Profile

Cover photo
Kevin Reid
Works at Google
Attended Clarkson University
246 followers|71,593 views
AboutPostsPhotosYouTube

Stream

Kevin Reid

Shared publicly  - 
 
 
Holy crap, this musical marble machine is amazing!

Have been watching various videos of the guy building the machine over the past year.  First became aware of it when he emailed me asking questions about gearing a long time ago.
25 comments on original post
1
2
Add a comment...

Kevin Reid

Shared publicly  - 
 
 
Introduction [link]. The prepper culture begs to be taken with a grain of salt. In some ways, it has all the makings of a doomsday cult: a tribe of unkempt misfits who hoard gold bullion, study herbalism, and preach about the imminent collapse of our society. Today, we see such worries as absurd ...
1 comment on original post
1
1
Add a comment...

Kevin Reid

Shared publicly  - 
 
OK, there is no way this is not a gross abuse of the SMTP spec. But it leverages an awesome quirk of +MantisBT's email reporting plugin.

I craft a message with a particular MessageID, and in that message, I included a References header that references...that same message ID. So, the message can be said to reference itself, which, while not explicitly disallowed by the RFCs, is definitely not what you're supposed to do with it.

This message I crafted? Pass it into Mantis. Mantis will create a ticket. Pass it into Mantis again, mantis will take the second copy and add it to that existing ticket as a bugnote. Keep passing it into Mantis, and you'll keep getting notes added to that ticket.

Close that ticket, and the next time a message with that same messageID and References header shows up, and Mantis will create a new ticket, mark it as related to the old, and future messages with that same messageID and References header will get added as bugnotes to the new ticket.

So, what's the upshot? Let's say you have an NMS that emails your bugtracker whenever a particular trigger is hit. You configure things so that those messages have a messageID that's only as unique as the trigger is; future triggerings of the same condition would produce the same message ID. So, now, if you have a flapping trigger, or if your trigger goes off from time to time without a clear indication of root cause, any future messages sent by the same trigger will get filed under the same ticket, and you build a history of the events. If you think you've got it licked, you close the ticket. Next time the trigger fires, a new ticket gets opened, marked as related to the old ticket, and you can walk back the ticket history for that trigger for a historical perspective, if you wish.

And your ticket bookkeeping and cleanup gets grossly reduced, too.
5 comments on original post
1
Add a comment...

Kevin Reid

Shared publicly  - 
 
I was reviewing my internet router/modem configuration, and found this menu setting in my WiFi access point:

     IPv6: Disabled

That's silly. Fixed. Now my entire home network has IPv6. Does yours?
1
Kevin Reid's profile photoRob Meijer's profile photo
3 comments
 
There could be a different reason behind it related to an annoyance I have with my current router. Given that I only have a single IPv4 adress at home, I tend to only run sshd on my VMs using IPv6. When I first got my current router a few years back I was puzzled by the fact that while I apparently still had IPv6 connectivity, I was unable to connect to any of my VMs. Turned out my router had some kind of TCP firewall that disallowed all incoming TCP connections on IPv6.  And worse, there was no way to turn it of globally. When I queried the manufacturer about this 'bug' they claimed it was a security feature as it allowed users the same level of security that IPv4 with NAT did.  

I could imagine a similar line of reasoning would make other manufacturers opt to just turn of IPv6 by default. The idea that home users who are used to being behind NAT will run services on their machines and that they don't expect these services to be made accessible through the internet when they upgrade to a dual stack router and/or ISP.
Add a comment...

Kevin Reid

Shared publicly  - 
 
(Not “FTL is probably impossible but necessary for stories”.)
Writing science fiction – especially hard science fiction, where you're expected to keep to what modern science believes is possible, rather than inventing force fields, lightswords, faster-than-light-drives, or other accoutrements of space opera – is a demanding task.
3
Rob Meijer's profile photoKevin Reid's profile photo
2 comments
 
+Rob Meijer​ I'm tempted to disagree with your analysis, but, like I said, the topic isn't FTL.
Add a comment...

