Chris X Edwards

--------------------------

I Shall Speak Of It

2015-04-24 11:48

Wer redet heute noch von der Vernichtung der Armenier?

{Who still speaks today of the extermination of the Armenians?}

— Attributed to Adolf Hitler**

** For the complicated details of this quote’s provenance, read all about the Obersalzberg Speech.

Exactly 100 years ago on the night of 23–24 April 1915 the government of the Ottoman Empire rounded up and imprisoned an estimated 250 Armenian intellectuals and community leaders. This set in motion events that eventually led to the deaths of around 1.8 million Armenians, predominantly civilians, who were systematically exterminated. I do not think I risk violating the spirit of Godwin’s Law by suggesting that these events were eerily analogous to Kristallnacht and the Holocaust of World War 2.

Being a part of a culture that perpetrates barbaric mass violence on innocent people is never going to end especially well. Even when you think you’ve gotten away with it your unlucky descendants will have to live with the legacy of your execrable deeds. Perhaps the only thing that more flagrantly demonstrates intellectual laziness than such craven violence is later pretending it did not happen or was justified, especially when confronted with substantial evidence to the contrary.

For almost a century there has been some controversy about the Armenian Genocide. Despite being the event that caused the the word "genocide" and the concept of crimes against humanity to be invented, Turkey emphatically denies that a "genocide" occurred. So much so that it is a crime there to assert that a genocide did in fact occur. In other countries (Switzerland for example) it is a crime to deny that such a genocide occurred.

What’s crazy is that few argue whether or not countless Armenian men, women, and children were systematically and horrifically murdered. They clearly were. Turkey seems mostly intent on denying that the word "genocide" should be applied to the case. To me that is essentially like saying, "Sure I murdered them but I didn’t have a very good plan." Basically second degree murder instead of first degree.

My perspective on murder differs from the law in that I’m rather unsympathetic to the defense of incompetence. I feel much safer around people who plan to kill a specific person for a specific reason and then do so than the people who at any time could just kill any one for no particular reason.

But does anyone really need to quibble about terminology in this case? Would the Turks prefer we call it the "Extermination of the Armenian People by Turkish Incompetence"? That would still be misleadingly generous. Imagine your most extenuated intentional murder scenario. Maybe a barroom brawl, road rage, cheating spouse, etc. How many instances of such a crime committed by the same person do you think the police would be willing to believe were spontaneous? Two, maybe three at the most. After "accidentally" running over the fourth pedestrian, the police will surely be reopening the files on one through three.

We’re not (just) talking about about the political assassination of a few hundred people. We’re talking about roughly half of Armenians. This doesn’t "just happen". The claim that the lack of good planning and high-quality German engineering somehow makes the crime less serious seems delusional to me. Furthermore, this isn’t an isolated case. The Turks claim that the chaos and general turbulence of World War 1 caused the depopulation of Armenians which would almost be plausible if it were not for the fact that non Armenians were hardly thus affected. Oh, and the other genocides. Yes, it seems the Ottoman Turks had been practicing for 1915 with previous massacres of Armenians.

The point of this is not to humiliate anyone (still alive anyway). On the contrary, the essential objective is to eliminate future opprobrium. The idea is to not let the important historic implications of a people’s actions become unaccounted externalities of their present policy. It is essential to remember the darkest parts of human history as accurately as possible if we are to most effectively prevent future occurrences.

I will not forget the Armenians.

Not Quite Linked In

2015-04-23 14:25

Since I’m on the topic of LinkedIn I’ll mention this interesting experience.

Someone I was supposed to talk to was introduced to me with an accompanying LinkedIn URL.

https://www.linkedin.com/pub/theguysname/5/428/205

Since I didn’t understand what the trailing numbers were I assumed the worst, that they were to track me. Instead of using that supplied URL, I searched for the guy’s name in the LinkedIn search box. I did this while not logged into my account. Strangely LinkedIn showed my picture, and assumed I was the user (probably based on IP number and/or cookies). The problem is that if I was not the user, say I had borrowed someone’s computer, it would be highly misleading to see my photo with the caption "Chris X Edwards - Research at Sandia National Laboratories" which LinkedIn showed. I presume it was a job ad, but to someone who is not me, it looks like I not only would condone working for a source of nuclear weapons but that I currently actually do!

I found the guy easily enough and was able to see his full employment history. LinkedIn then told me that to see his full profile I need to log in. Fine, be that way. After I logged in, however, I could no longer see his employment history. Or anything! They even suppressed his last name which I originally searched for!

With LinkedIn, it’s definitely a fine line between being a user and, as RMS calls them, being a "used".

Bespoke PC

2015-04-23 11:39

I tend to build work stations for people to actually do specific productive things on. A major reason to build these yourself is that you get exactly what you want. If you just want to browse the web or watch lolcats, anything will do, but if you have a specific and unusual agenda (like doing actual work with a computer), then the flexibility and customizability of building it yourself can not be matched by complete solution vendors.

