domestic terror organization it would hurt their funding, too. What's your point?</p>
]]></description><pubDate>Wed, 09 May 2018 03:11:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=17027181</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17027181</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17027181</guid></item><item><title><![CDATA[New comment by MichaelGG in "iOS 11.4 to Disable USB Port After 7 Days: What It Means for Mobile Forensics"]]></title><description><![CDATA[
<p>What are you talking about? You want to make it so that a company like Apple can just draw arbitrary bounds and say "no messing around beyond this point" and have that be internationally, legally, enforced?<p>We got that with the DMCA and DRM modules, phone unlocking, and console rooting.</p>
]]></description><pubDate>Wed, 09 May 2018 03:09:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=17027176</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17027176</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17027176</guid></item><item><title><![CDATA[New comment by MichaelGG in "Google Duplex: An AI System for Accomplishing Real World Tasks Over the Phone"]]></title><description><![CDATA[
<p>But I am saying that it is unlikely Google will let you spam reservations. This idea that some kid will now be able to make 1000 reservations via Google is not probable.</p>
]]></description><pubDate>Wed, 09 May 2018 03:07:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=17027170</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17027170</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17027170</guid></item><item><title><![CDATA[New comment by MichaelGG in "David Marcus Leaves Messenger to Focus on Blockchain Within Facebook"]]></title><description><![CDATA[
<p>That's just misusing the word blockchain. Like saying "cloud server" for "hosting" when there is nothing cloud about it.<p>Using digital signatures in a chain is not a novel idea. Without PoW you do not need the block part, just records.</p>
]]></description><pubDate>Wed, 09 May 2018 02:11:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=17026970</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17026970</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17026970</guid></item><item><title><![CDATA[New comment by MichaelGG in "Android P"]]></title><description><![CDATA[
<p>I get "Connection problem or invalid MMI code" on an AndroidOne device when I dial that. Edit oh hidden asterisks.</p>
]]></description><pubDate>Tue, 08 May 2018 22:32:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=17025909</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17025909</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17025909</guid></item><item><title><![CDATA[New comment by MichaelGG in "Never Write Your Own Database (2017)"]]></title><description><![CDATA[
<p>Here's a version <a href="https://github.com/michaelgg/cidb" rel="nofollow">https://github.com/michaelgg/cidb</a> -- Just some of the raw integer k-v storage part. It assumes you already have the hashed entries (you truncate them and the compression takes it from there). It is really what you should expect more from a college course IR project but since I never went to school... oh well.<p>I used this same library to encode telephone porting (LNP) instructions. That is a database of about 600M entries, mapping one phone number to another. With a bit of manipulation when creating the file, you go from 12GB+ naive encoding as strings (one client was using nearly 50GB after expanding it to a hashtable) to under a GB. Still better than any RMDBS can do and small enough to easily toss this in-RAM on every routing box.<p>Some day I'd like to write it in Rust and implement vectorized encoding and more compression schemes. Like an optimized SSTable just for integers.</p>
]]></description><pubDate>Tue, 08 May 2018 22:24:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=17025843</link><dc:creator>MichaelGG</dc:creator><comments>https://news.ycombinator.com/item?id=17025843</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17025843</guid></item></channel></rss>