Kevin Reid

Shared publicly  - 
 
 
Just like buffer overflow bugs are impossible can be eliminated by using a memory safe language, bugs like this recent OS X privilege escalation can be eliminated by using object capabilities.  What makes this particular OS X bug interesting is that it is the very same bug from a 1977 Unix-like system described in the first paper on the confused deputy problem.

http://cap-lore.com/CapTheory/ConfusedDeputy.html
http://waterken.sourceforge.net/aclsdont/current.pdf
DYLD_PRINT_TO_FILE: This is a path to a (writable) file. Normally, the dynamic linker writes all logging output (triggered by DYLD_PRINT_* settings) to file descriptor 2 (which is usually stderr). But this setting causes the dynamic linker to write logging output to the specified file.
2 comments on original post
3
1
Add a comment...
Have him in circles
246 people
Walter Connor's profile photo
Sean Anderson's profile photo
Craig Overend's profile photo
Dan Cocke's profile photo
Fake Unicode Consortium's profile photo
Georgi Himchev's profile photo
Yo'el Erez's profile photo
Max Shawabkeh's profile photo
Vipin Pavithran's profile photo

Kevin Reid

Shared publicly  - 
 
In software-defined radio, there are well-established ways of visually representing the entire received bandwidth, to see all signals within it. Here is a way to hear them as well, with caveats.
3
Add a comment...

Kevin Reid

Shared publicly  - 
 
The Museum of English Rural Life got a surprise on Wednesday — a 155-year-old mousetrap there managed to catch a mouse: So, this retired rodent had managed to sneak past University of Reading security, exterior doors and Museum staff, and clambered its way up into our Store. Upon finding itself there it would have found …
1
Add a comment...

Kevin Reid

Shared publicly  - 
 
He said thinking about infinities gave him headaches, so I suggested he see a doctor of philosophy.
1
Add a comment...

Kevin Reid

Shared publicly  - 
 
 
Full Moon, Full Earth
Image Credit: +NASA, NOAA/DSCOVR
http://apod.nasa.gov/apod/ap150807.html

The Moon was new on July 16. Its familiar nearside facing the surface of planet Earth was in shadow. But on that date a million miles away, the Deep Space Climate Observatory (DSCOVR) spacecraft's Earth Polychromatic Imaging Camera (EPIC) captured this view of an apparently Full Moon crossing in front of a Full Earth. In fact, seen from the spacecraft's position beyond the Moon's orbit and between Earth and Sun, the fully illuminated lunar hemisphere is the less familiar farside. Only known since the dawn of the space age, the farside is mostly devoid of dark lunar maria that sprawl across the Moon's perpetual Earth-facing hemisphere. Only the small dark spot of the farside's Mare Moscoviense (Sea of Moscow) is clear, at the upper left. Planet Earth's north pole is near 11 o'clock, with the North America visited by Hurricane Dolores near center. Slight color shifts are visible around the lunar edge, an artifact of the Moon's motion through the field caused by combining the camera's separate exposures taken in quick succession through different color filters. While monitoring the Earth and solar wind for space weather forcasts, about twice a year DSCOVR can capture similar images of Moon and Earth together as its crosses the orbital plane of the Moon.
74 comments on original post
1
Add a comment...
People
Have him in circles
246 people
Walter Connor's profile photo
Sean Anderson's profile photo
Craig Overend's profile photo
Dan Cocke's profile photo
Fake Unicode Consortium's profile photo
Georgi Himchev's profile photo
Yo'el Erez's profile photo
Max Shawabkeh's profile photo
Vipin Pavithran's profile photo
Education
  • Clarkson University
    Computer Science, 2010 - 2012
Contact Information
Home
Email
Work
Employment
  • Google
    Software Engineer, 2012 - present
  • Google
    Software Engineering Intern, 2011 - 2011
Basic Information
Gender
Male
Looking for
Friends, Dating, Networking
Relationship
Single
Other names
kpreid
Links
YouTube
Contributor to