This article clearly points out the second critical reason to build your own work stations.

Workstations are among the best builds you should put together yourself. Big box companies often markup their machines by thousands of dollars and get away with it because businesses and professionals are willing to pay for the extra cost. However, if you’re willing to spend a couple hours on a weekend building your own brand new workstation, you can easily save thousands of dollars! No longer will you need to pay Dell, HP, or Apple $5,000 for what you can build yourself for $2,000!

Here are my favorite useful resources for configuring new PC builds.

They Call Themselves Programmers

2015-04-21 07:12

Certainly there exist truly incredible programmers. However, I think the modern internet tends to be a distorting lens highlighting them and their accomplishments. In real life programmers and other computer professionals seem… well, let’s just say that the words "gamma distribution" come to mind.

Saw this on a Linux irc channel this morning.

06:55 <A> i work with people that think if you change .xls to .txt then
 you just changed the file
 06:56 <B> i'm dealing with a supposed IT department now that thinks about
 the same thing
 06:56 <A> they also think that giving me a PDF instead of the raw data is
 in fact helping me out
 06:57 <B> they are generating a "word XML document" which should have a
 .xml extension, but they call it .doc and pretend it's a binary
 doc format. so my spam filter swats it down.
 06:57 <C> where the heck do you work ?!
 06:57 <B> why they have chosen the ridiculous word xml format, i do not
 know
 06:58 <B> it's some reports generated by a 3rd party marketing company
 06:59 <A> "can you guys handle our new fixed width format" "this isnt
 fixed width. it's comma delimited" "YES WE KNOW THAT! but can
 you work with it?"
 06:59 <A> i should have just said yes no matter what since i dont even
 care what format it is
 07:00 <D> i wonder how they survived for so long
 07:00 <A> drugs
 07:01 <A> we have a whole branch in another location that doesnt know how
 to handle tab delimited files.
 07:01 <A> they call themselves "programmers" too

Not Quite Linked

2015-04-16 12:36

I can’t quite decide if LinkedIn is kind of creepy or unacceptably creepy. Perhaps the best argument-silencing reason that I host my own email is that I am certain that I have never sold you out to a social network through my email contacts. You’re welcome. But I am only half of the conversation. Person A and person B may have both sent me emails at some point. Or person A may have sent an email to me and also to person B. Now LinkedIn knows A and B might know each other and/or me.

While I do give companies I deal with unique email addresses so that I can track their activities, I do not do this for friends and other humans. To most people, what LinkedIn knows about you must just seem like the result of modern life. Having been scrupulously careful to disclose as little as possible through non-explicit mechanisms, the stuff LinkedIn comes up with for me can be pretty disturbing.

The hard push of this universal replicator to turn raw material into more opportunities to recursively do the same lead to some interesting data. My "People You May Know" suggestions seemed mildly interesting, but as I scrolled down I was impressed at how many people they could stitch together with a plausible connection to me from other people’s email contacts. Eventually, exactly 900 potential contacts were recommended.

Remember, these are not my LinkedIn contacts, but people near me in the network graph of email correspondents and LinkedIn connections. I thought this data set was large enough to be significant and insulated from me enough to not contain the biases of my actual contacts (though other biases clearly exist).

Lightly scraping these contacts I was able to extract some interesting observations.

  • 832 (92.4%) shared a contact with me

  • 415 (46.1%) shared 3 or more contacts with me

  • 185 (20.6%) work at some kind of university or institute

  • 160 (17.8%) are engineers of some kind (including network, software, etc.)

  • 112 (12.4%) are professors

  • 110 (12.2%) are software engineers or programmers

  • 82 (9.1%) are scientists

  • 66 (7.3%) work at UCSD (my employer and largest in the region)

  • 45 (5.0%) work at Google

  • 30 (3.3%) work at Microsoft

  • 13 (1.4%) work at Microsoft Research

The numbers are not perfect since I don’t really know exactly what these people are doing, but I would say the magnitudes are correct based on my analysis and quick correction of some messy data points. I think it’s interesting that LinkedIn has more people for me to consider from Google and (plus) Microsoft than from my own employer which is the largest (26,000) in San Diego. It’s also interesting that you’re more likely to hit a professor than a software engineer in my vicinity of the network graph, though many are no doubt both. Professors also probably have a high vertex degree.

If this were something serious, it would seem that I’m pretty well connected to the high stratosphere of STEM. My actual contacts and especially my real friends corroborate this. In real life I don’t know what this means or how useful it is. Although this may well imply that I have "knowledge and skills for jobs of the future" at this moment in time I think it mostly just means that I’m a nerd.

--------------------------

For older posts and RSS feed see the blog archives.
Chris X Edwards © 1999-2015